Telling Wonders: Ethnographic and Political Discourse in the Work of Herodotus 0472112031, 9780472112036

Citation preview

Telling Wonders

Telling Wonders Ethnographic and Political Discourse in the Work of Herodotus

Rosaria Vignolo Munson

Ann Arbor

Copyright 䉷 by the University of Michigan 2001 All rights reserved Published in the United States of America by The University of Michigan Press Manufactured in the United States of America 嘷 ⬁ Printed on acid-free paper 2004 2003 2002 2001

4 3 2 1

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, or otherwise, without the written permission of the publisher. A CIP catalog record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data applied for ISBN 0-472-11203-1 (cloth : alk. paper)

To Paul and Louise  µοι   Θωµατα µεγιστα

Preface

My first thanks go to Donald Lateiner, once my teacher and now my friend, most generous through the years with advice, support, and information of all kinds. He has just recently read this book in its final draft, but he has certainly had to hear a great deal about it for a long, long time: I hope he has not been disappointed. Last year William Turpin, in the busiest of all seasons for a department chairman, took the time to go through an earlier, bulkier version of my manuscript to help me decide how I could cut it down to about half its size. I am immensely grateful to him for his heroic help, as I am to Ellen Bauerle, then of the University of Michigan Press, for her wisdom in prescribing the shrinkage. Carolyn Dewald, my much admired fellow Herodotean and always a source of inspiration, has read all existing versions of this book. She made herself available for countless written and oral critiques of my work and for lengthy telephone discussions; she has been most generous in sharing her work in progress with me. I also thank Deborah Boedeker, Gregory Nagy, Helen North, Martin Ostwald, David Potter, Gil Rose, Grace Ledbetter, David Konstan, Elaine Fantham, Eric Munson, and John Carleo for their support, suggestions, or criticism at various stages of this project. David Chamberlain, John Graham, Bruce Grant, Andr´e Lardinois, Anthony Podlecki, and Kurt Raaflaub have kindly made available to me various sources and resources or their unpublished and forthcoming work. Colin Day, Collin Ganio, and Jennifer Wisinski of the University of Michigan Press have been extremely helpful, prompt, and encouraging. I am especially grateful to my copyeditor, Jill Butler Wilson, with whom I have carried out a tacit dialogue through colored pencils: she has done everything she could possibly do to save me from my own imprecision. This work could not have been completed without the generous leave policy and a Becker Fellowship from Swarthmore College. I owe a great deal to Stephe Bensch for proofreading the manuscript and suggesting

viii

Preface

numerous corrections. His sensitivity to language has made my prose somewhat less contorted and easier to read. My son, Paul, has also provided help in matters of style and a great deal of encouragement. My daughter, Louise, has been my closest companion while I was working on this project: she has tolerated my frequent absentmindedness with the gentle irony that is typical of her nature. I am also grateful to Luigi and Mariella Vignolo for following my progress and cheering me on from afar. Finally, I thank my beloved parents, Paolo and Mariarosa Vignolo: this book is a partial testimonial of how much they have done for me for so many years and a tribute to their memory.

Contents

Introduction

1

Chapter 1 Narrative and Metanarrative

20

Chapter 2 Comparison 45 Chapter 3 Interpretation and Evaluation Chapter 4 Thoma 232 Conclusion

266

Bibliography

275

General Index

293

Index of Passages

309

134

Preface

My first thanks go to Donald Lateiner, once my teacher and now my friend, most generous through the years with advice, support, and information of all kinds. He has just recently read this book in its final draft, but he has certainly had to hear a great deal about it for a long, long time: I hope he has not been disappointed. Last year William Turpin, in the busiest of all seasons for a department chairman, took the time to go through an earlier, bulkier version of my manuscript to help me decide how I could cut it down to about half its size. I am immensely grateful to him for his heroic help, as I am to Ellen Bauerle, then of the University of Michigan Press, for her wisdom in prescribing the shrinkage. Carolyn Dewald, my much admired fellow Herodotean and always a source of inspiration, has read all existing versions of this book. She made herself available for countless written and oral critiques of my work and for lengthy telephone discussions; she has been most generous in sharing her work in progress with me. I also thank Deborah Boedeker, Gregory Nagy, Helen North, Martin Ostwald, David Potter, Gil Rose, Grace Ledbetter, David Konstan, Elaine Fantham, Eric Munson, and John Carleo for their support, suggestions, or criticism at various stages of this project. David Chamberlain, John Graham, Bruce Grant, Andr´e Lardinois, Anthony Podlecki, and Kurt Raaflaub have kindly made available to me various sources and resources or their unpublished and forthcoming work. Colin Day, Collin Ganio, and Jennifer Wisinski of the University of Michigan Press have been extremely helpful, prompt, and encouraging. I am especially grateful to my copyeditor, Jill Butler Wilson, with whom I have carried out a tacit dialogue through colored pencils: she has done everything she could possibly do to save me from my own imprecision. This work could not have been completed without the generous leave policy and a Becker Fellowship from Swarthmore College. I owe a great deal to Stephe Bensch for proofreading the manuscript and suggesting

viii

Preface

numerous corrections. His sensitivity to language has made my prose somewhat less contorted and easier to read. My son, Paul, has also provided help in matters of style and a great deal of encouragement. My daughter, Louise, has been my closest companion while I was working on this project: she has tolerated my frequent absentmindedness with the gentle irony that is typical of her nature. I am also grateful to Luigi and Mariella Vignolo for following my progress and cheering me on from afar. Finally, I thank my beloved parents, Paolo and Mariarosa Vignolo: this book is a partial testimonial of how much they have done for me for so many years and a tribute to their memory.

Contents

Introduction

1

Chapter 1 Narrative and Metanarrative

20

Chapter 2 Comparison 45 Chapter 3 Interpretation and Evaluation Chapter 4 Thoma 232 Conclusion

266

Bibliography

275

General Index

293

Index of Passages

309

134

Introduction But the greatest wonder to me of all in that region, at least after the city itself, I am going to describe. It is their boats that go down the river to Babylon, all round and made of leather. For in the country of the Armenians, who live above the Assyrians, they cut ribs of willow and stretch around them watertight hides from the outside to serve as bottom, without broadening the stern or shaping the prow, but making it round like a shield. Then they fill this boat with straw, and after loading on the cargo, they launch it so that it is carried downstream. They mostly carry palmwood casks full of wine. The boat is steered by two paddles and two men standing up. One draws his paddle in and the other pushes his outward. These boats are made very large or of smaller size. The largest bear a cargo of up to five thousand talents. In each boat there is one live donkey, more if the boat is bigger. After they arrive by boat to Babylon and sell their cargo, they then auction off the frame and all the straw, but the hides they load on the donkeys and drive back to Armenia. For it is not in any way possible to sail upstream, because of the rapids of the river. For this reason, they make their boat not of wood but of leather. After they return to Armenia driving their donkeys, they make other boats in the same way. These are the boats they have. (Herodotus 1.194)

The Problem of the Ethnographies Reformulated In 490 and 480 b.c., the Greek city-states managed to defeat the aggression of the autocratically ruled Persians. Herodotus’ account of those events two generations later proceeds from the conflict’s remote antecedents. It describes the rise and expansion of the Persian Empire until the time when, in the plenitude of its resources, it came to threaten the Greeks. But Herodotus’ “story” also includes minute ethnographic descriptions, whose reason for being appears less plausible. For decades, we have read the Histories with the assumption, more or less resisted but

2

Telling Wonders

never entirely defeated, that the ethnographies belonged to another project, not only deriving from a different tradition of prose writing, but also conceived at a different stage of the author’s career and with different aims and concerns, a project whose disiecta membra were artfully but somewhat pointlessly inserted into the historical narrative.1 The lists of customs and geographical features in the ethnographies only occasionally provide information we need for our factual understanding of changes of dynasty and campaigns. We have come to agree that though they enrich the history in some general sense, they are not fully integrated with it. We have recourse to several strategies for dealing with the issue of “unity” in Herodotus. The stylistic and conceptual analysis of Greek archaic literature in general has taught us not to apply our post-Platonic notions of unity to earlier works.2 We also do not, especially in our present interdisciplinary generation, regard ancient Greek ethnography and history as formally separate fields. Modern literary criticism outside the classics, for its part, has abandoned the requirement of coherence in a text; we now revel in its inconsistencies, contradictions, and omissions.3 But the fact remains that the ancient modes of discourse of history and ethnography, as these genres are combined in Herodotus’ work, are objectively different. The one diachronically recounts unique events of the past and relies on chronological and causal continuity; the other synchronically describes permanent conditions and customary actions in the present in a discontinuous catalogue form. We are always bound to seek, therefore, a better understanding of their mutual cooperation and of the substantial connection between their respective aims. So far, the best solution of all has come from those who have broadened Herodotus’ subject matter almost ad infinitum. They have attributed the diversity of his material to his attempt to explain the human 1. Jacoby (1913, 281– 362, especially 327– 34, 341– 55) attributes the genesis of the work to a conglomeration of originally independent logoi; the ethnographic logoi were composed earlier and have been integrated into a historical context with little modification. For a later developmental view, see von Fritz 1936. The unitarians either reject the notion of development and advocate Herodotus’ adherence to an original plan (see, e.g., Lattimore 1958) or focus on the conceptual unity of the work as we have it. See especially Pohlenz 1937, on the assumptions of which the work of both Immerwahr (1966) and Lateiner (1989) is based. The integration of the ethnographic logoi and other digressions in Herodotus is discussed by Cobet (1971), who also provides excellent surveys on unitarians and separatists (1– 44, 188– 98). 2. See Frankel ¨ 1924; Van Groningen 1958. Cf. Pohlenz 1937, 67– 87. On the history of the notion of literary unity in ancient Greek practice and criticism, see Ford 1990. 3. See Belsey 1980, 109.

Introduction

3

condition in a global way and with whatever evidence was available to a prose writer.4 This is too vague a recognition, however, for it does not eliminate the notion that Herodotus has lingered along the way of a goaldirected narrative—leading to his account of the Persian Wars—to include many items of information just “interesting for their own sake.”5 The problem of unity in the Histories, in other words, is strictly connected with that of their purpose. We should approach both issues by asking, What is the basic relation between the ideological position that emerges from Herodotus’ history of the conflict with the other and that which he transmits through his description of multiple others? My representation of what the Histories communicate to their intended audience is especially indebted to two very different but mutually complementary insights of recent scholarship. The first of these has been most lucidly formulated by Fornara as a general principle and through specific illustrations from Herodotus’ narrative. Herodotus wrote and composed his history of the Persian Wars over a period of time in the 440s, the 430s, and perhaps even the 420s and later: these are the years of Athenian imperialism, ever more threatening and oppressive to some of the Greeks before and during the first stage of the so-called great Peloponnesian War against Sparta and her allies.6 Herodotus’ perspective is therefore comparable, for example, to that of an author writing about World War I in the course of World War II.7 His view of the past is bound to be 4. According to Immerwahr (1966, 5), Herodotus “first discovered history as a method of understanding the world as a whole.” Cf. Cobet 1971, 177– 87. Lateiner (1989, 16) suggests a richer and more specific set of answers when he asserts that “ethnographic information in the Histories . . . is documentation deployed to assert an historical thesis, namely that mankind has benefited from ethnic and political separation and self-determination.” 5. The most radical denial of purposefulness in Herodotus comes from Howald ( 1923, 1944), but see also Frankel ¨ 1924, especially 87. The notion that something in Herodotus may be there simply because “it is itself worth reporting” in some obvious but unspecified way is still pervasive in some of the best Herodotean scholarship: see, e.g., DarboPeschanski 1987, 11; Fornara 1971a, 21, 23. 6. The terminus post quem for the publication of the Histories as a whole is 430 b.c., the time of the latest datable event mentioned in the work (7.137). On the basis of more indirect internal evidence and echoes of the Histories in other texts, Fornara (1971b; 1981) argues for a publication date close to 414 b.c. Cobet (1977) favors the more orthodox date of ca. 425 b.c. The composition and dissemination of Herodotus’ work must in any case have stretched over a number of years; see Evans 1991, 90. My argument does not depend on precise dating after 430 b.c., but I will briefly revisit the issue in chap. 3, “Divine Communication.” 7. I am modifying for the sake of simplification a more subtle parallel made by Fornara (1971a, 40).

4

Telling Wonders

affected by later developments and by the momentous political circumstances and ideological trends of his own time. In the case of Herodotus, observes Fornara, the interpretation of the past in light of the present is merely implicit: he only infrequently mentions the historical developments in Greece after 479 and even eschews many opportunities to do so in the course of his narrative. This is remarkable in an author who is so interested in complete historical processes that he habitually jumps backward and forward in time to show how a story began or how it ended. What we have at several points, then, is evidence of a deliberate silence about recent history and present circumstances, coupled with an ironical intent in the report of earlier events.8 Herodotus’ narrative capitalizes on the audience’s knowledge of how things turned out and draws its force and meaning from those later outcomes. By describing a narrative of the past shaped by the circumstances of the present, Fornara draws us toward envisioning the Histories as a work that, through the narrative of their recent past, communicates to the Greeks (Herodotus’ implied audience) things they should learn about themselves.9 This is, after all, a traditional role of historical or mythical narratives documented elsewhere—in the speeches of Homeric heroes, Pindar’s epinician does, or the funeral orations of Athens. None of the narratives told by speakers in the Histories claim any other goal or effect, such as entertainment or pleasure.10 It is in conjunction with this didactic character of the Histories that I wish to explore the role of the ethnographic descriptions. These occupy a considerable portion of the narrative, yet Fornara excludes them from his argument about the relevance to the work of the historical and political circumstances of Herodotus’ time. In his view, the history of the Persian Wars, particularly the last three books of the Histories, are one thing, 8. See Fornara 1971b, 152– 53. All Herodotean references to events after 479 b.c. are listed and discussed in Cobet 1971, 59– 78. 9. See Fornara 1971a, 75– 91; Raaflaub 1987; Corcella 1984, 186– 235; Munson 1988; Stadter 1992; Moles 1996. This direction of Herodotean scholarship is partially indebted to Strasburger 1955. 10. This is unlike the songs of Demodocus and Phemius or Odysseus’ narrative in the Odyssey (1.325– 52; 8.62– 92, 266– 369, 471– 541; 11.333– 476). See Marincola 1997b, 11. Although ancient readers emphasize the enchanting effect of Herodotus’ narrative (see Dorati 2000, 33– 47), the internal model for reading the Histories is provided, e.g., by the useful stories of Solon (1.30.4– 5), Socles (5.92), and Leotychides (6.86α– δ). See Stadter 1992, 781– 82; 1997. Within the narrative of the Histories, one audience who sets out to listen to a (musical/poetic) performance for the sake of pleasure is discredited (1.24.5). Cf. 6.21; 6.129.

Introduction

5

while the ethnographies are quite another: they might as well belong to another author. I have no objection to Fornara’s support of Jacoby’s theory, for example, that the Egyptian section was composed earlier than most other parts of the work and that it bears signs of having been written as an ethnography and not at all for the purpose of occupying the place it now does within a history. To explain the composite nature of the work and in particular the stylistic peculiarities of book 2, it would help to postulate that Herodotus may have started his intellectual development by looking outward at exotic phenomena, before directing his gaze inward at the Greek world in his maturity—that he tended toward a synchronic approach before turning to a diachronic research of causes.11 But the Egyptian and other ethnographies have been interwoven with the historical sections in the extant work.12 In light of the insights inspired by Fornara himself about the message Herodotus conveys to his Greek audience about themselves, it is time to reconsider what this message is, whether it informs all of the Histories or merely certain parts, and whether the ethnographies dilute and put it on hold or contribute to it in the special way that is consonant with their genre. The argument that the Histories communicate a teaching to a contemporary audience by means of narrative leads us to my second interpretive tool for reading Herodotus. This is offered by Gregory Nagy in his Pindar’s Homer, a book that aims at explaining the common roots and parallel developments of different modes of discourse of distinct archaic and classical genres. Within this broader discussion, Nagy approaches Herodotus’ Histories as a performance based on that same tradition of the ainos that became embodied in other types of performance: in the fables of Aesop on the prose side and in the poetry of Hesiod, Archilochus, Theognis, and Pindar. According to Nagy, Herodotus’ performance of historie is the prose counterpart of the ainos of Pindar’s epinician odes. Both are authoritative speech acts with the power to convey explicit as well as implicit 11. See Fornara 1971a, 1– 23. The distinct features of Herodotus’ persona in book 2 are discussed by Marincola (1987). However Fornara’s statement (1971a, 18– 19) that book 2 differs from other parts of the work on account of “the utter absence of the moral or philosophical element” seems wrong to me. 12. As Drews (1973, 84– 85) observes, the Histories only go to prove that Herodotus still had ethnographic interests at the end of his career. Drews’ explanation, however, that Herodotus’ descriptions of foreign peoples have the function of magnifying the Persian Empire and therefore the accomplishment of the Greeks in resisting its attack does not account for a great deal of ethnographic material in the Histories.

6

Telling Wonders

messages.13 Pindar places the here and now of the athletic victory in direct relation with the heroic achievements of the past by conferring epic kleos (renown) on the members of a privileged group within the aristocratic class of his time. At the same time, he also implicitly warns about the impending danger of degeneration of the aristocracy itself and of the consequent rise of a tyrant in the polis.14 The Histories of Herodotus, for their part, deal centrally with a relatively late historical past and do so more ambivalently than Pindar treats his paradigmatic time of heroes. The sociopolitical composition of the implied audience, the ideological stance of the text, and the specific substance of the message it conveys differ in many important respects from their counterparts in Pindar. But the past and the present, the explicit and the implicit, praise formulated in terms of kleos and warning about the threat of tyranny for the state, and a message of certain retribution for hubris based on the moral ideology of Delphi are all part of Herodotus’ discourse, as they are of Pindar’s.15 Nagy’s parallel between Herodotus’ historie and the ainos dovetails with Fornara’s and Raaflaub’s insistence on the relevance of the historical context of Herodotus’ time to the narrative of the Histories.16 Herodotus’ deliberate silence concerning events closer to his own time, emphasized by Fornara, is suitable to the implicitness of the ainos, to the obscurity of the related tradition of oracular poetry, and to the position of an author who is advising a politically diversified public about sensitive contemporary issues.17 The approaches of both Fornara and Nagy have provided foundations to my understanding of the Histories as a speech act bearing advice and warnings that address the present attitudes and behavior of the Greek cities toward one another and their future prospects. From the point of view of Nagy’s analysis, our initial question about the role of the ethnographic material in the Histories becomes, To what use has the genre of Greek ethnographic writing been put as part of a 13. Nagy 1990, 215– 338, followed by Payen 1997, 62– 74. 14. The features of the ainos of Pindar are described by Nagy (1990, especially 146– 51, 164– 66, 173– 74, 181, 186– 87, 192– 99). 15. For the parallel between the mode of discourse of Herodotus and that of Pindar, see Nagy 1990, 215– 338. 16. Fornara 1971a, 75– 91; Raaflaub 1987; Nagy 1990, 305– 16. Nagy characterizes Herodotus’ narrative about Croesus the Tyrant as “an ainos which applies to Athens” (309). 17. See Nagy 1990, 332– 35 for the element of ainigma (riddle) in the ainos and for the connection between Herodotus and the ainos, on the one hand, and wisdom and oracular poetry, on the other.

Introduction

7

discourse analogous to the ainos? One sign of adaptation of ethnography to the mode of the ainos in Herodotus is perhaps the one and only quotation from Pindar to be found in the Histories, the gnomic statement “custom is king of all” (3.38.4). Herodotus twists it out of its Pindaric meaning and gives it his own, precisely to use it as an interpretive bridge between his historical narrative about a hybristic king and his entire ethnographic project of describing customs. Moreover, in his polemical essay On the Maliciousness of Herodotus, Plutarch attributes to Herodotus the unitarian agenda of praising the barbarians and criticizing the Greeks and accuses him of doing so “deceitfully.”18 Later on, we will have other opportunities to test the extent to which Plutarch’s rather pedestrian view can be credited. For the moment, we should notice that what he calls “deceit” is related in part to Herodotus’ speaking obscurely through the narrative of the past, the words of oracles, and the representation of foreign peoples. In reference to a particular passage, Plutarch says that Herodotus there uses the Pythian god as his mouthpiece just as at other times “invents words and assigns them to Scythians, Persians, or Egyptians in the way that Aesop assigns his to ravens and monkeys.” Since Aesop’s fables are ainoi—narratives whose implicit, often political teaching could be understood from the context in which the author performed them—Plutarch’s assertion confirms Nagy’s argument about the analogies between Herodotus’ discourse and that of the ainos.19 Although the discourse of Herodotus feeds on the same traditions and assumptions that govern the ainos, Nagy rightly insists that its essential component is historie, “inquiry.” This is the process (and the product) of collecting evidence by seeing what is possible to see and by hearing the οδεξις, available verbal testimony: the Histories are a ι στορ ιης απ which means the public presentation of those accumulated data. Herodotus’ historie is a “scientific” undertaking, but it is also comparable to the process of inquiry in cases of judicial arbitration.20 In an inquiry of arbitration, the evidence collected serves as the basis for a judgment about who is right and who is wrong in a dispute and for the recommendation of a settlement on 18. See especially Plut. De Malign. Herod. 12– 14 ⫽ Mor. 857A– F. 19. Plut. De Malign. Herod. 40 ⫽ Mor. 871D. Concerning this passage and the connection between Herodotus and Aesop, see Nagy 1990, 322– 26, with evidence from the Life of Aesop on the context in which Aesop told his fables. See also Payen 1997, 69, 71. 20. See Nagy 1990, 315– 20, for the historie of Herodotus as a juridical process in the light of attested cases of arbitration between Greek states. On Herodotus’ scientific attitude and rhetoric, see Lateiner 1986; Thomas 1997; Thomas 2000.

8

Telling Wonders

the part of the arbitrator or histor. In a similar way, the information presented in the Histories counts as evidence in view of judgments and recommendations by the histor Herodotus.21 The dispute is ultimately that among the Greek city-states in the time of Herodotus himself. The judgment and the recommendation represent the message of the Histories as a whole. This is communicated for the most part implicitly, through evidence that is both ethnological and historical. Assyrian Boats and Other Particulars The discussion of Greeks and barbarians that pervades Herodotus’ work is based on a dialectic between traditional notions of the Greeks, on the one hand, and Herodotus’ more or less overt disruption of these notions, on the other. Though recent scholarship has enhanced our understanding of Greek ideology about the barbaroi, it has not paid sufficient attention to the extent to which the several sources that constitute our evidence for this ideology respond to it in particular ways.22 Drawing from a generalized perception partially disseminated through poetic texts and owned, as it were, by his audience, Herodotus lures his modern readers into a false sense of recognition, only to undermine it. Though he represents otherness according to culturally determined— one might say, unconscious and inevitable—patterns of thought, he devalues its familiar implications through a series of concomitant strategies. He occasionally sets up the other as a model of what the Greeks would consider appropriate behavior; he complicates knowledge to both confirm and confuse ideological stereotypes; or he counterbalances his representation of difference with indications of unexpected similarities between his ethnographic subjects, other groups of barbarians, and different groups of Greeks.23 Herodotus’ ethnographies are thematically unitarian descriptions, each discussing the customs, geographical situation, and historical traditions of this or that people. They are made up of particulars, which contribute to a 21. The juridical meaning of histor is explained by Nagy (1990, 262, 315– 20). See also Connor 1993. Dewald (1987) applies the term to Herodotus for the sake of a different distinction: to indicate the narrator and implied author of the Histories as he emerges from the work (on narrator, implied author, and real author, see Booth 1983, 67– 77; Chatman 1978, 147– 58; Genette 1980, 213– 14). 22. See, e.g., Hartog 1988, xxiii; Rosellini and Sa¨ıd 1978; Hall 1989. 23. See Pelling 1997.

Introduction

9

general representation of the people in question but also possess a measure of autonomy separate from the monographic section to which they belong.24 The uncertainty about the role of ethnography in the Histories arises again and again from many individual ethnographic details, posing each time a series of typical questions: What is the meaning of this feature? How does it contribute to characterize a given ethnic group? In what other ways does it relate to its ethnographic surroundings or to any other part of the work? Examining Herodotus’ discussion of the boats of the Assyrians (1.194) will provide a preliminary demonstration of the value of particulars as evidence in the political discourse of the Histories. In the phrase that introduces the description, “the greatest wonder to me in that land, after the city itself” (1.194.1), the city is Babylon. Herodotus has discussed it earlier, right after the narrative statement that Cyrus, having subdued many other peoples of Asia, now planned to conquer it (1.178– 87). Babylon is an architectural marvel, defined by its river and bearing the stamp of royalty. Surrounded by a double circle of walls, it stretches on both shores of the Euphrates, with one of two symmetrical structures rising at the center of each half: the sanctuary of Zeus Belos and the royal palace (1.178– 83). Babylon has been built and adorned by its rulers. The two clever queens whom Herodotus singles out  for mention have added to its wondrous quality (αξιον θωµατος, 1.185.3; αξιοθ εητα, 1.184). One has endowed the city with impressive dikes to prevent flooding. The later one, Nitocris, has detoured the course of the Euphrates by means of canals to the north and an artificial lake, to protect Babylon from the Medes. By draining the waters of the river into the lake and letting them flow back again, she also built across the Euphrates a bridge that connects the two sides of the city in daytime and is removed at night to prevent crime (1.184– 86). The already announced narrative of Cyrus’ campaign against Babylon resumes after the description of the city. Accompanied by mule-drawn wagons carrying royal food supplies and the drinking water of his native Choaspes River in silver vessels, the Persian king crosses another stream, the Gyndes, which he divides into 360 channels in punishment for drowning one of his sacred white horses (1.188– 89). Water and mighty walls seem to make Babylon impregnable. Cyrus, however, turns the hydraulic works of Nitocris to his own advantage: he drains the Euphrates into the artificial lake she had built and invades the city through the riverbed. 24. See Darbo-Peschanski 1987, 138– 39.

10

Telling Wonders

Distracted by festive pleasures, the Babylonians do not see him coming, and the city falls into Persian hands (1.190– 92.1). The campaign narrative is followed by another, more properly ethnographic description, this time of the Assyro-Babylonians and their land. This description begins by listing the resources the newly conquered province added to the Persian Empire, amounting to a third of those provided by Asia as a whole (1.192). A chapter on agriculture (193) explains this fabulous wealth: a hand-operated irrigation system makes Assyria the most fertile of lands. Cereal plants grow to unbelievable proportions. Oil is derived from sesame, and the palm trees yield a fruit used to make food products as diverse as bread, wine, and honey. Here, the boats carrying the merchandise down the Euphrates are introduced. In the various sections in the ethnohistorical Babylonian narrative so far—the description of the city, the account of Cyrus’ campaign, and the Assyrian ethnography—dealing with the river has been a test of sophie (cleverness) by which all the protagonists are evaluated, either implicitly or explicitly: the rulers of Babylon used the river to increase its safety and order; the foreign king Cyrus conquered Babylon by means of waterworks; the careless city dwellers let themselves be caught unprepared; unspecified people authored the irrigation system; and finally, when we reach our passage, the Assyrians take advantage of the current of the Euphrates for their journey from Armenia to the markets of Babylon. These merchants mark a transition from royalty to the society whose customs are then described in areas of culture common to all nations around the world—dress, marriage, health care, funerals, and religion. Sophie remains intermittently a factor (see 1.196.1; 1.197), down to the report of the peculiar Fish Eaters, who, as the other natives do with palm dates, are able to process a single ingredient into various forms (1.200). Like all texts, this narrative and descriptive section is intersected by a number of different and partially overlapping cultural codes or subcodes.25 As a resource or an obstacle, the river is not merely a term of the 25. On the notion of “codes,” see Barthes 1970; Munson 1991. The primary code of any text is the linguistic (English, Greek, etc.). Second- or third-order cultural codes (more properly called “subcodes”) are represented by the words and ideas used when speaking about something: in a typical nineteenth-century fictional text, e.g., Barthes (1970, 18– 20) identifies a proairetic code (code of action), a hermeneutic code (signaling the existence of a mystery or progress toward its resolution), a semic code (or code of connotation), a symbolic code, and an indeterminate number of cultural codes that communicate by making reference to a certain body of cultural knowledge shared by the implied audience (e.g., the terminology of a medical textbook will constitute in effect a special language, a technical medical code).

Introduction

11

codes of action, geography, and military strategy but also a symbol of what is given, one’s proper share, or that which delimits it.26 It participates, in other words—as is frequently the case with concrete features in Herodotus’ narrative—in the symbolic code of the text.27 In relation to the river, another code emerges, that of kingship, and the actions of kings also acquire symbolic meaning.28 Characteristic of both native and foreign rulers in the Assyro-Babylonian section is a control over the natural environment, a control that appears morally problematic and therefore dangerous. Cyrus’ royalty is defined at the outset by his ownership of the water of the Choaspes, which he and all Persian kings carry around for their own special use wherever they go (1.188). Cyrus’ emasculation of the Gyndes corresponds at long range, backward and forward, to all the instances of expansionist violations of rivers in the Histories. The first of these occurs when Croesus initiates war against Cyrus and crosses the Halys, which some say he split into two streams (1.75). The last are represented by Xerxes’ abuse of the Hellespont (7.35) and by the advance against Greece of his large Persian force, which drinks rivers dry.29 Unlike the violations of Croesus and Xerxes, Cyrus’ assault on the Gyndes has no negative results. When he drains the Euphrates and captures Babylon and its wealth, his cleverness is crowned with success. Ethically, however, his conquest is on a continuum with two later royal thefts by Darius and Xerxes, mentioned in the same narrative (1.187, 1.183.3). Cyrus himself will meet with an unfortunate end in due time, after crossing another stream, the Araxes, in his attempt to subjugate the Massagetae (1.208).
26. Croesus’ crossing of the Halys involves an attempt to acquire land beyond his µοι ρα, “share,” (1.73.1). The symbolic interrelation of physical and moral boundaries in Herodotus is discussed by Lateiner (1989, 126– 44). See also Immerwahr 1954, 19– 28; Immerwahr 1966, 325; Konstan 1983; Stadter 1992, 785– 95; Payen 1997, 138– 45. 27. Physical objects as conveyors of meaning in Herodotus are discussed in Dewald 1993. On the symbolism of metals, see Kurke 1995. Symbolic action in Herodotus is discussed by Payen (1997, 29– 31 and passim). Any word or narrative element in a text normally belongs at once to many different codes and to codes within the codes (e.g., the word heart belongs to the codes of love, physiology, medicine, cardiology, etc.). The symbolic code draws terms from overlapping codes and helps to reshuffle and reassign them to yet other codes. So in Herodotus, the river becomes, through the symbolic code, a part of the political code and the code of ethics, the languages in which we talk about such issues. 28. The actions and features typical of kings (i.e., all the terms of the “monarchical code”) are also terms of the symbolic code when they not only apply to real kings but become the means for talking about, for example, the human condition and human actions in general. See chap. 2, “Analogy as an Interpretive Tool.” 29. See 7.21.1, 7.43.1, 7.58.3, 7.108.2– 109.1, 7.109.2, 7.127.2. There are twentyfive rivers mentioned in the account of Xerxes’ march from Critalla to Thessaly alone (7.26– 131).

12

Telling Wonders

The digging, draining, and bridge building of the Babylonian queens have more benign motives and do not entail trespassing into someone else’s land. These acts also represent, however, monarchic imperialism over the environment. Like Nitocris’ attempt to impose order on Babylon from above, these acts are useless given the citizens’ lack of discipline and prove counterproductive in the face of the enemy attack. The folks whom Herodotus describes as making their living through the river have neither leisure for lapses nor monarchic opportunities to exercise intrusive mastery on earth and waters. The boats of Assyrian merchants are constructed of inexpensive materials (timber, hides, and straw), which are then recycled and reassembled. Unlike royal masonry work, they do not mark the land. To some extent, like the leather trousers of the Persians of long ago (1.71.2), they connote primitivity.30 They are crude in shape, all round, with no stern or prow. More “like shields” than like ships, reactive rather than offensive, they are designed not to cross the seas, to be steered in different directions and against the waves, or to dominate the elements but to drift downstream helped along by paddles. Very large or less large, they only require a crew of two men. Once in Babylon, the merchants sell their wares and the framework of their boats, then retrace their way by land on their donkeys, because the swiftness of the current prevents them from sailing back upstream. They display not only industry and sophie but also what we call common sense and the Greeks would include in their notion of sophrosune.31 The description of the collapsible boats and other details in the ethnography characterize the Assyrians in a way that on the whole tends to redefine the issue of difference: though their boats are peculiar, their market of brides preposterous, and their diet exotic, a talent for problem solving makes the Assyro-Babylonians similar to the Greeks. Sophie is a virtue the Greeks claimed to possess to a higher degree than all barbarian nations (see, e.g., 1.60.3). To counteract such ethnocentric assumptions by means of specific evidence and praise is an important goal of Herodotus’ historie. 30. Connotation, activated through signs or indices (the semic code: cf. n. 25), is closely akin to symbolism but involves a more immediate and less constructed connection between signifier and signified. See Barthes 1970, 6– 9, 19; Prince 1987, s.v. “seme.” 31. See Geertz 1983, 71– 93, on our tradition of regarding common sense as a transcultural category (“what anyone clothed and in his right mind knows”). On sophrosune, see North 1966, especially 28, where North notes that although Herodotus uses sophr- words sparingly, he “is the most fertile source in Greek prose of stories illustrating traditional ideas of sophrosune.”

Introduction

13

At the same time, however, Herodotus’ particulars often appear, as in this case, to have less to do with the construction of the national identity of a foreign people than with projecting onto a faraway setting pieces of a problematic that is entirely Greek. One well-known ideological contradiction in the mid– fifth century was that of the simultaneous desirability of exceptional (i.e., symbolically “royal”) status and normal (or citizen) status, as is illustrated in the Histories, for example, by the contest of happiness between Croesus of Lydia and Tellus of Athens.32 Throughout Herodotus’ work, the representation of the otherness of foreign peoples competes with the representation of the cross-cultural otherness of kings. By setting the pragmatics of everyday life, exotic but legitimate, side by side with the behavior of the powerful agents of history, his ethnographies represent a crucial part of a discourse at once “democratic” (almost in the modern popular sense of the word) and anti-imperialistic. To the talking Scythians, Persians, and Egyptians whom Herodotus uses, according to Plutarch, for saying what he wants to say, we may here add the silent Assyrians. The description of their boats, like the ainos of Tellus, provides a foil for the actions of kings. The ethnographic style of the Histories, in the Assyro-Babylonian logos and elsewhere, consists mostly in the factual description of particular features, with little interpretation of their meaning. What has been characterized as Ruth Benedict’s typical mode, the general description of a people’s worldview, repeated again and again and illustrated by examples of what they do and say, is in general alien to Herodotus.33 In the Histories, the point of what “they” do and the reasons why “we” should find it interesting have to emerge from the description of what they do. The near absence of explicit interpretive commentary fits the tradition of Greek ethnographic writing as well as Herodotus’ mode of implicit communication. In the case we have examined and several other times in the Histories, celebration of the subject stands in lieu of interpretation. The initial sentence of the passage communicates to the recipient of the narrative that the feature is important or, in other words, meaningful. The phrase  “great wonder” [θωµα µεγιστον] establishes its membership in the broad category of “achievements great and wonderful, some performed by Greeks, some by non-Greeks,” that, according to the first sentence of the 32. 1.30.3– 5. On the Greek political and ethical discourse concerning tyrannical power, see McGlew 1993, especially 30– 32, 196– 212. On kingship as symbol, see my n. 28 in this introduction. 33. For the description of Benedict’s style, see Geertz 1988, 108– 20.

14

Telling Wonders

work, are deserving of permanent renown. The researcher and narrator named in that first sentence, Herodotus of Halicarnassus, here comes into the open—with “the greatest wonder to me”—to claim the inclusion as his own idiosyncratic choice. When he says, “the greatest wonder to me after the city,” he creates a paradox and helps to establish at the outset the symbolic nexus between two discrepant items, “leather boats” and “Babylon,” the people and their kings. This is, on the whole, a more eloquent directive than simply a red mark on the page. It is almost as eloquent and interpretive, in an implicit way, as the famous title that announces one of Ruth Benedict’s ethnographic inquiries, The Chrysanthemum and the Sword.34 The Histories as Performance Announcements and glosses such as the one we have just considered signal the presence of the one who is telling the story and his reactions to it. They represent the metanarrative component of the discourse, the study of which is a tool for better understanding the message, ideology, and tone of a text. In the case of Herodotus, the metanarrative often enhances our awareness of a direct verbal communication from a speaking author to a listening audience. The Histories as we have them are obviously a written text, and they contain internal signs of writing as a tool of composition and publication.35 These signs are less conspicuous, however, than those indicating or imitating an oral situation for Herodotus’ speech act.36 As Nagy maintains, the work is “not a public oral performance as such,” but it is nevertheless “a public demonstration of an oral performance, by way of writing.”37 34. Benedict (1989, 2), with characteristic explicitness, explains her title: “Both the sword and the chrysanthemum are part of the picture. The Japanese are, to the highest degree, both aggressive and unaggressive, both militaristic and aesthetic, both insolent and polite . . .” 35. Examples are the authorial “I write” occurring side by side with “I say” in programmatic statements (see chap. 1, “Types of Introductions and Conclusions” and “SelfReferential Glosses”); long genealogies and catalogues (see Thomas 1989, 189; Harrison 1998); perhaps the past tense in metanarrative sentences of the type “it could be seen still in my time” (see Rosler ¨ 1991); and, of course, the statement in the first sentence that Herodotus intends to preserve the glory of great deeds and save them from oblivion. 36. See Pohlenz 1937, 208– 11; Lang 1984, especially 1– 17; Munson 1993a; Stadter 1997. 37. Nagy (1990, 220) so argues on the basis of the first sentence. For the notion of performance applied to the Homeric poems, see Martin 1989, 1– 42. We do not, of course,

Introduction

15

We have some external evidence, more or less direct, of historical Herodotean performances.38 Our written text may theoretically stand in relation to these as a special version of a larger potential performance, a reservoir for potential performances, or a collection or summary designed to extend the life and usefulness of actual or possible contemporary performances. Since I approach the Histories in this way, considering it a text that is unitarian but not definitive, I will not emphasize its overall structure (each narrative in the exact place in which it stands) and integrity (the original presence of all the narratives it now contains) as primary conveyers of meaning.39 Thematic and factual cross-references so multifariously pervade the work and spill over onto the unmentioned real world of Herodotus’ audiences that, to communicate its message, a given narrative may relate in fairly specific ways to the immediately surrounding narratives though it is not entirely dependent either on them or on what is now the whole. The narratives of the Histories have the potential for joining up with many other narratives that are now far away or simply not there at all. Like the Assyrian boats, Herodotus’ Histories lend themselves to being disassembled and reassembled into smaller (or, theoretically, larger) functional wholes. The apparent elasticity of the Histories recalls to some extent the circumstances of rhapsodic performances of the Homeric texts in classical times. It is also internally indicated by their relative open-endedness, by the unfulfilled promises they contain of narratives that could have been included but are not, and by the contrived and not inevitable character of many of the transitions between narratives.40 The one and only postulate the same sort of orality for the Histories as one may for the composition of the Homeric poems. But see the parallel drawn in Murray 1987, 107. Most importantly, Herodotus’ use of writing at a moment of transition between preliterate and literate conditions does not entail a widespread readership from the beginning, and an audience of listeners is more likely. On the related issue of literacy, reading, and the aural reception of literature in the fifth century b.c., see Davison 1962; Flory 1980; Havelock 1982, 146– 47; Thomas 1989, 15– 34. 38. See Plut. De Malign. Herodot. 26 ⫽ Mor. 862A– B (⫽ Diyllus, FGrHist 73 F 3); Lucian Herodot. 1– 2; Marcellin. Life of Thuc. 54; Eus. Chron. Olymp. 83.4. This evidence is rejected by Powell (1939, 31– 36) and Johnson (1994). But see Parke 1946, 86– 87; Momigliano 1978, 62– 66; Evans 1991, 89– 132; Dorati 2000, 17– 52. 39. Structure has been emphasized by Wood (1972, especially 19) and productively studied by Immerwahr (1966). See, however, Immerwahr 1966, 308. I mean not to devalue the meaningfulness of the present structure but to suggest that other combinations of logoi would be just as meaningful. 40. On the end of the Histories, see Dewald 1997, with full bibliography on the previous scholarly discussion. On its unfulfilled promises, see Drews 1970 (discussing

16

Telling Wonders

performance available to us, in other words, possesses a retroactive unity and also partakes of the incompleteness or fluidity of the various potential performances of which we can conceive. As a display in front of an audience, the Histories are a “performance” in the ordinary sense of the word. They are also a “performance” in a narrower sense, from the point of view of speech-act theory. They represent an utterance that does something, performing a range of worldchanging operations. At the surface level, they inform, preserve the memory of events, and establish a record of praise and blame. Implicitly and at a deeper level, they predict, advise, warn, attempt to persuade, or promote a certain predisposition in the audience.41 The substance of these implicit predictions, advice, and warnings represents, once again, the message of the work. The various functions (or illocutionary acts) that the Histories perform at different levels through the narrative are facilitated by the metanarrative. Narrative statements (whether recounting or descriptive) are, theoretically, world-describing utterances: any other eventual functions they might perform (e.g., warning) are especially indirect and disguised.42 The narrative records what is real, simply because it is real. With the metanarrative, however, the author of the narrative emerges from the text. As an explicit mediator between reality and the audience, he communicates to the audience about his record of reality and its reason for being.43 1.106.2 and 1.184); Nagy 1990, 235 n. 91 (on 7.213.3). The artificiality of Herodotean transitions between different logoi is noted by Jacoby (1913, 343– 46). Within modern (written) literature, an approximate parallel to the rhapsodic nature of the Histories is provided by what Genette (1980, 149) calls the “unity after the event” of Proust’s A` la recherche du temps perdu, a work made up of heteroclyte material that was being continuously rearranged and revised by the author even during the last stages of publication. See also Eco’s (1962) definition and description of what constitutes an opera aperta and an opera da finire. 41. For the conjunction of the ordinary and speech-act meanings of the terms performance and performative, see Martin 1989, especially 5, 47. Speech-act theory was initiated by Austin (1962); for subsequent developments, see especially Searle 1976; Bach and Harnish 1979. Application of speech-act theory to works of literature is discussed by van Dijk (1976), Pratt (1977), Chatman (1978, 161– 66), and Searle (1979, 58– 75). 42. They are true-or-false statements, such as “Napoleon won the battle of Austerlitz” or “the earth is flat” (see Austin 1962, 132– 49). In Searle’s subsequent classification of five categories of illocutionary acts (1976), they are called “representatives.” 43. The presence of the narrator as a more or less overt mediator of narrative statements (and his absence in “nonnarrated” narratives) is discussed by Chatman (1978, 146– 266). See also Genette 1980, 212– 62; for Homer, see Richardson 1990, especially 1– 8. For Herodotus, see Dewald 1987, forthcoming a; de Jong 1998.

Introduction

17

Metanarrative signs are of course embedded in the narrative throughout. Narrative as pure transcript of the real world exists only in theory, because words can achieve exact mimesis of other words only, not of actions and states.44 But in Herodotus we are faced with an especially massive metanarrative component consisting of introductions, conclusions, and glosses that in some way summarize and explain the narrative (“X is like Y,” “I will tell X,” “X means Y,” “I have finished with X”).45 These statements stand out from minimally narrated narrative statements: Prince writes, “their appearance is similar to that of a (fragmentary) text in the text, representing a language that is other in the language of the text and establishing some of the interpersonal coordinates of a communicative situation.”46 Many contain a grammatical sign of the narrator (e.g., “the greatest wonder of all to me . . . I am going to describe”), but others do not. Since they manipulate the narrative, they are also more likely to evaluate, convey advice, question, argue, promise, express the narrator’s state of mind, and even bring a state of affairs into existence (e.g., convict)—in sum, to reveal purposes that go beyond simple representing or reporting.47 The Contents of This Book Different types of metanarrative sentences are briefly classified in chapter 1, which serves to establish the basic distinctions and terminology on which the rest of the analysis in this book is based. I then proceed to examine two overlapping functions pursued in Herodotus’ discourse by narrative and metanarrative means: comparison and interpretation. The 44. Hence, although we say that nonnarrated narrative (reality recording itself) and even its closest approximation, minimally narrated narrative, are theoretically “mimetic,” Plato uses the term mimesis only in reference to narrative in its dramatic form (Rep. 392d– 394c). In the case of a speech reported by a narrator in oratio recta, no mediation is visible between “reality” and the text. By contrast, actions require the mediation of a narrator, who puts them into words through the narrator’s own discourse. 45. Dewald (1987, 148– 50) compares Thucydides’ narrative surface to “perfectly transparent glass through which one is encouraged to imagine one is directly perceiving the res gestae narrated” and argues that Herodotus’ narratorial interruptions resemble rather “those little decals—flowers, rainbows and whatnot—scattered by the cautious on the surface of the glass.” 46. This quotation, well suited to Herodotus’ discourse, is part of Prince’s description of metanarrative in general (1982, 127). 47. See, e.g., Herodotus’ “conviction” of Ephialtes (7.214.3), discussed in my chap. 3, “Interpretation in the History.”

18

Telling Wonders

first of these crisscrosses the text in two separate directions: diachronically for the events of history and synchronically for ethnographic and other phenomena separated not by time but by space. Figuring out how each of the two types work and how they cohere with one another is the task in chapter 2. Through the second and most encompassing function, interpretation, the text signals what facts mean and evaluates their worth. Here once again, history and ethnography go their separate ways. As I show in chapter 3, the political message of the work depends on the cooperation between a relativistic ethnographer, who interprets little and evaluates cautiously, and an absolutist historian, who explains historical action in moral terms. Herodotus’ message is nuanced, communicated not without obscurity and even, despite the charm and lightheartedness of his style, a certain degree of effort. It requires that we bridge our distance from the intended audience to which it is directed by a painstaking analysis that will not, however, entirely protect us from interpretive risks. The opacity of the Histories partially results from the uneasy series of mediations on which they rely. What is far away in space and time serves to talk about the present of narration, but the burning issues of the audience’s here and now are treated, as I have said, as almost unmentionable. Particulars not only communicate general principles but also provide contradictory evidence. Reality is represented through the narratives and arguments of sources and characters, some more reliable than others; the narrator often declines to unify the plurality of voices.48 Herodotus’ use of the terms of the Greek/barbarian antithesis frustrates his project of subverting that antithesis and redistributing the criteria of otherness. Herodotus’ attempt to reconcile the Greeks among themselves is based on a representation of a culturally special and homogeneous Greekness; yet this goal collides with his need to assign them to the shuffle of humanity, showing that they are almost just as different from each other as from non-Greeks. The instability and uncertainty of a logos that undermines itself as it goes along reflect the gaps and inconsistencies of the ideology to which Herodotus responds from within. They are also the mark of a historian who, if we will not find him unfailingly honest and objective, is still, in the way he displays evidence, more honest and objective than most.49 48. See Dewald 1987, 160– 63, 167– 70. 49. On Herodotus’ early reputation for dishonesty or lack of seriousness as a historian, see Evans 1968. Some modern scholars deny both Herodotus’ accuracy and the authenticity of his source citations. See especially Fehling 1989; West 1985; Hartog 1988, especially

Introduction

19

Herodotus’ speech act stands at the intersection of the three fundamental kinds of discourse. Just as we can distinguish sentences that are declarative (impart knowledge), imperative (give orders), and interrogative (request information), so we can, on the basis of these categories, identify corresponding types of text.50 As the exposition of the results of an inquiry and as a narrative of past events, the Histories conform to the declarative modality of discourse. I argue, however, that by virtue of its message, the work is also an imperative text: it “invites the reader to adopt a position of struggle rather than stability, specifically struggle in relation to something which is marked in the text as . . . existing outside discourse, in the real world.” But to the extent that the recipient of Herodotus’ logos is caught up in contradictions and invited “to produce answers [the text] implicitly or explicitly raises,” the Histories also constitute an interrogative text.51 The combination of the declarative, imperative, and interrogative modalities in the discourse of Herodotus’ Histories is inscribed in the recurrent concept of thoma, “wonder,” which is the topic of chapter 4. Appeals to wonder belong to the celebratory function of the text, and “a wonder” is a fact that wants to be narrated—incredible if true, hard to imagine if unknown. The narrator shares it as information, demands a reaction, and leaves us wondering what it means for our understanding of the Histories. With the term thoma, the text legitimizes the question that listeners and readers are bound to ask about any other apparently optional fact included in the Histories: why is this here? Since thoma words advertise onetime occurrences, lasting individual achievements, geographical features, and cultural artifacts, an exploration of their meaning will throw further light on the ability of metanarrative to bridge the distinction between ethnography and history.

364. Cf. Pritchett 1993; Fowler 1996, 76– 80. My work investigates meaning rather than factual truthfulness; an investigation of the meaning Herodotus attributes to facts he narrates would be even more urgent if he had invented them rather than seen or heard them. In practice, however, I believe that Fehling and others have not proven their case and that the gulf between the narrator and the real author is not that wide; see the more subtle proposal by Marincola (1997b, 14– 19). The dishonesty of Thucydides for the sake of advocacy is a newer issue. See Badian 1993, 125– 62; cf. Hornblower’s narratological discussion of Thucydides’ rhetoric in presenting the evidence (1994a). See Moles 1993. 50. See Benveniste 1971, 110, quoted by Belsey (1980, 90). 51. See Belsey 1980, 91– 92, for the definitions quoted.

Chapter 1

Narrative and Metanarrative

The aim of this chapter is to identify different levels of narrative in Herodotus’ text. I first define narrative in the strict sense, as opposed to metanarrative, and then distinguish self-referential from referential metanarrative. My discussion is especially indebted to three sets of works: narratological studies outside the field of classics,1 studies that apply narratological principles to Herodotus,2 and the work of other scholars who have devoted special attention to the formal aspects of Herodotus’ narrative.3 The definitions I present here are largely my own and formulated strictly as a function of my overall interpretive task. I keep unfamiliar terms to a minimum and avoid making theoretical points for their own sake. Hurried readers more interested in substantive issues of interpretation than in the approach offered here have the option of skipping this chapter and referring back to it later if needed. What Is Metanarrative? The Histories contain a multiplicity of stories shaped and held together by discourse and transformed by it into a single story with a logical, if rambling and open-ended, plot.4 Transitions between stories may be deter1. Genette 1980; Bal 1985; Chatman 1978; Labov and Waletzky 1966; Labov 1972. See especially Prince 1977, 1982, 1987; Barthes 1986. 2. Dewald 1987, 1999, forthcoming a; Fowler 1996; de Jong 1987, 1998; Richardson 1990; Hornblower 1994a; Rood 1998. 3. Especially Immerwahr 1966; Beck 1971; Wood 1972; Muller ¨ 1980; Pearce 1981; Munson 1983; Hartog 1988; Marincola 1987. 4. These narratives more or less correspond to the units Immerwahr (1966, 14) calls logoi. See also especially Immerwahr 1966, 46– 48, 329– 62. On the distinction between story and discourse, see Chatman 1978, 19: “the story is the what in a narrative that is depicted, discourse the how.” Other narratologists make more refined distinctions and use different terminologies. For example, Bal (1985, 1– 10) adds a useful definition of text as an upper layer of the communication: “a text is a finite, structured whole composed of language signs.” In other words (those of de Jong 1987, 31), “that which the hearer/reader

20

Narrative and Metanarrative

21

mined by historical landmarks along a chronological sequence, by changes of time, or by changes of place and subject matter, but always on the basis of some factual connection. On the whole, the narrative proceeds chronologically, but the discourse interrupts the story sequence by constantly introducing explanations and expansions of this or that story element.5 In most cases, these formally subordinated narratives recount events belonging to a specific previous or later story time (flashbacks or follow-ups) or are descriptions in the present tense.6 In my definition, “narrative” includes both the recounting of events in the past and description.7 Description, in whatever tense, displays objects, beings, situations, and actions “in their spatial, rather than temporal existence, their topological rather than chronological functioning, their simultaneity, rather than succession.”8 In Herodotus’ ethnographic descriptions, the present tense describes circumstances that may also obtain at the time reached by the historical narrative to which the description is attached. Whether it does or not, the ethnographic present is at any rate a real present, referring to the time of narration.9 Just as he instructs the audience about what happened in the past, so Herodotus teaches them about the contemporary world. Whereas narrative represents the story as it is manipulated by the discourse, metanarrative speaks about the narrative and exists as a function of the discourse. Minimally narrated narrative consists of passages that approximate the concept of pure narrative, or objective mimesis, of hears/reads is a text.” When I say “Herodotus,” unless the context makes clear I am indicating either the historical author or the narrator, I am referring to the “text” in this sense. 5. The discourse devices used in archaic and early classical Greek literature for connecting semiautonomous items of a chain are discussed by Frankel ¨ (1924, especially 62– 67) and Van Groningen (1958, 36– 50). 6. See Pearce 1981. Genette (1980, 35– 85) calls narratives involving a change of time “anachronic” (either “analeptic” or “proleptic”). In a few cases, Herodotus’ inserted narratives are chronologically parallel (see, e.g., 3.39– 60) or indeterminate (see 1.23– 24). 7. This definition, functional to my analysis, differs from that of most literary theorists, for whom narrative only concerns specific events in a temporal sequence of two or more. See, e.g., Labov and Waletzky 1966, 20– 32; Labov 1972, 359– 62; Bal 1985, 1– 10; Prince 1982, 1– 4; Prince 1987, s.v. “narrative.” 8. Prince 1987, s.v. “description.” 9. Hartog (1988, 254– 55) inexplicably denies this. Even if we wished to attribute a certain timelessness to the “gnomic” present, that would not apply to the ethnographic present. See, e.g., the timing of the Persian ethnography discussed in chap. 3, “Persian Ideology.”

22

Telling Wonders

external facts.10 Certain propositions, however, fall partially or entirely outside of the narrative and are equivalent to or contain titles, proems, repetitions, postscripts, or explanations that fulfill the role of glosses to the narrative itself. These metanarrative sentences especially appear as a sort of “padding” between adjacent or concentric narratives.11 At 7.57.1, for example, the minimally narrated narrative sentence “a mare gave birth to a hare” represents the core of a larger story sequence: *

s-i s-ii s-iii s-iv

Xerxes’ army crossed the Hellespont. A mare gave birth to a hare. They saw it. They proceeded on their way.

In Herodotus’ discourse, however, event s-ii stands out by itself. What precedes and follows is predominantly metanarrative, containing event s-i in a subordinated clause and incorporating event s-iii: 1a.

I

n G

When all [the Persian troops] had crossed, while they were moving on their way, a great prodigy [τερας µεγα] appeared to them of which Xerxes took no account, though it was υµβλητον]: easy to interpret [ευσ for [γα ρ] a mare gave birth to a hare. This was easy to interpret because Xerxes was about to lead an army against Greece with the greatest pomp and magnificence but would come back to the same place running for his life.

10. See my introduction, n. 44 and corresponding text. All narratives are of course “narrated” to different degrees, and we could discuss the internal signs of narration in each case. Here I am concerned with making a basic distinction. 11. For the combination of an introductory and a concluding statement framing a narrative in Herodotus, see especially Immerwahr 1966, 12, 52– 58. Cf. Pohlenz 1937, 72, 208– 10; Beck 1971, 11– 17, 57– 59; Muller ¨ 1980, 79– 80. On the concept of metanarrative, I am applying very broadly the definition by Prince (1977, 1– 2): “Chaque fois que le discours narratif (au sens large) renvoie au code qui le sous-tend ou, plus sp´ecifiquement chaque fois qu’il accomplˆıt (paraˆıt accomplir) une function de glose par rapport a` l’un de ses propres e´ l´ements, nous avons affaire a` des signes m´etanarratifs.” See also Prince 1980; 1982, 115– 28. The “shifters” discussed by Barthes (1986, 128– 30) and Fowler’s “markers of the historian’s voice” (1996, 69– 76), including, among others, all statements in the narrator’s first person (for which see Dewald 1987), are all part of the metanarrative as I define it.

Narrative and Metanarrative

23

It will become clear later why I identify statement I as an introduction and statement G as a concluding gloss rather than as a conclusion. What matters now is that both statements I and G are predominantly at a different level of discourse with respect to the central narrative sentence. Their main function is to “read,” summarize, or explain. They perform, in other words, some of the operations a reader/listener might perform, and they do so from a perspective that, like that of the recipient, is not an integral part of the action narrated. This commentary, moreover, leads the narrator to postpone s-iv after he has attached to this story the narrative of a chronologically anterior omen, similar to the one just narrated. The result is a narrative preceded by its own summarizing introduction (7.57.2), which in the present context represents a gloss to the preceding narrative of the mare/hare omen. This is followed by a sentence (CC) that both concludes preceding narratives and narrates story function s-iv. 1b. ⫽ G

I n

CC n


 Also another prodigy [ετερον . . . τερας] occurred for him when he was still in Sardis: for a mule gave birth to another mule with a double set of genitals, male and female, the male on top. Taking neither of these two into account, Xerxes moved forward. (7.57.2– 7.58.1).

A contrasting example to this set of metanarrative interferences is provided, for example, by a minimally narrated narrative reporting what Astyages learned about the meaning of his daughter’s two successive dreams and how he reacted to the information (1.107– 108.3). Astyages is the embedded focalizer of the events; whoever is telling this story (Herodotus or one of the sources mentioned at 1.95.1) is almost invisible.12 In the case of the hare giving birth to the mare, in contrast, while 12. On focalization, see Genette 1980, 185– 210; Bal 1985, 100– 118; de Jong 1987, 29– 36 (in the Iliad); Hornblower 1994a (in Thucydides); Dewald 1999 (in Herodotus and Thucydides). The narrator is first and foremost the one “who speaks”; the focalizer, the one “who sees.” While narrating always entails focalizing (so that the narrator is by definition the primary focalizer), the narrator may report the focalization of someone else (the embedded focalizer); or he may report the character’s act of seeing as a pure event, as Herodotus does when he says that the hare omen “appeared to them.” The distinction between narrative and metanarrative in the history can be described in terms of different focalization when the illusion that there is no narrator is achieved not by means of objective recording of

24

Telling Wonders

the agent in the narrative marches on, the narrator, Herodotus (and this time we are sure it is he), comes forward to communicate his perception. He and his audience come to share an understanding about the discrepan υµβλητον)  cies between the clearness of divine communication (ευσ and men’s failure to respond appropriately and between the initial magnificence of Xerxes’ expedition and its anticipated outcome. The stories of Astyages and Mandane, on the one hand, and that of the omens during Xerxes’ march, on the other, illustrate different discourse possibilities in the Histories. Metanarrative introductions or conclusions may subdivide the narrative at any point; the resulting narrative sections may be theoretically as extensive as the entire work, as small as the smallest segment, or of any extent in between. Introductions (most fre give a preliminary summary that identifies quently with continuative δε) a section of the following narrative as a unit. Conclusions summarize in some way what has been narrated, identifying it as a unit that has ended. Rather than being connected with δε to what precedes, most of these  (or και . . . µεν,  µεν  νυν, µ εν  δη),  to conclusions have anticipatory µεν enhance the mechanical connection of the passage that has just ended with what follows.13 Introductions and conclusions contribute to clarifying the subdivisions of a complex work, but their purely organizational function is secondary to my analysis. Especially interesting, however, is how their form, force, and interpretive potential indicate a more self-consciously didactic undertaking than that performed, for example, by Homeric poetry. Just as the histor is personally involved in investigating his subject in a way that the Muse-inspired bard is not, so he is also in close contact with his public. The way in which he speaks to them and guides their listening, however, is often ambiguous and reflects the complexity of his message. Types of Introductions and Conclusions I begin this discussion of metanarrative by treating introductory and concluding statements because in Herodotus, they are particularly numerous, discrete, and visible. They represent in themselves glosses to the text and thereby attract the presence of other glosses of various types, which can be external events but by means of a narrative focalized through a character. In metanarrative statements, by contrast, we always perceive the presence of the narrator-focalizer. See Marincola 1997b, 9. 13. See Frankel ¨ 1924, 83; Muller ¨ 1980, 77– 78; Immerwahr 1966, 46– 58.

Narrative and Metanarrative

25

found scattered along the narrative (e.g., in parenthetical statements or at the end of a sequence) or within it (in qualifiers). Introductions and conclusions are, in other words, privileged pockets of metanarrative communication. I will briefly survey their basic forms before discussing their general effect on the recipient of the narrative. All introductions and conclusions contain a summary of the narrative they identify, but what I call a summary conclusion is just that—an autonomous plain restatement of the whole or of parts of the preceding account, with no other fixed characteristics.14 For example, the sentence  ο Περσ  δ η υπ  ησι ε δεδουλωντο  2. Λυδο ι µ εν [The Lydians, then, were definitively enslaved by the Persians] (1.94.7) does not mention a new event in the narrative sequence but rather recaps the earlier account of Croesus’ war against Cyrus by rephrasing its result, which has already been recorded (though in different words) along with  δη anticipate a all the other stages of the action. The particles µ εν continuative δε in the introduction to the narrative that follows (1.95.1). The pluperfect tense of the summarizing verb marks the point at which the narrative had arrived before the intervening Lydian ethnography (1.92– 94)—Where were we? Ah, yes: the Lydians had lost their freedom. When an element of summarization on which the emphasis of the sentence lies is either replaced or accompanied by a backward-looking    demonstrative—a form of ουτος, τοιουτος, τοσουτος, or, less often,
οδε—the conclusion is no longer a plain summary. I call it a retrospective sentence. An example is    διε␸θαρησαν.  τουτω   τω 3. κα ι ουτοι µ εν   µορω [and these, then, were killed in this way.] or [and this is how these were killed.] (5.21.1) Here the demonstrative refers back to the unfolding of the action itself in the preceding narrative. “In this way” means “as it has been narrated.”15 14. On the terminology I use here to distinguish different types of introductory and concluding statements, see Munson 1983, 28. 15. I use the terms retrospective and prospective in a more restricted way than does Van Groningen (1958, 42– 43; see also Beck 1971, 7– 10). All else being equal, conclusions in
which the demonstrative is adverbial, especially ουτω(ς), tend to lean back less heavily than those where the demonstrative is subject or object. Elements of summarization that do not

26

Telling Wonders

Like retrospectives, the third and last type of conclusions in Herodotus consists of a nonautonomous, backward-looking sentence whose metanarrative status is formally identifiable. I call it the programmatic conclusion because it makes reference to the narrator’s compositional plan by expressing the idea that the preceding narrative has been narrated and ends at this point. It may or may not include the appearance of the grammatical first person referring to the narrator. Examples follow.   νυν περι  πεπαυµαι.  4. Ροδωπιος µεν [I am through [talking] about Rhodopis] (2.135.6)  των τοσαυτα  αναθηµα

 ε ι ρησθω  5. κα ι περ ι µ εν [And about offerings let this much be said] (1.92.4) Among opening statements, programmatic introductions incorporate a reference to an act of narration that is about to occur. Herodotus’ introduction to his description of Assyrian boats (1.194.1, discussed in the introduction) belongs to this type, 6. But the greatest wonder of all for me . . . I am going to describe [ε#ρχοµαι ␸ρα σων]. The introductory counterpart of retrospective conclusions are prospective sentences, where the primary element of summarization is similarly represented or accompanied by a deictic that points to the narrative or narrative segment that the statement identifies as a unit. In a prospective sentence, the deictic is a forward-looking demonstrative implicitly signify
or τοιοσδε, ing “as it will be narrated” (it is usually a form of οδε but 16  ουτος is also found). An example is 

ι σι ο
ιδε κατεστεασι.  7. νοµοι δ ε αυτο [Their customs are the following.] (1.196.1) bear the main emphasis of the sentence can at any rate be replaced or accompanied by a backward-looking demonstrative pronoun without the conclusion being necessarily “retro  νυν δεδεµενοι  spective” according to my definition. E.g., at 5.72.4, ουτοι µεν ε τελε υτησαν, translated “these [i.e., the Cilonians, just mentioned, then, died in chains” (not “these then were the men who died in chains,” and unlike “these were the only peoples in the cavalry”), is a plain summary conclusion.
16. The prospective value of the adverb ουτω(ς) is sometimes weaker than that of deictic pronouns used as subject or object, in which case the introduction is almost the equivalent of a summary (see, e.g., 1.7.1). A prospective, however, is never as weak as the weakest retrospectives. See n. 15 in the present chapter.

Narrative and Metanarrative

27

Finally, unlike prospective sentences and programmatic introductions, the plain summary introduction does not formally look forward to anything. It consists of a statement grammatically and logically autonomous from the report that follows. If taken out of context, it gives no indication of its introductory function. For example, the sentence 8. There are many other offerings of Croesus in Greece beside those mentioned (1.92.1) happens to represent the heading for a subsequent discussion of specific items. However, the very similar sentence at 1.183.3 (“there are also many private offerings”) does not. Plain summary introductions to narratives may be, in other words, formally identical to summary narratives.17 In fact, another way to analyze summary introductions, especially when the narrative segment they identify is short and connected with γαρ, is to regard the summary introductions as being the narrative and take the following segment as an explanatory gloss that provides further details.18 What identifies a sentence as a summary introduction is the fact that it is more abstract and “processed” than what follows; for example, it may contain broad categorizations or other interpretive elements (see the word prodigy in statement I of passages 1a and 1b quoted earlier). In undecidable cases, the only principle that matters is that when the text contains more than one statement of the type “X happened” in reference to something that happens once in the story, the excess of discourse constitutes a metanarrative phenomenon. The Rhetorical Value of Introductions and Conclusions All introductory and concluding statements in the Histories either can be assigned to one of the three basic types I have described for each or consist of a mixture or series of these. They provide “reading” directions at least by virtue of the fact that they intervene at a certain point to summarize the narrative in a certain way. Statement 2 quoted earlier, 17. With the term summary narrative, I am adapting the concept of summary that Genette (1980, 35– 85, especially 40) develops in reference to novelistic narrative and that Richardson (1990, 35, especially 31– 35) applies to Homeric narrative. My metanarrative summary statements (introductions and conclusions) have much in common with Richardson’s “appositive summaries,” forward- or backward-looking. 18. See, e.g., the narrative at 7.125, analyzed in chapter 4, “Wondering Why.” On explanatory glosses, see “Referential Glosses” later in the present chapter.

28

Telling Wonders

“The Lydians, then, were definitively enslaved by the Persians,” contains no additional glosses but fulfills a function of gloss by bringing out the meaning of the narrative according to the monarchical code. The verb  “enslave” (used metaphorically) is a particularly strong term in δουλοω, this code. It has appeared only once so far in the Histories (1.27.4) but becomes more frequent in subsequent narratives of conquest, especially Persian.19 If we think in terms of “performance,” the sentence seems to require a moment of silence as it underlines a milestone in the story and a major break in the narrative. Following upon the Lydian ethnography and at some distance from the preceding historical account, it concludes the entire Lydian narrative by reminding the audience that the actions of rulers affect communities. Croesus’ defeat by the Persians has caused the “enslavement” of an ethnos whose contributions to culture and initial resourcefulness have just been described (1.94.1– 7). If narrative always entails interpretation, a preliminary summary or a restatement of part of the narrative represents an additional opportunity to interpret, whether by attributing the narrated event to a general class, by privileging a single moment or feature, or by referring to one or more of the cultural codes according to which the narrative can be read.20 Moreover, introductions and conclusions, including those that seem expendable from the point of view of what they actually say, scan and pause the narrative, endow it with a certain rhythm and tone, and perform a “phatic” function vis-a-vis ` the audience.21 Some retrospectives, for example, are equivalent to mere verbalized punctuation marks that leave the listener time to react. This is especially true when both the primary and the secondary elements of summarization are replaced with  19. Similarly, the word δουλος appears in the sense of “subject” beginning with the rule of Cyrus (1.89.1). For cultural codes and the monarchical code in particular, see my introduction, text and n. 28. 20. Cf. Prince’s definition of metanarrative quoted in n. 11 in this chapter. The introduction (1.7.1) and conclusion (1.14.1) framing the Gyges-Candaules episode, e.g., bring out the political code in a story of love, betrayal, and revenge by emphasizing the resulting change of dynasty. They therefore draw attention to the interface between the public and the private spheres in the actions of individuals in power, a major tenet of Herodotean thought. 21. Of six constitutive functions of language that Jacobson (1960, 353– 56) distinguishes in any speech event, the “phatic” focuses on the contact between speaker and addressee through messages “primarily serving to establish, to prolong, or to discontinue the communication, to check whether the channel works (‘Hello, do you hear me?’), to attract the attention of the interlocutor or to confirm his continued attention (‘Are you listening?’ or in Shakespearean diction, ‘Lend me your ears! . . .’).”

Narrative and Metanarrative

29

backward-looking demonstratives. At the end of a riveting narrative of how Alexander of Macedon had an entire Persian delegation killed over dinner and managed to cover up the murder, we find,  µεν  νυν ουτω
 9. ταυτα κ η ε γενετο. [these things, then, happened approximately in this way.] (5.22.2) So, this is it. Period, end of story. What do you know? Prospective sentences, which constitute an implicit promise of something to come, may be almost as inexpressive in their substance as the retrospective sentence I have quoted. When a prospective sentence mentions a set of facts, it serves as the chapter heading for the narrative of those facts (as in “Their [i.e., the Babylonians’] customs are the following” in statement 7). Prospective sentences, however, also point at short range to an individual element about which they anticipate very little information. Thus, forward pointers of the type “he did/devised the following thing” frequently appear to create a tiny moment of suspense before the narrative of clever, unexpected, or outrageous actions. No matter how colorless, a prospective sentence stimulates the recipient to formulate a question (“The following thing happened.” What happened? “He saw the following dream.” What dream? “They sacrifice in the following way.” What way?), and marking a pause before the beginning of the narrative, it signals, “Listen! and you will know.”22 The narrative has then the chance for a new start.23 While the deictic allows the prospective to withhold information momentarily at the same time as it announces a topic, summary introductions must anticipate a complete thought about a topic before the narrative 22. Each logical pause in a narrative implies a question about what will come next, and forward-looking introductions provide the terms in which the narrator wants the recipient to ask the question. In one case (3.6.2), an introductory open-ended question encoding the recipient explicitly formulates a riddle that the recipient may not have thought about and is in turn followed by a programmatic introduction: “Where on earth, one might ask, are the empty jars used? I will explain this too.” All questions in the text that are not in speeches belong to the metanarrative and encode the addressee, but different types fulfill different roles: celebratory (7.21.1), interpretive (1.75.6), or introductory, as here (cf. Iliad 1.8 and 5.703; see Richardson 1990, 179; de Jong 1987, 49). See Lang 1984, 38– 41; Lateiner 1989, 72– 73. For a more detailed discussion of Herodotus’ use prospectives, see Munson 1993a. 23. Examples are “Regarding Croesus himself, this is what happened. He had a son” (1.85.1), and “Smerdis was unmasked in the following way. Otanes was the son of Prexaspes . . .” (3.68.1).

30

Telling Wonders

begins, often with γα ρ, to substantiate and expand on the summary. For this reason, summary introductions tend to be a more suitable tool for interpretation and evaluation than are “strong” prospectives (i.e., prospectives that lean heavily forward because the deictic fulfills most of the summarizing function). See, for example, the introduction to the narrative of the death of Croesus’ son. 10. After Solon left, a great nemesis from god overtook Croesus, one can imagine, because he thought he was the happiest of all men. (1.34.1) To call what happened to Croesus a “great nemesis” and to make a statement on the causality of the event on that basis is a striking interpre ε ι κα σαι] tive maneuver, and the self-referential “one can imagine” [ως identifies it as such. In programmatic introductions and conclusions, the summary that announces the topic of the following narrative or restates some aspect of the preceding narrative is by definition joined to a self-referential gloss of narration, by which Herodotus comes into the open as the one who is ultimately verbalizing the story and putting it together. The narrator displays his control over his material simply by announcing what he is going to say, by cutting a story short, by explaining his reasons for narrating or not narrating something, or by expressing which criteria govern his whole work. Not all programmatic statements are showstoppers. Nevertheless, the introductions or conclusions that bear the greatest rhetorical force and are most expressive about the substance and point of the narrative tend to include a programmatic element. Thus, Herodotus’ first sentence is a mixed summary-prospective-programmatic introduction; it identifies the entire work as a single, though diversified, narrative. This statement signals at the outset the tensions and complications of the Histories themselves, torn between unity and dispersion, fact and meaning, diachrony and synchrony, syntaxis and parataxis.    µητε 

οδεξις
 Ηροδοτου Αλικαρνησσεος ι στορι ης απ ηδε, ως τα    ε ξι τηλα γενηται,

   γενοµενα ε ξ ανθρ ωπων τω   χρονω µητε ε#ργα 
Ελλησι, τα δ ε βαρβαροισι µεγα λα τε κα ι θωµαστα , τα µ εν



 # + α ι τ ι ην ε ποαποδεχθ εντα, ακλε α γενηται, τα τε αλλα κα ι δι ην 

ηλοισι.  λεµησαν αλλ

Narrative and Metanarrative

31

[This is the demonstration of the research of Herodotus of Halicarnassus, lest the events of men may become evanescent with time, or great and wonderful actions, some performed by Greeks, some by non-Greeks, may become inglorious, both the other things and also for what cause they came to war with one another.] The words summarizing the narrative generically and at long range are represented by the subjects of the double purpose clause that gives the reason for narrating. “Great and wonderful actions, some performed by  Greeks, some by non-Greeks” [ε#ργα µεγαλα τε κα ι θωµαστα , τα µ εν   


Ελλησι, τα δ ε βαρβαροισι αποδεχθεντα] are the terms of the narrator’s code of celebration, here advertising the narratability of all the “events of 

 men” [τα γενοµενα ε ξ ανθρ ωπων] treated in the Histories, be they historical or ethnographic. A more specific summary is provided by the final # colon of ambiguous grammatical status, where anticipatory αλλος in τα τε # αλλα κα ι [both the other things and also] serves as a bridge between the + α ι τι ην ε πολεµησαν  broader subject matter and the narrower topic δι ην

ηλοισι  αλλ [for what cause they came to war with one another].24 The word α ι τι η means “cause,” “grievance” (of an offended party), or “responsibility” (of an offender). As a historical, juridical, and ethical term, it anticipates the combination of these three codes in the immediately following narratives about the remote origin of the East-West conflict and, at long range, in the narratives of many other wars both between Greeks and nonGreeks and between members of each group.25 The first sentence also notably names the real author, Herodotus. This person will henceforth become the “I” of the text, the histor. He will appear again in his historical dimension, as he does here, in a particular type of metanarrative statement that I call glosses of historie, that is, 24. The best analysis of the first sentence is by Erbse (1956, especially 217– 19 for this point). Cf. also Krischer 1965; Drexler 1972, 6– 9; Moles 1993, 92– 94. Initial sentences of # other fifth-century prose works are discussed by Fowler (1996, 69– 70). Anticipatory αλλος (as opposed to “additional” as in statement 8 earlier in the present chapter) has the function of narrowing the focus of the narrative that follows by postponing (or excluding) certain aspects of the topic being treated. Wood (1972, 14) rightly regards it as one of the signs of “a perspective . . . which views discrete events as parts of a whole, which sees always the meaning of events.” 25. See α ι τι ους at 1.1, where it means “those at fault,” “those responsible,” and  . . . υπα   ρξαντα αδ

ι κων therefore also the “causes” and is eventually picked up by τον ε#ργων [the one who initiated unjust actions] at 1.5.3. See Pagel 1927, 8; Immerwahr 1956, 245. As Payen argues (1997, 88– 91), the narrative of the Histories imposes a broad interpretation of the phrase “they came to war with one another.”

32

Telling Wonders

references to Herodotus’ fact-finding process.26 The gloss of historie in  

οδεξις.

οδεξις this sentence is in the expression ι στορι ης απ The word απ constitutes the programmatic element (or gloss of narration) of the sentence. It means “performance” in the sense that the product of historie becomes a display of narrative; but it also points to the disclosure of the preliminary discovery process, in a sort of contrapuntal narrative of the actions and reasonings of the histor as researcher—where he went, where he learned things, how he put together what he saw and heard, and so

on.27 But another type of gloss is embedded here: since the root αποδεκin this same sentence also advertises the narrated (“great and wonderful 



οδεξις deeds performed” [αποδεχθ εντα]), the term απ is attracted into the celebratory code as well. The narrator, Herodotus, in other words, presents his own speech act as a performance in some way comparable to the great and wonderful performances of the characters of his narrative.28 Self-Referential Glosses As my preceding analysis of the first sentence of Herodotus’ Histories shows, an introductory or concluding statement is a gloss for the way in which it summarizes the narrative, but at the same time, it may also contain a number of other glosses. There are two major categories of glosses, selfreferential and referential. A self-referential gloss belongs to the level of metanarrative that is most distant from the story level, because it defines and qualifies—“talks about”—another piece of the text (i.e., a narrative, part of it, or another gloss) as a verbal product, rather than focalizing through the narrator a referent in the world of the narrated. Glosses of narration. We have already encountered glosses of narration: they identify a narrative, a portion of a narrative, or another metanarrative statement as something that the narrator, Herodotus, has said or will say, or they postpone or recall narratives (“as I will narrate later,” “as has been said before”). In the negative form, glosses of narration describe the text by default; they state that it lacks certain features by decision of the narrator. 26. See Dewald 1999, 232, 236. 27. See Dewald 1987, 167; Marincola 1987, 127.  28. See Erbse 1956, 209– 11; Nagy 1990, 218– 21 with nn. 24 and 35. Αποδεξις is

also a part of a semic code and connotes wisdom because another use of αποδε ι κνυµαι has to do with the display of advice, opinion, or sophie (cf., e.g., 2.146.1, 7.139.1, 8.8.3: 

 γνωµην αποδ εξασθαι, applied to the narrator).

Narrative and Metanarrative

33

Herodotus identifies the unfinished product of his narration as his  λογος. He also applies this word to some of his individual narratives or parts of the whole.29 Herodotus denotes his act of narration by both verbs of speaking and verbs of writing; he thereby shows his adherence to the two modes and his double intention to communicate in the present and to create a lasting record.30 Several of Herodotus’ terms for narrating participate in the metaphor of a journey to the places that will become part of the narrated.31 The narrator as such is a traveler who may or may not decide to embark on the roads of other people’s logoi.32 The itinerary is his work in progress, his logos, which “seeks” or “wants” (verb ε πιδι ζοµαι or δι ζοµαι) more or less predictable destinations or topics.33 This metaphor of the journey creates the illusion of an overlap between the code of narration and the code of historie (where verbs of motion are in most cases used literally). Thus it reinforces our sense of the identity between the narrator and the researcher.34 29. See, e.g., 1.5.3, 1.140.3, 1.184. See also Payen 1997, 63– 66. 30. Nagy (1990, 219) writes, “Saying something is in the case of Herodotus the equivalent of writing something because it is ultimately being written down in the Histories.” Cf. Hartog 1988, 276– 89. According to Immerwahr (1966, 15), γρα␸ω emphasizes exactitude (1.95.1, 2.70.1, 2.123.1) or the notoriety of a person or event (2.123.3, 7.214.3). Cf. Hartog 1988, 285. Herodotus’ vocabulary of narration is vast. Verbs with the root µνη (ε πιµνησθεω, 2.3.2; etc.) may suggest the idea of memorializing through words (cf. the memorial of words at 7.226.2; see Immerwahr 1960, 267). Applicable to both oral or written speech are verbs of mentioning, indicating, explaining, or displaying, including the particularly fluid σηµαι νω (on which see Hartog 1988, 366; Nagy 1990, 165). Other  performative verbs of narrating include “go on at length” (µηκυνω, 2.35.1, 3.60.1), “use”  in the sense of mentioning information he knows (χρωµαι, 8.85.2), “refrain from saying”  

ερχοµαι,  (ε πεχω, 7.139.1), “stop” (παυοµαι, 2.135.6), “revert” (αν 1.140.3, 7.239.1),

ι ηµι, 3.95.2), “forget about” (ε πιλανθανοµαι, 4.43.7). “omit” (απ  31. The act of narrating is often expressed with verbs of going: e.g., ε#ρχοµαι ε ρεων # λεξων  (1.5.3, 2.99.1), ηια (4.82). At 1.5.3, the narrator will proceed forward in the logos     (προβησοµαι ε ς το προσω του λογου), going through (ε πεξι ων) cities great and small. Bidding goodbye to a subject is equivalent to leaving a place (see 2.117, 4.96.2). 32. At 2.20.1 (where metaphorical and literal terms of the code of narration mix in a striking way), the narrator states that some Greeks wanting to be distinguished (or perhaps  “visible”: ε πι σηµοι) for their cleverness, “said three roads [οδους]” (cf. 2.22.1), two of which he does not deem worth mentioning, except that he wants only to point to them   about Cyrus (σηµηναι). At 1.95.1, he knows three other “roads of logoi” [λογων οδους] besides the logos according to which he will write. For the journey metaphor applied in the narrative to the speeches of characters, see 1.117.2, 2.115.3. 33. See 1.95.1, 4.30.1. The verb δι ζοµαι is often found in the narrative in conjunction with verbs of motion, as when Heracles goes all over Scythia looking for his mares (see 4.9.1). 34. On Herodotus as alter ego of the Homeric Odysseus, see Nagy 1990; Marincola 1997, 1– 3.

34

Telling Wonders

Narrating is only one of the activities the text attributes to the authorial first person, but one in which Herodotus displays unparalleled selfassurance.35 Occasional apologies compensate for his expressions of control over what he narrates,36 and these phrases sometimes take the form of an appeal to the narrator’s obligation or lack of obligation to mention something or to include certain types of material. The verb δει appears frequently in this context and bears little emphasis (“And yes, I must still explain . . . where was the dirt from the ditch utilized,” 1.179.1). It suggests “a sense of strain inherent in the problem of composition,”37 the need to be detailed and exact, perhaps the fear to go on for too long and appear trivial. Other times, we find stronger phrases expressing a graver commitment to the integrity of his logos. The logos that seeks/wants (his logos, always with a dative of possession) is his narrative agenda. This is never sharply defined for the recipient, but it imposes on the narrator a duty that overrides personal inclinations, other moral issues, or external pressures.38 Glosses of source. Complementary to glosses of narration are glosses of source, which identify a narrative or statement as originating from other narrators outside the narrative (“so and so say” or “it is said”).39 The logoi of others have the important function of giving a voice to the other, who challenges the subjectivity of the audience.40 At the same time, 35. See Dewald 1987, 164. The problems connected with the first-person plural are discussed by Chamberlain, forthcoming. 36. At 2.45.3, the narrator exceptionally apologizes to the gods and heroes, as if they were a special part of his audience. 37. Dewald 1987, 165.

38. On the narrator’s obligation with αναγκ-, see 7.139.1, 2.3.2, 2.65.2, 7.99.1. Cf.  οκειται),  2.123.1 (ε µο ι . . . υπ 7.152.3 (ο␸ει λω, ου . . . ο␸ει λω). The idea of obligation is discussed further in chap. 4. 39. See Dewald 1987, 153; Jacoby 1913, 398– 99; Darbo-Peschanski 1987, 91– 97. Sometimes sources do more than “say”: they demonstrate (see, e.g., 5.45.1, 5.45.2), know exactly, calculate (see 2.145.3), agree or disagree with one another (see, e.g., 1.23.1), swear (see 4.105.2), accuse (see 6.14.1), have nothing to say (see 3.111.1), and so on. Though Herodotus refers to the provenance of his evidence throughout his work, he does so irregularly, and many evidently received narratives are not marked by “they say” or any such gloss or slide from the second to the first narrative level of the narrator’s own voice. See Darbo-Peschanski 1987, 113– 18, 124– 25; Hartog 1988, 291– 94. Dewald (1999) discusses the problems of the focalization, and therefore the status, of unattributed narratives. 40. Glosses of sources identifying foreign logoi resemble cognitive statements of the type “the Persians say.” These, however, record a people’s cultural beliefs in an ethnographic context and are therefore just as much a part of the narrative as are statements of the type “the Persians do.” An example is “The Taurians say that the divinity to whom they sacrifice is Iphigeneia, the daughter of Agamemnon” (4.103.2).

Narrative and Metanarrative

35

the logoi reported in the Histories also expose the self-interest or subjectivity of the speakers themselves.41 More often than not, they provide unreliable evidence, contradict one another, and lead to contention. If Herodotus’ first sentence puts the narrator in charge, his second sentence (“The Persian logioi say that the Phoenicians were the ones responsible for the conflict,” 1.1.1) introduces a verbal quarrel that remains unresolved.42 This is only the first of many quarrels in the metanarrative of the text that echo, reflect, and metaphorically represent the disputes and struggles among the characters in the narrative and in the real world. Glosses of historie. When a source “says” something in the past tense and directly to the researcher, the gloss of source is colliminal with a gloss of historie, which identifies a part of the text as containing the results of Herodotus’ inquiry.43 No longer a passive recipient of information, the histor travels, “wants to know,” participates in interviews, collects hearsay, verifies by autopsy. The more extended passages that place him at the scene are small narratives in their own right, though at a different level, and represent the closest thing in Herodotus to what modern ethnographers call “personal narrative.”44 To put it in literary terms, the outside narrator enters the narrative and almost becomes a character.45 Evaluations of accuracy, glosses of evidence, knowledge, and ignorance. Other than underlining the provenance of a statement or a narrative in the text, self-referential signs give indications about their reliability. In a text mostly composed of the received logoi of others, these are important directions. Evaluations of accuracy46 corroborate or decline to 41. See Dewald 1987, 168. 42. See Dewald 1999, especially 236. For an inventory of Herodotean alternative versions, see Lateiner 1989, 104– 8. The interface between metanarrative and narrative quarrels is discussed in chap. 3, “Disputes, Arbitration, and the Subjectivity of Opinions.” 43. Jacoby 1913, 247– 76, 395– 400; Macan 1895, 1:lxxxi– lxxxii; How and Wells 1928, 1:16– 20; Marincola 1987; Dewald 1987, 155– 59; Hartog 1988, 261– 73; DarboPeschanski 1987, 84– 87. 44. Pratt 1986; Geertz 1988. 45. See Marincola 1987, especially 127– 28. Marincola gives a complete survey of Herodotus’ statements of inquiry and shows that the extended type only occurs in book 2, where the statements represent implicit polemics against previous accounts. In these passages, we find verbs of setting forth, traveling, going, arriving, sailing (see ε#πλευσα at 2.44.1), being in  someone’s company (συνεγενοµην, 3.55.2), and so on. Occasionally the histor’s sources become part of the narrative as well, as do the Egyptian priests in the famous piromis scene (2.143) and the disingenuous scribe of the temple of Athena on Sais (2.28.2). 46. Accuracy is not a very satisfactory term here, but I am reluctant to use truth, a word that Herodotus himself avoids in most contexts. See chap. 3, “Specific Glosses of Interpretation.”

36

Telling Wonders

corroborate, refute or reject tout court.47 Herodotus guarantees his own statements with glosses of evidence (“it is evident that such and such is the case”)48 or of knowledge.49 When he says he does not know (or he “cannot tell exactly”), he is either unable to corroborate someone else’s logos or acknowledging the incompleteness of his own.50 Glosses of opinion. Among the most pervasive self-referential metanarrative signs are those that identify a referential gloss or a statement of fact (e.g., the number of troops at 7.184.1) as representing the result of Herodotus’ mental activity—estimate, reasoning, conjecture, judgment, and so on. To these I apply the blanket term opinion.51 Glosses of opinion both weaken and enhance the authority of the text. When they accompany a referential gloss that proclaims how great the significance of a fact is (celebration), discloses what its significance is (interpretation), or evaluates its worth, they underline Herodotus’ own subjectivity, separate from that of other voices in and outside the text.52 They constitute the most forceful markers of Herodotus’ own ideological and philosophical position. For example, 47. Rejections qualify received information as not trustworthy. A refutation is an explanatory gloss attached to a rejection (called an ε#λεγχος at 2.23; cf. 2.22.4). See, e.g., 3.45.3. 48. When Herodotus is confronted with controversial issues, the gloss of evidence often introduces a referential gloss that explains the evidentiary basis for something (see, e.g., 5.22) See Darbo-Peschanski 1987, 131– 47; Thomas 1997 and 2000, 168– 200. 49. With such generalized expressions as “I learn” [πυνθανοµαι] (e.g., 7.114.2), the category of glosses of historie overlaps with that of glosses of knowledge, which emphasize the results rather than the process of the research. An example occurs at 1.20: “I know, having learned it from the Delphians.” See Lang 1984, 11– 17. Glosses of knowledge   include the limitative phrase “the first [or greatest, etc.] we know about” (πρωτος των  ις #ιδµεν, etc.). See Shimron 1973; Darbo-Peschanski 1987, 113; Hartog 1988, 289– 90. ηµε 50. See, e.g., 4.25.1 (no one knows), 4.180.4, 6.14.1. See the lists “Ignorance Universal” and “Certainty Impossible” in Lateiner 1989, 69– 71. See also Flory 1987, 49. 51. In the programmatic statement with gloss of historie at 2.99.1, Herodotus mentions  the role of his own γνωµη, “judgment,” in processing the data of eyewitness and oral report. Besides verbs of seeming, thinking, and conjecturing, Herodotus’ vocabulary of opinion also  (3.111.1, 3.38.2, 4.195.4, 5.97.2, 7.167.1, includes expressions with ο-ικα and ο ι κ ως 7.239.2), the dative of reference µοι (see passage 17 later in the present chapter), and certain first-person verbs in self-referential or referential glosses (e.g., πει θοµαι, “I am persuaded”;  “I laugh”; α ι νεω,  “I praise”: see passage 15 in the present chapter). See Beck 1971, γελω, 70– 72; Dewald 1987, 162; Lateiner 1989, 98; Hohti 1977; Darbo-Peschanski 1987, 164– 89, especially 184. 52. Darbo-Peschanski (1987, 186) writes ‘Lorsque l’enquˆeteur consent a` e´ valuer ses propres discours comme ceux de ses informateurs . . . il donne . . . a` son jugement le caract`ere relatif d’un avis susc´eptible d’ˆetre discut´e.” See also Dewald forthcoming a.

Narrative and Metanarrative

37

  η µετερην]  11. In my judgment [κατα γνωµην τ ην the following is the wisest custom, which I learn that the Illyrian Eneti also have. (1.196.1) 12. . . . since I believe [νοµι ζων] that all men know equally about the gods . . . (2.3.2) To the extent, however, that opinion identifies information that may not be accurate, it is a compromise that replaces the vacuum of “being unable to tell” but falls short of “knowing.”53 Various glosses encoding the addressee. What the narrator Herodotus knows, corroborates, and believes (in the sense of nomizein) provides reference points for the understanding of the world he wants to communicate. Glosses that express his ignorance or uncertainty are the explicit marks of an interrogative text. They put the recipient of the narrative in charge. We perceive the presence of the recipient in the text especially through the mediation of self-referential metanarrative. He is the anony
mous οστις who is asked by a gloss of noncorroboration at 2.123.1 to “use” the reported logoi if he finds them credible, and he is the τις who is invited to choose between different versions at 5.45.2. This “whoever” or “someone” is also enlisted in glosses of historie as a potential histor, traveler, observer, or eyewitness (“it is evident even for one who has not heard about this but just sees, at least if he has some intelligence . . . ,” 2.5.1).54 Direct addresses in the second-person singular coopt the listener to the inquiry.55 Yet the narrator also expresses some doubt that his audience will be up to the task with which he entrusts them. When glosses of narration explicitly refer to their likely reaction to what is or could be narrated, they 53. I am paraphrasing Romm (1989, 100 n. 12). The nonauthoritative aspect of Herodotus’ opinion is emphasized by Darbo-Peschanski (1987, 146– 47, 184– 89). 54. The attributive participle also encodes the recipient in this capacity, e.g., at 2.135.3:   ] can see . . .”; cf. 1.105.4, 2.31. In some cases, mostly “anyone who wants [τω   βουλοµενω in referential metanarrative (2.154.4, 3.116.3, etc), the first-person plural encodes the audience because it means “we the Greeks.” On the role of the narratee, see Prince 1973; 1982, 16– 24; 1987, s.v. “narratee.”  55. An example occurs at 1.139: “if you look into this, you will find . . .” [ε ς τουτο  ησεις   διζηµενος ευρ . . .); see also 3.6.2 (the recipient as interviewer); 1.199.4, 4.28.1, 3.12.1 (the recipient as prospective or hypothetical experimenter); 2.5.2, 2.29.5, 2.30.1, 2.97.2 (the recipient as prospective traveler). See Dewald 1987, 155; 1990, 220. De Jong (1998) notices the similarity with the use of the second person in the Hippocratics (e.g., Airs, Waters, Places 8.10).

38

Telling Wonders

especially mention disbelief.56 One famous passage (7.139.1), which I examine later (chapter 3.2), attributes to them hostility. Referential Glosses A referential gloss provides directions on how to receive the narrative by commenting not on the narrative itself but on the narrated. Such glosses often represent the propositional content of glosses of the self-referential group—in other words, they correspond to item Y in a statement of the type “X is evident because Y” or “it is my opinion that Y.” But referential metanarrative can stand on its own. Referential glosses constitute the level of metanarrative that is closest to the narrative. For this reason, they fulfill their function indirectly and often in a subtler way than statements that identify a piece of text as coming from a certain source, as being Herodotus’ opinion, or as representing—or not—an accurate report. Consider, for example, what I call glosses of testimony. These consist of references to poetic and other written testimony of narrated events or to tangible vestiges of the past that are generally well known, verifiable by Herodotus’ contemporary audience, or allegedly verified by the narrator/researcher. Glosses of testimony sometimes appear to be the referential content of an implicit gloss of corroboration or evidence. An example is the item y in “X is evident/ proven/accurate because Y,” whatever X in the narrative may be in the particular case. Thus, the Spartans were defeated by the Arcadians (X), and their chains were visible “still in my time” in the temple of Tegea (Y): here Y memorializes event x and confirms its gravity (1.66.4).57 In many cases, however, what notations of this sort contribute to Herodotus’ account, what they confirm or go to prove, is not entirely transparent. At 5.77.3, for example, we encounter another mention of chains: the chains of Chalcidians and Boeotians, crushingly defeated by the Athenians in 56. Glosses of narration automatically encode a recipient of the narrative—“I say/ narrate/write (for the benefit of someone)”—but they are also a type of gloss where the recipient is likely to be explicitly mentioned. At 1.193.4, the disbelief of the audience motivates the narrator’s negative program. Disbelief is attributed to “some Greeks” (i.e., some of the audience) in a gloss of narration at 6.43.3 and in an implicit gloss of corroboration at 3.80.1. See Hartog 1988, 289– 90; Packman 1991, especially 406. Other glosses of narration refer to the audience’s cultural knowledge as the basis for narrating or explaining something: see 3.37.2, 3.103, 4.81.4, 4.99.5. 57. On these glosses providing evidence for the “greatness” of an event, see Immerwahr 1966, 269.

Narrative and Metanarrative

39

their first battle after the liberation of Athens from the tyrannical regime, still hang at the Athenian Acropolis on the wall “half-burned by the Medes.” This notice juxtaposes the distant past, the more recent past, and the present of narration in an allusive way that requires decoding. It goes beyond a testimonial function in the most obvious sense.58 Explanatory glosses. Questions about the function of metanarrative intrusions sometimes emerge within the broad referential category of explanatory glosses (to which glosses of testimony also sometimes belong). These provide new factual information apparently designed to clarify some element of the context in which they occur, but their explanatory value is not always clear. At the end of the narrative of the murder of the Persian ambassadors at the hands of Alexander of Macedonia during a banquet, for example, we find a gloss (itself emphasized by selfreferential glosses of knowledge and source and generating a gloss of evidence with its explanation) stating that the kings of Macedonia are Greek (5.22). Is this information designed for the sake of apologia or irony? Or has the narrator simply taken the opportunity to insert information that will be useful later on?59 Explanatory glosses in general occupy an intermediate position between the metanarrative and the narrative, because along the main narrative line that proceeds in chronological order from Croesus to Xerxes, insertions marking a change of time always more or less begin as explanatory glosses. Thus, the long flashback on Cyrus’ antecedents is introduced by the summarizing statement “who was this Cyrus who conquered Croesus” (1.95.1). It interrupts the chronological narrative, in other words, like the delayed and much expanded gloss of identification that typically accompanies the entrance of a character in the history to give a few facts about his family, his position, and his accomplishments.60 Similarly, even the most structurally autonomous ethnographic description can be regarded as a gloss explaining the background of a people that has played a role in the narrative or is about to do so. In comparison with inserted semiautonomous narratives, the metanarrative status of an 58. For other glosses of testimony, see, e.g., 1.12.2, 1.24.8, 2.123.3, 2.131.3, 4.11.4, 4.12.1, 4.166.2, 5.58.3, 6.14.3, 7.167.2, 7.178.2. At 3.38.4, the Pindaric quotation is used somewhat differently, to corroborate a gloss of interpretation. 59. See Badian 1994, especially 121, on the ambivalence of Herodotus’ whole Macedonian narrative. 60. See, e.g., 1.6, 1.23, 5.32, 6.35.1, 6.131, 8.79.1 (cited as passage 16 in the present chapter). Cf. the discussion of Homeric character introductions in Richardson 1990, 36– 51.

40

Telling Wonders

explanatory gloss is based on its brevity. Consequently, it is, or we expect it to be, immediately functional as a short-range explanation and not autonomous from the narrative to which it is attached. An isolated ethnographic gloss within the Gyges-Candaules episode, for example, is clearly designed to help a Greek audience to evaluate Candaules’ behavior in displaying his wife naked. 13. Among the Lydians, as also among almost all other nonGreeks, to be seen naked even for a man brings great shame. (1.10.3) While historical glosses bring out the historical code in an ethnographic description (e.g., by explaining the origin of certain customs or monuments), ethnographic glosses underline the “code of customs” in the history. Both testify to the mutual interdependence of the two genres in Herodotus’ work. Sometimes, we need to know about culture to understand history, or we need to know about historical occurrences to understand culture. But again, specific cases raise different issues: not only whether and how an ethnographic gloss actually contributes to the context where it appears, but also whether it is complementary or contradictory in relation to the ethnographic information about the same people given elsewhere, especially in autonomous ethnographic narratives; or how these isolated bits of ethnographic knowledge work as a group when they can be classified cross-culturally, according to the areas of culture they discuss.61 Glosses of comparison. An explanatory intrusion in the text also occurs every time the narrator brings into the narrative an extraneous referent for the purpose of affirming its resemblance to or difference from something in the story. The resulting gloss of comparison often explains an unfamiliar phenomenon through one that is, from the point of view of the audience, more familiar. But this is clearly not the case, for example, with the numerous comparisons proclaiming the similarity or equivalence among customs and beliefs of different peoples (see the earlier examples 11 and 13). The function that these and many other comparative glosses perform individually and the cumulative effect of 61. See, e.g., the collections of ethnographic glosses concerning oath taking or purification rituals around the world (1.74.1, 3.8.1) and the metalinguistic glosses (translations of terms) scattered along the narrative. On the latter group, see Hartog 1988, 237– 48; Harrison 1998; Chamberlain 1999.

Narrative and Metanarrative

41

explicit comparison in the text are, as I shall show in the next chapter, more profoundly interpretive. Glosses of interpretation. Interpretation is explanation at a higher conceptual level. Interpretive glosses occur when the voice of the narrator comments on some of the more covert or questionable aspects of the narrated. Herodotus’ decoding of dreams, omens, symbolic objects, oracles, logoi, and other utterances belongs at the more explicit end of the explanatory/interpretive spectrum. At the other end, interpretation is communicated by the summarizing elements of introductory and concluding statements and, closer still to the edge of “pure narrative,” even by the words and codes with which the text verbalizes the story in the narrative itself. Between these two poles, we find a number of statements that discuss, more or less conspicuously or problematically, why an event is important; its value, meaning, or origin; and its less obvious motive or result. In most cases, interpretive glosses do not add new facts, as pure and simple explanations tend to do, but rather process those given in the narrative. The operation is often made more visible by a gloss of opinion or by some other self-referential gloss—for example, one of knowledge or evidence. An example is 14. And it is clear to me [δηλοι τε µοι] that the whole situation on the barbarian side depended on the Persians, since also these fled even before they engaged with the enemy, just because they saw the Persians flee. (9.68) Other metanarrative signs that alert us to the presence of an interpretive gloss are a generalized form of discourse (e.g., “all men know equally about the gods” in example 12), certain types of negations and questions, hypothetical constructions, particles signifying “perhaps” or “somehow” (especially κως), or a listing of alternative choices about what may have happened or why.62 62. For multiple choices, see, e.g., 1.86.2 (motives of Cyrus), 7.54.3 (motives of Xerxes), 7.239.2 (motives of Demaratus), 8.87.3 (motives of Artemisia). For an interpretive negation, see, e.g., 6.61.1. Negative statements are always a part of the metanarrative, because what we call the story does not include nonevents. See Prince 1982, 18– 19; de Jong 1987, 61– 68, especially 67; Hornblower 1994a, 152– 53. The same is true for hypothetical constructions (see, e.g., 7.139.2– 4). Important interpretive glosses with κως and κου appear at, e.g., 3.108.2, 7.191.2, and 6.98.1. Cf. Lateiner 1989, 31– 32. Interpretive glosses in the form of a question (often in combination with a hypothetical construction) appear at

42

Telling Wonders

In the absence of any marker of interpretation whatsoever, statements that attribute undisclosed motives to characters can present special problems.63 In other cases, the perception that we are in the presence of a gloss depends first and foremost on the level at which the text processes the raw external data. No self-referential sign of interpretation marks the statement that the Athenian war against Aegina, by forcing the Athenians to build ships, proved to be the salvation of Greece at the time of the Persian invasion (7.144.2). Yet the causal connection it establishes a posteriori between two otherwise unrelated occurrences clearly reveals the mental process and deliberate intervention of the narrator. By creating a thoughtprovoking paradox, he goads the reader to search for further meaning.64 Evaluations of worth. Embedded in many narratives is a judgment that some of the actions narrated are “Bad” or “Good” on either moral or intellectual/strategic grounds or by both standards at once.65 In other cases, the narrator himself, explicitly and in his own voice, makes evaluations of worth either by using evaluative words in the course of the narrative (e.g., 9.78.1: “he uttered a most impious speech”) or by using glosses of praise or blame.66 In the following retrospective/prospective that marks the transition between two items of the Persian ethnography, the evaluative verb incorporates a self-referential gloss. “I praise” is more or less equivalent to “I record/believe that [gloss of narration and opinion] it is good [evaluation].”   νοµον,    µεν  νυν τονδε  δ ε κα ι τονδε 15. α ι ν εω τον α ι ν εω [I praise that custom [reported earlier], and I also praise the following one.] (1.137.1) 2.11.4 (twice), 2.125.7, 4.46.3, 1.75.6, 2.15.2, 2.22.2, 2.45.2, 2.45.3, 2.57.2 (the last six occur in refutations). Generalizations are discussed in chap. 3. 63. This is also due to the fact that the narrator of the Histories swings back and forth between the positions of omniscient narrator—that is, with “the privilege . . . of obtaining an inside view of another character” (Booth 1983, 160; cf. Chatman 1978, 212, 215– 16)— and researcher. In the latter stance, he distinguishes seen from unseen and marks attributions of motives by a self-referential gloss. Dewald (1987, 161 n. 161) counts twenty-two  and δοκεειν  cases in which δοκεω ε µοι have this function. 64. Interpretive glosses also include glosses of anticipation of doom, which underline the decisive role of a functional event in the plot in triggering an overdetermined negative outcome. An example occurs at 1.8.2: “After not much time, since Candaules was bound to end up badly, he said to Gyges the following.” See also 2.161.3, 5.92δ1, 6.135.3, 9.109.2, 4.79.1. For discussion of these passages, see Hohti 1975; Gould 1989, 72– 78; Munson forthcoming. 65. I borrow these deliberately vague expressions from Asad 1986. 66. See Prince 1982, 11.

Narrative and Metanarrative

43

Praise (reinforced by glosses of opinion and historie and by a pun with the proper name) is the most emphatic component in the identification of

Aristides (8.79.1). The gloss is designed to underline the element of αρετ η (moral excellence) in the narrative of Salamis, otherwise dominated by an ethically more ambiguous σο␸ ι η (cleverness) of Themistoclean stamp.67 16. When the generals were gathered together, there crossed over from Aegina Aristides, the son of Lysimachus, an Athenian who had been ostracized by the people and about whom I have come to believe, when I was inquiring about his character, that  ε γω he was the best man in Athens and the most just [τον    τροπον, 

υ τον # # νενοµικα, πυνθανοµενος αυτο αριστον ανδρα    γενεσθαι ε ν Αθηνησι κα ι δικαιοτατον]. Advertisements of narratability. In Herodotus, evaluation is intimately joined with explanation and interpretation because bad or good behavior and foolish or wise actions determine the course of history. At the same time, Herodotus’ positive evaluation of a fact in the world of the narrated sometimes also coincides with a different metanarrative function: the celebration of his subject as such. Thus, a conclusion/introduction system within the narrative about Sperthias and Boulis at the Persian court represents both an evaluation of worth and what I call an advertisement of narratability (or celebratory gloss). It is phrased in terms designed to recall the text’s initial advertisement in the promise to preserve the renown of “great and wonderful deeds.” 
τε η τολµα   ανδρ

 θωµατος 

ι η κα ι ταδε 17. αυτη τουτων των ων αξ  τουτοισι  προς τα ε#πεα. [This boldness [i.e., that reported earlier] from the part of these men is worthy of wonder, and in addition also the words they said, as follows.] (7.135.1) Here and in other cases where the narrative is about brave deeds, “goodness” of conduct and what is deserving of mention are one and the same. This attitude conforms to the Homeric tradition, later pursued by 67. Cf., e.g., 8.124.2. In Herodotus, when the moral and the strategic standards of evaluation are separate—which is by no means always the case—the split between the two centers especially around the fluid term σο␸ ι η, which can be used to mean “intelligence” in a narrower sense and does not necessarily convey moral approval.

44

Telling Wonders

the praise poetry of Pindar.68 In Herodotus, however, what I have called evaluation of worth is more specifically moralistic than is praise of greatness of the Homeric type; and conversely, his field of celebration (“great and wonderful deeds”) is both different and broader than the range of what Homer, Pindar, or (in his own way) Thucydides claim is worth preserving.69 Perhaps because Herodotus’ concept of narratability is so unpredictable, the advertisement we find in the first sentence needs continuous tending along the logos. Terms of the celebratory code in fact recur throughout the Histories, often joined to programmatic statements and glosses of opinion. These express the narrator’s authority in determining what he will tell for no other reason than that it is, or he considers it to be, worthy of being told.70 Phenomena of very different types and magnitude, historical or ethnographic, are emphasized in this manner. Explanation in a broad sense, including interpretation and evaluation, purports to indicate why something in the world of the narrated occurs and what constitutes its importance, meaning, or worth. Celebration, in contrast, is simply designed to signal that a feature in the world of the narrated possesses some sort of importance, meaning, or worth. Earlier in this chapter, I showed that in many cases, self-referential glosses put the reader in charge of filling in the blanks left in the text and of interpreting the reality the text represents. So far as referential glosses have an unstable or multiple role, they achieve a similar effect, albeit even more implicitly. Referential metanarrative represents the main focus of my analysis. In the next two chapters, I examine how Herodotus explains through explanation. In chapter 4, I explore, albeit through the study of a single term, how he directs us to explain (if that is what he does), or what else he does, through celebration. 68. On Pindar’s praise poetry in relation to the Homeric tradition, see Nagy 1990,  150– 207. For Homeric praise of αρετ η in Herodotus, see my chap. 3, “Interpretation in the History.” 69. Thucydides magnifies his subject as the greatest war and the most worthy of report (1.1– 19, especially 1– 2; cf. 7.87). Cf. Herodotus at 7.20.2. The tradition of ancient historians’ magnification of their subject is broadly surveyed by Marincola (1997a, 34– 43). On praise and Homeric glory in Thucydides, see Immerwahr 1966, 177– 279.  70. These celebratory terms include words of the θωµα family; the phrases “great 

οδεξις, deed” or “great work,” “display of deeds” (ε#ργων απ 2.101.1), “deeds greater than   human” (2.148.6); the noun λαµπροτης (“brilliance” (2.101.1)); the adjective µεγας,  µεγιστος, as well as other superlatives; words and expressions indicating fame (verb

 ευδοκιµ εω), originality (“the first we know about” to do something), primacy (“the first in his time”), or uniqueness; various expressions equivalent to “worthy of being told”  

ολογος)  (αξι or “greater than words” (λογου µεζω, 2.148.1).

Telling Wonders

Telling Wonders Ethnographic and Political Discourse in the Work of Herodotus

Rosaria Vignolo Munson

Ann Arbor

Copyright 䉷 by the University of Michigan 2001 All rights reserved Published in the United States of America by The University of Michigan Press Manufactured in the United States of America 嘷 ⬁ Printed on acid-free paper 2004 2003 2002 2001

4 3 2 1

No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, or otherwise, without the written permission of the publisher. A CIP catalog record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data applied for ISBN 0-472-11203-1 (cloth : alk. paper)

To Paul and Louise  µοι   Θωµατα µεγιστα

Preface

My first thanks go to Donald Lateiner, once my teacher and now my friend, most generous through the years with advice, support, and information of all kinds. He has just recently read this book in its final draft, but he has certainly had to hear a great deal about it for a long, long time: I hope he has not been disappointed. Last year William Turpin, in the busiest of all seasons for a department chairman, took the time to go through an earlier, bulkier version of my manuscript to help me decide how I could cut it down to about half its size. I am immensely grateful to him for his heroic help, as I am to Ellen Bauerle, then of the University of Michigan Press, for her wisdom in prescribing the shrinkage. Carolyn Dewald, my much admired fellow Herodotean and always a source of inspiration, has read all existing versions of this book. She made herself available for countless written and oral critiques of my work and for lengthy telephone discussions; she has been most generous in sharing her work in progress with me. I also thank Deborah Boedeker, Gregory Nagy, Helen North, Martin Ostwald, David Potter, Gil Rose, Grace Ledbetter, David Konstan, Elaine Fantham, Eric Munson, and John Carleo for their support, suggestions, or criticism at various stages of this project. David Chamberlain, John Graham, Bruce Grant, Andr´e Lardinois, Anthony Podlecki, and Kurt Raaflaub have kindly made available to me various sources and resources or their unpublished and forthcoming work. Colin Day, Collin Ganio, and Jennifer Wisinski of the University of Michigan Press have been extremely helpful, prompt, and encouraging. I am especially grateful to my copyeditor, Jill Butler Wilson, with whom I have carried out a tacit dialogue through colored pencils: she has done everything she could possibly do to save me from my own imprecision. This work could not have been completed without the generous leave policy and a Becker Fellowship from Swarthmore College. I owe a great deal to Stephe Bensch for proofreading the manuscript and suggesting

viii

Preface

numerous corrections. His sensitivity to language has made my prose somewhat less contorted and easier to read. My son, Paul, has also provided help in matters of style and a great deal of encouragement. My daughter, Louise, has been my closest companion while I was working on this project: she has tolerated my frequent absentmindedness with the gentle irony that is typical of her nature. I am also grateful to Luigi and Mariella Vignolo for following my progress and cheering me on from afar. Finally, I thank my beloved parents, Paolo and Mariarosa Vignolo: this book is a partial testimonial of how much they have done for me for so many years and a tribute to their memory.

Contents

Introduction

1

Chapter 1 Narrative and Metanarrative

20

Chapter 2 Comparison 45 Chapter 3 Interpretation and Evaluation Chapter 4 Thoma 232 Conclusion

266

Bibliography

275

General Index

293

Index of Passages

309

134

Preface

My first thanks go to Donald Lateiner, once my teacher and now my friend, most generous through the years with advice, support, and information of all kinds. He has just recently read this book in its final draft, but he has certainly had to hear a great deal about it for a long, long time: I hope he has not been disappointed. Last year William Turpin, in the busiest of all seasons for a department chairman, took the time to go through an earlier, bulkier version of my manuscript to help me decide how I could cut it down to about half its size. I am immensely grateful to him for his heroic help, as I am to Ellen Bauerle, then of the University of Michigan Press, for her wisdom in prescribing the shrinkage. Carolyn Dewald, my much admired fellow Herodotean and always a source of inspiration, has read all existing versions of this book. She made herself available for countless written and oral critiques of my work and for lengthy telephone discussions; she has been most generous in sharing her work in progress with me. I also thank Deborah Boedeker, Gregory Nagy, Helen North, Martin Ostwald, David Potter, Gil Rose, Grace Ledbetter, David Konstan, Elaine Fantham, Eric Munson, and John Carleo for their support, suggestions, or criticism at various stages of this project. David Chamberlain, John Graham, Bruce Grant, Andr´e Lardinois, Anthony Podlecki, and Kurt Raaflaub have kindly made available to me various sources and resources or their unpublished and forthcoming work. Colin Day, Collin Ganio, and Jennifer Wisinski of the University of Michigan Press have been extremely helpful, prompt, and encouraging. I am especially grateful to my copyeditor, Jill Butler Wilson, with whom I have carried out a tacit dialogue through colored pencils: she has done everything she could possibly do to save me from my own imprecision. This work could not have been completed without the generous leave policy and a Becker Fellowship from Swarthmore College. I owe a great deal to Stephe Bensch for proofreading the manuscript and suggesting

viii

Preface

numerous corrections. His sensitivity to language has made my prose somewhat less contorted and easier to read. My son, Paul, has also provided help in matters of style and a great deal of encouragement. My daughter, Louise, has been my closest companion while I was working on this project: she has tolerated my frequent absentmindedness with the gentle irony that is typical of her nature. I am also grateful to Luigi and Mariella Vignolo for following my progress and cheering me on from afar. Finally, I thank my beloved parents, Paolo and Mariarosa Vignolo: this book is a partial testimonial of how much they have done for me for so many years and a tribute to their memory.

Contents

Introduction

1

Chapter 1 Narrative and Metanarrative

20

Chapter 2 Comparison 45 Chapter 3 Interpretation and Evaluation Chapter 4 Thoma 232 Conclusion

266

Bibliography

275

General Index

293

Index of Passages

309

134

Introduction But the greatest wonder to me of all in that region, at least after the city itself, I am going to describe. It is their boats that go down the river to Babylon, all round and made of leather. For in the country of the Armenians, who live above the Assyrians, they cut ribs of willow and stretch around them watertight hides from the outside to serve as bottom, without broadening the stern or shaping the prow, but making it round like a shield. Then they fill this boat with straw, and after loading on the cargo, they launch it so that it is carried downstream. They mostly carry palmwood casks full of wine. The boat is steered by two paddles and two men standing up. One draws his paddle in and the other pushes his outward. These boats are made very large or of smaller size. The largest bear a cargo of up to five thousand talents. In each boat there is one live donkey, more if the boat is bigger. After they arrive by boat to Babylon and sell their cargo, they then auction off the frame and all the straw, but the hides they load on the donkeys and drive back to Armenia. For it is not in any way possible to sail upstream, because of the rapids of the river. For this reason, they make their boat not of wood but of leather. After they return to Armenia driving their donkeys, they make other boats in the same way. These are the boats they have. (Herodotus 1.194)

The Problem of the Ethnographies Reformulated In 490 and 480 b.c., the Greek city-states managed to defeat the aggression of the autocratically ruled Persians. Herodotus’ account of those events two generations later proceeds from the conflict’s remote antecedents. It describes the rise and expansion of the Persian Empire until the time when, in the plenitude of its resources, it came to threaten the Greeks. But Herodotus’ “story” also includes minute ethnographic descriptions, whose reason for being appears less plausible. For decades, we have read the Histories with the assumption, more or less resisted but

2

Telling Wonders

never entirely defeated, that the ethnographies belonged to another project, not only deriving from a different tradition of prose writing, but also conceived at a different stage of the author’s career and with different aims and concerns, a project whose disiecta membra were artfully but somewhat pointlessly inserted into the historical narrative.1 The lists of customs and geographical features in the ethnographies only occasionally provide information we need for our factual understanding of changes of dynasty and campaigns. We have come to agree that though they enrich the history in some general sense, they are not fully integrated with it. We have recourse to several strategies for dealing with the issue of “unity” in Herodotus. The stylistic and conceptual analysis of Greek archaic literature in general has taught us not to apply our post-Platonic notions of unity to earlier works.2 We also do not, especially in our present interdisciplinary generation, regard ancient Greek ethnography and history as formally separate fields. Modern literary criticism outside the classics, for its part, has abandoned the requirement of coherence in a text; we now revel in its inconsistencies, contradictions, and omissions.3 But the fact remains that the ancient modes of discourse of history and ethnography, as these genres are combined in Herodotus’ work, are objectively different. The one diachronically recounts unique events of the past and relies on chronological and causal continuity; the other synchronically describes permanent conditions and customary actions in the present in a discontinuous catalogue form. We are always bound to seek, therefore, a better understanding of their mutual cooperation and of the substantial connection between their respective aims. So far, the best solution of all has come from those who have broadened Herodotus’ subject matter almost ad infinitum. They have attributed the diversity of his material to his attempt to explain the human 1. Jacoby (1913, 281– 362, especially 327– 34, 341– 55) attributes the genesis of the work to a conglomeration of originally independent logoi; the ethnographic logoi were composed earlier and have been integrated into a historical context with little modification. For a later developmental view, see von Fritz 1936. The unitarians either reject the notion of development and advocate Herodotus’ adherence to an original plan (see, e.g., Lattimore 1958) or focus on the conceptual unity of the work as we have it. See especially Pohlenz 1937, on the assumptions of which the work of both Immerwahr (1966) and Lateiner (1989) is based. The integration of the ethnographic logoi and other digressions in Herodotus is discussed by Cobet (1971), who also provides excellent surveys on unitarians and separatists (1– 44, 188– 98). 2. See Frankel ¨ 1924; Van Groningen 1958. Cf. Pohlenz 1937, 67– 87. On the history of the notion of literary unity in ancient Greek practice and criticism, see Ford 1990. 3. See Belsey 1980, 109.

Introduction

3

condition in a global way and with whatever evidence was available to a prose writer.4 This is too vague a recognition, however, for it does not eliminate the notion that Herodotus has lingered along the way of a goaldirected narrative—leading to his account of the Persian Wars—to include many items of information just “interesting for their own sake.”5 The problem of unity in the Histories, in other words, is strictly connected with that of their purpose. We should approach both issues by asking, What is the basic relation between the ideological position that emerges from Herodotus’ history of the conflict with the other and that which he transmits through his description of multiple others? My representation of what the Histories communicate to their intended audience is especially indebted to two very different but mutually complementary insights of recent scholarship. The first of these has been most lucidly formulated by Fornara as a general principle and through specific illustrations from Herodotus’ narrative. Herodotus wrote and composed his history of the Persian Wars over a period of time in the 440s, the 430s, and perhaps even the 420s and later: these are the years of Athenian imperialism, ever more threatening and oppressive to some of the Greeks before and during the first stage of the so-called great Peloponnesian War against Sparta and her allies.6 Herodotus’ perspective is therefore comparable, for example, to that of an author writing about World War I in the course of World War II.7 His view of the past is bound to be 4. According to Immerwahr (1966, 5), Herodotus “first discovered history as a method of understanding the world as a whole.” Cf. Cobet 1971, 177– 87. Lateiner (1989, 16) suggests a richer and more specific set of answers when he asserts that “ethnographic information in the Histories . . . is documentation deployed to assert an historical thesis, namely that mankind has benefited from ethnic and political separation and self-determination.” 5. The most radical denial of purposefulness in Herodotus comes from Howald ( 1923, 1944), but see also Frankel ¨ 1924, especially 87. The notion that something in Herodotus may be there simply because “it is itself worth reporting” in some obvious but unspecified way is still pervasive in some of the best Herodotean scholarship: see, e.g., DarboPeschanski 1987, 11; Fornara 1971a, 21, 23. 6. The terminus post quem for the publication of the Histories as a whole is 430 b.c., the time of the latest datable event mentioned in the work (7.137). On the basis of more indirect internal evidence and echoes of the Histories in other texts, Fornara (1971b; 1981) argues for a publication date close to 414 b.c. Cobet (1977) favors the more orthodox date of ca. 425 b.c. The composition and dissemination of Herodotus’ work must in any case have stretched over a number of years; see Evans 1991, 90. My argument does not depend on precise dating after 430 b.c., but I will briefly revisit the issue in chap. 3, “Divine Communication.” 7. I am modifying for the sake of simplification a more subtle parallel made by Fornara (1971a, 40).

4

Telling Wonders

affected by later developments and by the momentous political circumstances and ideological trends of his own time. In the case of Herodotus, observes Fornara, the interpretation of the past in light of the present is merely implicit: he only infrequently mentions the historical developments in Greece after 479 and even eschews many opportunities to do so in the course of his narrative. This is remarkable in an author who is so interested in complete historical processes that he habitually jumps backward and forward in time to show how a story began or how it ended. What we have at several points, then, is evidence of a deliberate silence about recent history and present circumstances, coupled with an ironical intent in the report of earlier events.8 Herodotus’ narrative capitalizes on the audience’s knowledge of how things turned out and draws its force and meaning from those later outcomes. By describing a narrative of the past shaped by the circumstances of the present, Fornara draws us toward envisioning the Histories as a work that, through the narrative of their recent past, communicates to the Greeks (Herodotus’ implied audience) things they should learn about themselves.9 This is, after all, a traditional role of historical or mythical narratives documented elsewhere—in the speeches of Homeric heroes, Pindar’s epinician does, or the funeral orations of Athens. None of the narratives told by speakers in the Histories claim any other goal or effect, such as entertainment or pleasure.10 It is in conjunction with this didactic character of the Histories that I wish to explore the role of the ethnographic descriptions. These occupy a considerable portion of the narrative, yet Fornara excludes them from his argument about the relevance to the work of the historical and political circumstances of Herodotus’ time. In his view, the history of the Persian Wars, particularly the last three books of the Histories, are one thing, 8. See Fornara 1971b, 152– 53. All Herodotean references to events after 479 b.c. are listed and discussed in Cobet 1971, 59– 78. 9. See Fornara 1971a, 75– 91; Raaflaub 1987; Corcella 1984, 186– 235; Munson 1988; Stadter 1992; Moles 1996. This direction of Herodotean scholarship is partially indebted to Strasburger 1955. 10. This is unlike the songs of Demodocus and Phemius or Odysseus’ narrative in the Odyssey (1.325– 52; 8.62– 92, 266– 369, 471– 541; 11.333– 476). See Marincola 1997b, 11. Although ancient readers emphasize the enchanting effect of Herodotus’ narrative (see Dorati 2000, 33– 47), the internal model for reading the Histories is provided, e.g., by the useful stories of Solon (1.30.4– 5), Socles (5.92), and Leotychides (6.86α– δ). See Stadter 1992, 781– 82; 1997. Within the narrative of the Histories, one audience who sets out to listen to a (musical/poetic) performance for the sake of pleasure is discredited (1.24.5). Cf. 6.21; 6.129.

Introduction

5

while the ethnographies are quite another: they might as well belong to another author. I have no objection to Fornara’s support of Jacoby’s theory, for example, that the Egyptian section was composed earlier than most other parts of the work and that it bears signs of having been written as an ethnography and not at all for the purpose of occupying the place it now does within a history. To explain the composite nature of the work and in particular the stylistic peculiarities of book 2, it would help to postulate that Herodotus may have started his intellectual development by looking outward at exotic phenomena, before directing his gaze inward at the Greek world in his maturity—that he tended toward a synchronic approach before turning to a diachronic research of causes.11 But the Egyptian and other ethnographies have been interwoven with the historical sections in the extant work.12 In light of the insights inspired by Fornara himself about the message Herodotus conveys to his Greek audience about themselves, it is time to reconsider what this message is, whether it informs all of the Histories or merely certain parts, and whether the ethnographies dilute and put it on hold or contribute to it in the special way that is consonant with their genre. The argument that the Histories communicate a teaching to a contemporary audience by means of narrative leads us to my second interpretive tool for reading Herodotus. This is offered by Gregory Nagy in his Pindar’s Homer, a book that aims at explaining the common roots and parallel developments of different modes of discourse of distinct archaic and classical genres. Within this broader discussion, Nagy approaches Herodotus’ Histories as a performance based on that same tradition of the ainos that became embodied in other types of performance: in the fables of Aesop on the prose side and in the poetry of Hesiod, Archilochus, Theognis, and Pindar. According to Nagy, Herodotus’ performance of historie is the prose counterpart of the ainos of Pindar’s epinician odes. Both are authoritative speech acts with the power to convey explicit as well as implicit 11. See Fornara 1971a, 1– 23. The distinct features of Herodotus’ persona in book 2 are discussed by Marincola (1987). However Fornara’s statement (1971a, 18– 19) that book 2 differs from other parts of the work on account of “the utter absence of the moral or philosophical element” seems wrong to me. 12. As Drews (1973, 84– 85) observes, the Histories only go to prove that Herodotus still had ethnographic interests at the end of his career. Drews’ explanation, however, that Herodotus’ descriptions of foreign peoples have the function of magnifying the Persian Empire and therefore the accomplishment of the Greeks in resisting its attack does not account for a great deal of ethnographic material in the Histories.

6

Telling Wonders

messages.13 Pindar places the here and now of the athletic victory in direct relation with the heroic achievements of the past by conferring epic kleos (renown) on the members of a privileged group within the aristocratic class of his time. At the same time, he also implicitly warns about the impending danger of degeneration of the aristocracy itself and of the consequent rise of a tyrant in the polis.14 The Histories of Herodotus, for their part, deal centrally with a relatively late historical past and do so more ambivalently than Pindar treats his paradigmatic time of heroes. The sociopolitical composition of the implied audience, the ideological stance of the text, and the specific substance of the message it conveys differ in many important respects from their counterparts in Pindar. But the past and the present, the explicit and the implicit, praise formulated in terms of kleos and warning about the threat of tyranny for the state, and a message of certain retribution for hubris based on the moral ideology of Delphi are all part of Herodotus’ discourse, as they are of Pindar’s.15 Nagy’s parallel between Herodotus’ historie and the ainos dovetails with Fornara’s and Raaflaub’s insistence on the relevance of the historical context of Herodotus’ time to the narrative of the Histories.16 Herodotus’ deliberate silence concerning events closer to his own time, emphasized by Fornara, is suitable to the implicitness of the ainos, to the obscurity of the related tradition of oracular poetry, and to the position of an author who is advising a politically diversified public about sensitive contemporary issues.17 The approaches of both Fornara and Nagy have provided foundations to my understanding of the Histories as a speech act bearing advice and warnings that address the present attitudes and behavior of the Greek cities toward one another and their future prospects. From the point of view of Nagy’s analysis, our initial question about the role of the ethnographic material in the Histories becomes, To what use has the genre of Greek ethnographic writing been put as part of a 13. Nagy 1990, 215– 338, followed by Payen 1997, 62– 74. 14. The features of the ainos of Pindar are described by Nagy (1990, especially 146– 51, 164– 66, 173– 74, 181, 186– 87, 192– 99). 15. For the parallel between the mode of discourse of Herodotus and that of Pindar, see Nagy 1990, 215– 338. 16. Fornara 1971a, 75– 91; Raaflaub 1987; Nagy 1990, 305– 16. Nagy characterizes Herodotus’ narrative about Croesus the Tyrant as “an ainos which applies to Athens” (309). 17. See Nagy 1990, 332– 35 for the element of ainigma (riddle) in the ainos and for the connection between Herodotus and the ainos, on the one hand, and wisdom and oracular poetry, on the other.

Introduction

7

discourse analogous to the ainos? One sign of adaptation of ethnography to the mode of the ainos in Herodotus is perhaps the one and only quotation from Pindar to be found in the Histories, the gnomic statement “custom is king of all” (3.38.4). Herodotus twists it out of its Pindaric meaning and gives it his own, precisely to use it as an interpretive bridge between his historical narrative about a hybristic king and his entire ethnographic project of describing customs. Moreover, in his polemical essay On the Maliciousness of Herodotus, Plutarch attributes to Herodotus the unitarian agenda of praising the barbarians and criticizing the Greeks and accuses him of doing so “deceitfully.”18 Later on, we will have other opportunities to test the extent to which Plutarch’s rather pedestrian view can be credited. For the moment, we should notice that what he calls “deceit” is related in part to Herodotus’ speaking obscurely through the narrative of the past, the words of oracles, and the representation of foreign peoples. In reference to a particular passage, Plutarch says that Herodotus there uses the Pythian god as his mouthpiece just as at other times “invents words and assigns them to Scythians, Persians, or Egyptians in the way that Aesop assigns his to ravens and monkeys.” Since Aesop’s fables are ainoi—narratives whose implicit, often political teaching could be understood from the context in which the author performed them—Plutarch’s assertion confirms Nagy’s argument about the analogies between Herodotus’ discourse and that of the ainos.19 Although the discourse of Herodotus feeds on the same traditions and assumptions that govern the ainos, Nagy rightly insists that its essential component is historie, “inquiry.” This is the process (and the product) of collecting evidence by seeing what is possible to see and by hearing the οδεξις, available verbal testimony: the Histories are a ι στορ ιης απ which means the public presentation of those accumulated data. Herodotus’ historie is a “scientific” undertaking, but it is also comparable to the process of inquiry in cases of judicial arbitration.20 In an inquiry of arbitration, the evidence collected serves as the basis for a judgment about who is right and who is wrong in a dispute and for the recommendation of a settlement on 18. See especially Plut. De Malign. Herod. 12– 14 ⫽ Mor. 857A– F. 19. Plut. De Malign. Herod. 40 ⫽ Mor. 871D. Concerning this passage and the connection between Herodotus and Aesop, see Nagy 1990, 322– 26, with evidence from the Life of Aesop on the context in which Aesop told his fables. See also Payen 1997, 69, 71. 20. See Nagy 1990, 315– 20, for the historie of Herodotus as a juridical process in the light of attested cases of arbitration between Greek states. On Herodotus’ scientific attitude and rhetoric, see Lateiner 1986; Thomas 1997; Thomas 2000.

8

Telling Wonders

the part of the arbitrator or histor. In a similar way, the information presented in the Histories counts as evidence in view of judgments and recommendations by the histor Herodotus.21 The dispute is ultimately that among the Greek city-states in the time of Herodotus himself. The judgment and the recommendation represent the message of the Histories as a whole. This is communicated for the most part implicitly, through evidence that is both ethnological and historical. Assyrian Boats and Other Particulars The discussion of Greeks and barbarians that pervades Herodotus’ work is based on a dialectic between traditional notions of the Greeks, on the one hand, and Herodotus’ more or less overt disruption of these notions, on the other. Though recent scholarship has enhanced our understanding of Greek ideology about the barbaroi, it has not paid sufficient attention to the extent to which the several sources that constitute our evidence for this ideology respond to it in particular ways.22 Drawing from a generalized perception partially disseminated through poetic texts and owned, as it were, by his audience, Herodotus lures his modern readers into a false sense of recognition, only to undermine it. Though he represents otherness according to culturally determined— one might say, unconscious and inevitable—patterns of thought, he devalues its familiar implications through a series of concomitant strategies. He occasionally sets up the other as a model of what the Greeks would consider appropriate behavior; he complicates knowledge to both confirm and confuse ideological stereotypes; or he counterbalances his representation of difference with indications of unexpected similarities between his ethnographic subjects, other groups of barbarians, and different groups of Greeks.23 Herodotus’ ethnographies are thematically unitarian descriptions, each discussing the customs, geographical situation, and historical traditions of this or that people. They are made up of particulars, which contribute to a 21. The juridical meaning of histor is explained by Nagy (1990, 262, 315– 20). See also Connor 1993. Dewald (1987) applies the term to Herodotus for the sake of a different distinction: to indicate the narrator and implied author of the Histories as he emerges from the work (on narrator, implied author, and real author, see Booth 1983, 67– 77; Chatman 1978, 147– 58; Genette 1980, 213– 14). 22. See, e.g., Hartog 1988, xxiii; Rosellini and Sa¨ıd 1978; Hall 1989. 23. See Pelling 1997.

Introduction

9

general representation of the people in question but also possess a measure of autonomy separate from the monographic section to which they belong.24 The uncertainty about the role of ethnography in the Histories arises again and again from many individual ethnographic details, posing each time a series of typical questions: What is the meaning of this feature? How does it contribute to characterize a given ethnic group? In what other ways does it relate to its ethnographic surroundings or to any other part of the work? Examining Herodotus’ discussion of the boats of the Assyrians (1.194) will provide a preliminary demonstration of the value of particulars as evidence in the political discourse of the Histories. In the phrase that introduces the description, “the greatest wonder to me in that land, after the city itself” (1.194.1), the city is Babylon. Herodotus has discussed it earlier, right after the narrative statement that Cyrus, having subdued many other peoples of Asia, now planned to conquer it (1.178– 87). Babylon is an architectural marvel, defined by its river and bearing the stamp of royalty. Surrounded by a double circle of walls, it stretches on both shores of the Euphrates, with one of two symmetrical structures rising at the center of each half: the sanctuary of Zeus Belos and the royal palace (1.178– 83). Babylon has been built and adorned by its rulers. The two clever queens whom Herodotus singles out  for mention have added to its wondrous quality (αξιον θωµατος, 1.185.3; αξιοθ εητα, 1.184). One has endowed the city with impressive dikes to prevent flooding. The later one, Nitocris, has detoured the course of the Euphrates by means of canals to the north and an artificial lake, to protect Babylon from the Medes. By draining the waters of the river into the lake and letting them flow back again, she also built across the Euphrates a bridge that connects the two sides of the city in daytime and is removed at night to prevent crime (1.184– 86). The already announced narrative of Cyrus’ campaign against Babylon resumes after the description of the city. Accompanied by mule-drawn wagons carrying royal food supplies and the drinking water of his native Choaspes River in silver vessels, the Persian king crosses another stream, the Gyndes, which he divides into 360 channels in punishment for drowning one of his sacred white horses (1.188– 89). Water and mighty walls seem to make Babylon impregnable. Cyrus, however, turns the hydraulic works of Nitocris to his own advantage: he drains the Euphrates into the artificial lake she had built and invades the city through the riverbed. 24. See Darbo-Peschanski 1987, 138– 39.

10

Telling Wonders

Distracted by festive pleasures, the Babylonians do not see him coming, and the city falls into Persian hands (1.190– 92.1). The campaign narrative is followed by another, more properly ethnographic description, this time of the Assyro-Babylonians and their land. This description begins by listing the resources the newly conquered province added to the Persian Empire, amounting to a third of those provided by Asia as a whole (1.192). A chapter on agriculture (193) explains this fabulous wealth: a hand-operated irrigation system makes Assyria the most fertile of lands. Cereal plants grow to unbelievable proportions. Oil is derived from sesame, and the palm trees yield a fruit used to make food products as diverse as bread, wine, and honey. Here, the boats carrying the merchandise down the Euphrates are introduced. In the various sections in the ethnohistorical Babylonian narrative so far—the description of the city, the account of Cyrus’ campaign, and the Assyrian ethnography—dealing with the river has been a test of sophie (cleverness) by which all the protagonists are evaluated, either implicitly or explicitly: the rulers of Babylon used the river to increase its safety and order; the foreign king Cyrus conquered Babylon by means of waterworks; the careless city dwellers let themselves be caught unprepared; unspecified people authored the irrigation system; and finally, when we reach our passage, the Assyrians take advantage of the current of the Euphrates for their journey from Armenia to the markets of Babylon. These merchants mark a transition from royalty to the society whose customs are then described in areas of culture common to all nations around the world—dress, marriage, health care, funerals, and religion. Sophie remains intermittently a factor (see 1.196.1; 1.197), down to the report of the peculiar Fish Eaters, who, as the other natives do with palm dates, are able to process a single ingredient into various forms (1.200). Like all texts, this narrative and descriptive section is intersected by a number of different and partially overlapping cultural codes or subcodes.25 As a resource or an obstacle, the river is not merely a term of the 25. On the notion of “codes,” see Barthes 1970; Munson 1991. The primary code of any text is the linguistic (English, Greek, etc.). Second- or third-order cultural codes (more properly called “subcodes”) are represented by the words and ideas used when speaking about something: in a typical nineteenth-century fictional text, e.g., Barthes (1970, 18– 20) identifies a proairetic code (code of action), a hermeneutic code (signaling the existence of a mystery or progress toward its resolution), a semic code (or code of connotation), a symbolic code, and an indeterminate number of cultural codes that communicate by making reference to a certain body of cultural knowledge shared by the implied audience (e.g., the terminology of a medical textbook will constitute in effect a special language, a technical medical code).

Introduction

11

codes of action, geography, and military strategy but also a symbol of what is given, one’s proper share, or that which delimits it.26 It participates, in other words—as is frequently the case with concrete features in Herodotus’ narrative—in the symbolic code of the text.27 In relation to the river, another code emerges, that of kingship, and the actions of kings also acquire symbolic meaning.28 Characteristic of both native and foreign rulers in the Assyro-Babylonian section is a control over the natural environment, a control that appears morally problematic and therefore dangerous. Cyrus’ royalty is defined at the outset by his ownership of the water of the Choaspes, which he and all Persian kings carry around for their own special use wherever they go (1.188). Cyrus’ emasculation of the Gyndes corresponds at long range, backward and forward, to all the instances of expansionist violations of rivers in the Histories. The first of these occurs when Croesus initiates war against Cyrus and crosses the Halys, which some say he split into two streams (1.75). The last are represented by Xerxes’ abuse of the Hellespont (7.35) and by the advance against Greece of his large Persian force, which drinks rivers dry.29 Unlike the violations of Croesus and Xerxes, Cyrus’ assault on the Gyndes has no negative results. When he drains the Euphrates and captures Babylon and its wealth, his cleverness is crowned with success. Ethically, however, his conquest is on a continuum with two later royal thefts by Darius and Xerxes, mentioned in the same narrative (1.187, 1.183.3). Cyrus himself will meet with an unfortunate end in due time, after crossing another stream, the Araxes, in his attempt to subjugate the Massagetae (1.208).
26. Croesus’ crossing of the Halys involves an attempt to acquire land beyond his µοι ρα, “share,” (1.73.1). The symbolic interrelation of physical and moral boundaries in Herodotus is discussed by Lateiner (1989, 126– 44). See also Immerwahr 1954, 19– 28; Immerwahr 1966, 325; Konstan 1983; Stadter 1992, 785– 95; Payen 1997, 138– 45. 27. Physical objects as conveyors of meaning in Herodotus are discussed in Dewald 1993. On the symbolism of metals, see Kurke 1995. Symbolic action in Herodotus is discussed by Payen (1997, 29– 31 and passim). Any word or narrative element in a text normally belongs at once to many different codes and to codes within the codes (e.g., the word heart belongs to the codes of love, physiology, medicine, cardiology, etc.). The symbolic code draws terms from overlapping codes and helps to reshuffle and reassign them to yet other codes. So in Herodotus, the river becomes, through the symbolic code, a part of the political code and the code of ethics, the languages in which we talk about such issues. 28. The actions and features typical of kings (i.e., all the terms of the “monarchical code”) are also terms of the symbolic code when they not only apply to real kings but become the means for talking about, for example, the human condition and human actions in general. See chap. 2, “Analogy as an Interpretive Tool.” 29. See 7.21.1, 7.43.1, 7.58.3, 7.108.2– 109.1, 7.109.2, 7.127.2. There are twentyfive rivers mentioned in the account of Xerxes’ march from Critalla to Thessaly alone (7.26– 131).

12

Telling Wonders

The digging, draining, and bridge building of the Babylonian queens have more benign motives and do not entail trespassing into someone else’s land. These acts also represent, however, monarchic imperialism over the environment. Like Nitocris’ attempt to impose order on Babylon from above, these acts are useless given the citizens’ lack of discipline and prove counterproductive in the face of the enemy attack. The folks whom Herodotus describes as making their living through the river have neither leisure for lapses nor monarchic opportunities to exercise intrusive mastery on earth and waters. The boats of Assyrian merchants are constructed of inexpensive materials (timber, hides, and straw), which are then recycled and reassembled. Unlike royal masonry work, they do not mark the land. To some extent, like the leather trousers of the Persians of long ago (1.71.2), they connote primitivity.30 They are crude in shape, all round, with no stern or prow. More “like shields” than like ships, reactive rather than offensive, they are designed not to cross the seas, to be steered in different directions and against the waves, or to dominate the elements but to drift downstream helped along by paddles. Very large or less large, they only require a crew of two men. Once in Babylon, the merchants sell their wares and the framework of their boats, then retrace their way by land on their donkeys, because the swiftness of the current prevents them from sailing back upstream. They display not only industry and sophie but also what we call common sense and the Greeks would include in their notion of sophrosune.31 The description of the collapsible boats and other details in the ethnography characterize the Assyrians in a way that on the whole tends to redefine the issue of difference: though their boats are peculiar, their market of brides preposterous, and their diet exotic, a talent for problem solving makes the Assyro-Babylonians similar to the Greeks. Sophie is a virtue the Greeks claimed to possess to a higher degree than all barbarian nations (see, e.g., 1.60.3). To counteract such ethnocentric assumptions by means of specific evidence and praise is an important goal of Herodotus’ historie. 30. Connotation, activated through signs or indices (the semic code: cf. n. 25), is closely akin to symbolism but involves a more immediate and less constructed connection between signifier and signified. See Barthes 1970, 6– 9, 19; Prince 1987, s.v. “seme.” 31. See Geertz 1983, 71– 93, on our tradition of regarding common sense as a transcultural category (“what anyone clothed and in his right mind knows”). On sophrosune, see North 1966, especially 28, where North notes that although Herodotus uses sophr- words sparingly, he “is the most fertile source in Greek prose of stories illustrating traditional ideas of sophrosune.”

Introduction

13

At the same time, however, Herodotus’ particulars often appear, as in this case, to have less to do with the construction of the national identity of a foreign people than with projecting onto a faraway setting pieces of a problematic that is entirely Greek. One well-known ideological contradiction in the mid– fifth century was that of the simultaneous desirability of exceptional (i.e., symbolically “royal”) status and normal (or citizen) status, as is illustrated in the Histories, for example, by the contest of happiness between Croesus of Lydia and Tellus of Athens.32 Throughout Herodotus’ work, the representation of the otherness of foreign peoples competes with the representation of the cross-cultural otherness of kings. By setting the pragmatics of everyday life, exotic but legitimate, side by side with the behavior of the powerful agents of history, his ethnographies represent a crucial part of a discourse at once “democratic” (almost in the modern popular sense of the word) and anti-imperialistic. To the talking Scythians, Persians, and Egyptians whom Herodotus uses, according to Plutarch, for saying what he wants to say, we may here add the silent Assyrians. The description of their boats, like the ainos of Tellus, provides a foil for the actions of kings. The ethnographic style of the Histories, in the Assyro-Babylonian logos and elsewhere, consists mostly in the factual description of particular features, with little interpretation of their meaning. What has been characterized as Ruth Benedict’s typical mode, the general description of a people’s worldview, repeated again and again and illustrated by examples of what they do and say, is in general alien to Herodotus.33 In the Histories, the point of what “they” do and the reasons why “we” should find it interesting have to emerge from the description of what they do. The near absence of explicit interpretive commentary fits the tradition of Greek ethnographic writing as well as Herodotus’ mode of implicit communication. In the case we have examined and several other times in the Histories, celebration of the subject stands in lieu of interpretation. The initial sentence of the passage communicates to the recipient of the narrative that the feature is important or, in other words, meaningful. The phrase  “great wonder” [θωµα µεγιστον] establishes its membership in the broad category of “achievements great and wonderful, some performed by Greeks, some by non-Greeks,” that, according to the first sentence of the 32. 1.30.3– 5. On the Greek political and ethical discourse concerning tyrannical power, see McGlew 1993, especially 30– 32, 196– 212. On kingship as symbol, see my n. 28 in this introduction. 33. For the description of Benedict’s style, see Geertz 1988, 108– 20.

14

Telling Wonders

work, are deserving of permanent renown. The researcher and narrator named in that first sentence, Herodotus of Halicarnassus, here comes into the open—with “the greatest wonder to me”—to claim the inclusion as his own idiosyncratic choice. When he says, “the greatest wonder to me after the city,” he creates a paradox and helps to establish at the outset the symbolic nexus between two discrepant items, “leather boats” and “Babylon,” the people and their kings. This is, on the whole, a more eloquent directive than simply a red mark on the page. It is almost as eloquent and interpretive, in an implicit way, as the famous title that announces one of Ruth Benedict’s ethnographic inquiries, The Chrysanthemum and the Sword.34 The Histories as Performance Announcements and glosses such as the one we have just considered signal the presence of the one who is telling the story and his reactions to it. They represent the metanarrative component of the discourse, the study of which is a tool for better understanding the message, ideology, and tone of a text. In the case of Herodotus, the metanarrative often enhances our awareness of a direct verbal communication from a speaking author to a listening audience. The Histories as we have them are obviously a written text, and they contain internal signs of writing as a tool of composition and publication.35 These signs are less conspicuous, however, than those indicating or imitating an oral situation for Herodotus’ speech act.36 As Nagy maintains, the work is “not a public oral performance as such,” but it is nevertheless “a public demonstration of an oral performance, by way of writing.”37 34. Benedict (1989, 2), with characteristic explicitness, explains her title: “Both the sword and the chrysanthemum are part of the picture. The Japanese are, to the highest degree, both aggressive and unaggressive, both militaristic and aesthetic, both insolent and polite . . .” 35. Examples are the authorial “I write” occurring side by side with “I say” in programmatic statements (see chap. 1, “Types of Introductions and Conclusions” and “SelfReferential Glosses”); long genealogies and catalogues (see Thomas 1989, 189; Harrison 1998); perhaps the past tense in metanarrative sentences of the type “it could be seen still in my time” (see Rosler ¨ 1991); and, of course, the statement in the first sentence that Herodotus intends to preserve the glory of great deeds and save them from oblivion. 36. See Pohlenz 1937, 208– 11; Lang 1984, especially 1– 17; Munson 1993a; Stadter 1997. 37. Nagy (1990, 220) so argues on the basis of the first sentence. For the notion of performance applied to the Homeric poems, see Martin 1989, 1– 42. We do not, of course,

Introduction

15

We have some external evidence, more or less direct, of historical Herodotean performances.38 Our written text may theoretically stand in relation to these as a special version of a larger potential performance, a reservoir for potential performances, or a collection or summary designed to extend the life and usefulness of actual or possible contemporary performances. Since I approach the Histories in this way, considering it a text that is unitarian but not definitive, I will not emphasize its overall structure (each narrative in the exact place in which it stands) and integrity (the original presence of all the narratives it now contains) as primary conveyers of meaning.39 Thematic and factual cross-references so multifariously pervade the work and spill over onto the unmentioned real world of Herodotus’ audiences that, to communicate its message, a given narrative may relate in fairly specific ways to the immediately surrounding narratives though it is not entirely dependent either on them or on what is now the whole. The narratives of the Histories have the potential for joining up with many other narratives that are now far away or simply not there at all. Like the Assyrian boats, Herodotus’ Histories lend themselves to being disassembled and reassembled into smaller (or, theoretically, larger) functional wholes. The apparent elasticity of the Histories recalls to some extent the circumstances of rhapsodic performances of the Homeric texts in classical times. It is also internally indicated by their relative open-endedness, by the unfulfilled promises they contain of narratives that could have been included but are not, and by the contrived and not inevitable character of many of the transitions between narratives.40 The one and only postulate the same sort of orality for the Histories as one may for the composition of the Homeric poems. But see the parallel drawn in Murray 1987, 107. Most importantly, Herodotus’ use of writing at a moment of transition between preliterate and literate conditions does not entail a widespread readership from the beginning, and an audience of listeners is more likely. On the related issue of literacy, reading, and the aural reception of literature in the fifth century b.c., see Davison 1962; Flory 1980; Havelock 1982, 146– 47; Thomas 1989, 15– 34. 38. See Plut. De Malign. Herodot. 26 ⫽ Mor. 862A– B (⫽ Diyllus, FGrHist 73 F 3); Lucian Herodot. 1– 2; Marcellin. Life of Thuc. 54; Eus. Chron. Olymp. 83.4. This evidence is rejected by Powell (1939, 31– 36) and Johnson (1994). But see Parke 1946, 86– 87; Momigliano 1978, 62– 66; Evans 1991, 89– 132; Dorati 2000, 17– 52. 39. Structure has been emphasized by Wood (1972, especially 19) and productively studied by Immerwahr (1966). See, however, Immerwahr 1966, 308. I mean not to devalue the meaningfulness of the present structure but to suggest that other combinations of logoi would be just as meaningful. 40. On the end of the Histories, see Dewald 1997, with full bibliography on the previous scholarly discussion. On its unfulfilled promises, see Drews 1970 (discussing

16

Telling Wonders

performance available to us, in other words, possesses a retroactive unity and also partakes of the incompleteness or fluidity of the various potential performances of which we can conceive. As a display in front of an audience, the Histories are a “performance” in the ordinary sense of the word. They are also a “performance” in a narrower sense, from the point of view of speech-act theory. They represent an utterance that does something, performing a range of worldchanging operations. At the surface level, they inform, preserve the memory of events, and establish a record of praise and blame. Implicitly and at a deeper level, they predict, advise, warn, attempt to persuade, or promote a certain predisposition in the audience.41 The substance of these implicit predictions, advice, and warnings represents, once again, the message of the work. The various functions (or illocutionary acts) that the Histories perform at different levels through the narrative are facilitated by the metanarrative. Narrative statements (whether recounting or descriptive) are, theoretically, world-describing utterances: any other eventual functions they might perform (e.g., warning) are especially indirect and disguised.42 The narrative records what is real, simply because it is real. With the metanarrative, however, the author of the narrative emerges from the text. As an explicit mediator between reality and the audience, he communicates to the audience about his record of reality and its reason for being.43 1.106.2 and 1.184); Nagy 1990, 235 n. 91 (on 7.213.3). The artificiality of Herodotean transitions between different logoi is noted by Jacoby (1913, 343– 46). Within modern (written) literature, an approximate parallel to the rhapsodic nature of the Histories is provided by what Genette (1980, 149) calls the “unity after the event” of Proust’s A` la recherche du temps perdu, a work made up of heteroclyte material that was being continuously rearranged and revised by the author even during the last stages of publication. See also Eco’s (1962) definition and description of what constitutes an opera aperta and an opera da finire. 41. For the conjunction of the ordinary and speech-act meanings of the terms performance and performative, see Martin 1989, especially 5, 47. Speech-act theory was initiated by Austin (1962); for subsequent developments, see especially Searle 1976; Bach and Harnish 1979. Application of speech-act theory to works of literature is discussed by van Dijk (1976), Pratt (1977), Chatman (1978, 161– 66), and Searle (1979, 58– 75). 42. They are true-or-false statements, such as “Napoleon won the battle of Austerlitz” or “the earth is flat” (see Austin 1962, 132– 49). In Searle’s subsequent classification of five categories of illocutionary acts (1976), they are called “representatives.” 43. The presence of the narrator as a more or less overt mediator of narrative statements (and his absence in “nonnarrated” narratives) is discussed by Chatman (1978, 146– 266). See also Genette 1980, 212– 62; for Homer, see Richardson 1990, especially 1– 8. For Herodotus, see Dewald 1987, forthcoming a; de Jong 1998.

Introduction

17

Metanarrative signs are of course embedded in the narrative throughout. Narrative as pure transcript of the real world exists only in theory, because words can achieve exact mimesis of other words only, not of actions and states.44 But in Herodotus we are faced with an especially massive metanarrative component consisting of introductions, conclusions, and glosses that in some way summarize and explain the narrative (“X is like Y,” “I will tell X,” “X means Y,” “I have finished with X”).45 These statements stand out from minimally narrated narrative statements: Prince writes, “their appearance is similar to that of a (fragmentary) text in the text, representing a language that is other in the language of the text and establishing some of the interpersonal coordinates of a communicative situation.”46 Many contain a grammatical sign of the narrator (e.g., “the greatest wonder of all to me . . . I am going to describe”), but others do not. Since they manipulate the narrative, they are also more likely to evaluate, convey advice, question, argue, promise, express the narrator’s state of mind, and even bring a state of affairs into existence (e.g., convict)—in sum, to reveal purposes that go beyond simple representing or reporting.47 The Contents of This Book Different types of metanarrative sentences are briefly classified in chapter 1, which serves to establish the basic distinctions and terminology on which the rest of the analysis in this book is based. I then proceed to examine two overlapping functions pursued in Herodotus’ discourse by narrative and metanarrative means: comparison and interpretation. The 44. Hence, although we say that nonnarrated narrative (reality recording itself) and even its closest approximation, minimally narrated narrative, are theoretically “mimetic,” Plato uses the term mimesis only in reference to narrative in its dramatic form (Rep. 392d– 394c). In the case of a speech reported by a narrator in oratio recta, no mediation is visible between “reality” and the text. By contrast, actions require the mediation of a narrator, who puts them into words through the narrator’s own discourse. 45. Dewald (1987, 148– 50) compares Thucydides’ narrative surface to “perfectly transparent glass through which one is encouraged to imagine one is directly perceiving the res gestae narrated” and argues that Herodotus’ narratorial interruptions resemble rather “those little decals—flowers, rainbows and whatnot—scattered by the cautious on the surface of the glass.” 46. This quotation, well suited to Herodotus’ discourse, is part of Prince’s description of metanarrative in general (1982, 127). 47. See, e.g., Herodotus’ “conviction” of Ephialtes (7.214.3), discussed in my chap. 3, “Interpretation in the History.”

18

Telling Wonders

first of these crisscrosses the text in two separate directions: diachronically for the events of history and synchronically for ethnographic and other phenomena separated not by time but by space. Figuring out how each of the two types work and how they cohere with one another is the task in chapter 2. Through the second and most encompassing function, interpretation, the text signals what facts mean and evaluates their worth. Here once again, history and ethnography go their separate ways. As I show in chapter 3, the political message of the work depends on the cooperation between a relativistic ethnographer, who interprets little and evaluates cautiously, and an absolutist historian, who explains historical action in moral terms. Herodotus’ message is nuanced, communicated not without obscurity and even, despite the charm and lightheartedness of his style, a certain degree of effort. It requires that we bridge our distance from the intended audience to which it is directed by a painstaking analysis that will not, however, entirely protect us from interpretive risks. The opacity of the Histories partially results from the uneasy series of mediations on which they rely. What is far away in space and time serves to talk about the present of narration, but the burning issues of the audience’s here and now are treated, as I have said, as almost unmentionable. Particulars not only communicate general principles but also provide contradictory evidence. Reality is represented through the narratives and arguments of sources and characters, some more reliable than others; the narrator often declines to unify the plurality of voices.48 Herodotus’ use of the terms of the Greek/barbarian antithesis frustrates his project of subverting that antithesis and redistributing the criteria of otherness. Herodotus’ attempt to reconcile the Greeks among themselves is based on a representation of a culturally special and homogeneous Greekness; yet this goal collides with his need to assign them to the shuffle of humanity, showing that they are almost just as different from each other as from non-Greeks. The instability and uncertainty of a logos that undermines itself as it goes along reflect the gaps and inconsistencies of the ideology to which Herodotus responds from within. They are also the mark of a historian who, if we will not find him unfailingly honest and objective, is still, in the way he displays evidence, more honest and objective than most.49 48. See Dewald 1987, 160– 63, 167– 70. 49. On Herodotus’ early reputation for dishonesty or lack of seriousness as a historian, see Evans 1968. Some modern scholars deny both Herodotus’ accuracy and the authenticity of his source citations. See especially Fehling 1989; West 1985; Hartog 1988, especially

Introduction

19

Herodotus’ speech act stands at the intersection of the three fundamental kinds of discourse. Just as we can distinguish sentences that are declarative (impart knowledge), imperative (give orders), and interrogative (request information), so we can, on the basis of these categories, identify corresponding types of text.50 As the exposition of the results of an inquiry and as a narrative of past events, the Histories conform to the declarative modality of discourse. I argue, however, that by virtue of its message, the work is also an imperative text: it “invites the reader to adopt a position of struggle rather than stability, specifically struggle in relation to something which is marked in the text as . . . existing outside discourse, in the real world.” But to the extent that the recipient of Herodotus’ logos is caught up in contradictions and invited “to produce answers [the text] implicitly or explicitly raises,” the Histories also constitute an interrogative text.51 The combination of the declarative, imperative, and interrogative modalities in the discourse of Herodotus’ Histories is inscribed in the recurrent concept of thoma, “wonder,” which is the topic of chapter 4. Appeals to wonder belong to the celebratory function of the text, and “a wonder” is a fact that wants to be narrated—incredible if true, hard to imagine if unknown. The narrator shares it as information, demands a reaction, and leaves us wondering what it means for our understanding of the Histories. With the term thoma, the text legitimizes the question that listeners and readers are bound to ask about any other apparently optional fact included in the Histories: why is this here? Since thoma words advertise onetime occurrences, lasting individual achievements, geographical features, and cultural artifacts, an exploration of their meaning will throw further light on the ability of metanarrative to bridge the distinction between ethnography and history.

364. Cf. Pritchett 1993; Fowler 1996, 76– 80. My work investigates meaning rather than factual truthfulness; an investigation of the meaning Herodotus attributes to facts he narrates would be even more urgent if he had invented them rather than seen or heard them. In practice, however, I believe that Fehling and others have not proven their case and that the gulf between the narrator and the real author is not that wide; see the more subtle proposal by Marincola (1997b, 14– 19). The dishonesty of Thucydides for the sake of advocacy is a newer issue. See Badian 1993, 125– 62; cf. Hornblower’s narratological discussion of Thucydides’ rhetoric in presenting the evidence (1994a). See Moles 1993. 50. See Benveniste 1971, 110, quoted by Belsey (1980, 90). 51. See Belsey 1980, 91– 92, for the definitions quoted.

Chapter 1

Narrative and Metanarrative

The aim of this chapter is to identify different levels of narrative in Herodotus’ text. I first define narrative in the strict sense, as opposed to metanarrative, and then distinguish self-referential from referential metanarrative. My discussion is especially indebted to three sets of works: narratological studies outside the field of classics,1 studies that apply narratological principles to Herodotus,2 and the work of other scholars who have devoted special attention to the formal aspects of Herodotus’ narrative.3 The definitions I present here are largely my own and formulated strictly as a function of my overall interpretive task. I keep unfamiliar terms to a minimum and avoid making theoretical points for their own sake. Hurried readers more interested in substantive issues of interpretation than in the approach offered here have the option of skipping this chapter and referring back to it later if needed. What Is Metanarrative? The Histories contain a multiplicity of stories shaped and held together by discourse and transformed by it into a single story with a logical, if rambling and open-ended, plot.4 Transitions between stories may be deter1. Genette 1980; Bal 1985; Chatman 1978; Labov and Waletzky 1966; Labov 1972. See especially Prince 1977, 1982, 1987; Barthes 1986. 2. Dewald 1987, 1999, forthcoming a; Fowler 1996; de Jong 1987, 1998; Richardson 1990; Hornblower 1994a; Rood 1998. 3. Especially Immerwahr 1966; Beck 1971; Wood 1972; Muller ¨ 1980; Pearce 1981; Munson 1983; Hartog 1988; Marincola 1987. 4. These narratives more or less correspond to the units Immerwahr (1966, 14) calls logoi. See also especially Immerwahr 1966, 46– 48, 329– 62. On the distinction between story and discourse, see Chatman 1978, 19: “the story is the what in a narrative that is depicted, discourse the how.” Other narratologists make more refined distinctions and use different terminologies. For example, Bal (1985, 1– 10) adds a useful definition of text as an upper layer of the communication: “a text is a finite, structured whole composed of language signs.” In other words (those of de Jong 1987, 31), “that which the hearer/reader

20

Narrative and Metanarrative

21

mined by historical landmarks along a chronological sequence, by changes of time, or by changes of place and subject matter, but always on the basis of some factual connection. On the whole, the narrative proceeds chronologically, but the discourse interrupts the story sequence by constantly introducing explanations and expansions of this or that story element.5 In most cases, these formally subordinated narratives recount events belonging to a specific previous or later story time (flashbacks or follow-ups) or are descriptions in the present tense.6 In my definition, “narrative” includes both the recounting of events in the past and description.7 Description, in whatever tense, displays objects, beings, situations, and actions “in their spatial, rather than temporal existence, their topological rather than chronological functioning, their simultaneity, rather than succession.”8 In Herodotus’ ethnographic descriptions, the present tense describes circumstances that may also obtain at the time reached by the historical narrative to which the description is attached. Whether it does or not, the ethnographic present is at any rate a real present, referring to the time of narration.9 Just as he instructs the audience about what happened in the past, so Herodotus teaches them about the contemporary world. Whereas narrative represents the story as it is manipulated by the discourse, metanarrative speaks about the narrative and exists as a function of the discourse. Minimally narrated narrative consists of passages that approximate the concept of pure narrative, or objective mimesis, of hears/reads is a text.” When I say “Herodotus,” unless the context makes clear I am indicating either the historical author or the narrator, I am referring to the “text” in this sense. 5. The discourse devices used in archaic and early classical Greek literature for connecting semiautonomous items of a chain are discussed by Frankel ¨ (1924, especially 62– 67) and Van Groningen (1958, 36– 50). 6. See Pearce 1981. Genette (1980, 35– 85) calls narratives involving a change of time “anachronic” (either “analeptic” or “proleptic”). In a few cases, Herodotus’ inserted narratives are chronologically parallel (see, e.g., 3.39– 60) or indeterminate (see 1.23– 24). 7. This definition, functional to my analysis, differs from that of most literary theorists, for whom narrative only concerns specific events in a temporal sequence of two or more. See, e.g., Labov and Waletzky 1966, 20– 32; Labov 1972, 359– 62; Bal 1985, 1– 10; Prince 1982, 1– 4; Prince 1987, s.v. “narrative.” 8. Prince 1987, s.v. “description.” 9. Hartog (1988, 254– 55) inexplicably denies this. Even if we wished to attribute a certain timelessness to the “gnomic” present, that would not apply to the ethnographic present. See, e.g., the timing of the Persian ethnography discussed in chap. 3, “Persian Ideology.”

22

Telling Wonders

external facts.10 Certain propositions, however, fall partially or entirely outside of the narrative and are equivalent to or contain titles, proems, repetitions, postscripts, or explanations that fulfill the role of glosses to the narrative itself. These metanarrative sentences especially appear as a sort of “padding” between adjacent or concentric narratives.11 At 7.57.1, for example, the minimally narrated narrative sentence “a mare gave birth to a hare” represents the core of a larger story sequence: *

s-i s-ii s-iii s-iv

Xerxes’ army crossed the Hellespont. A mare gave birth to a hare. They saw it. They proceeded on their way.

In Herodotus’ discourse, however, event s-ii stands out by itself. What precedes and follows is predominantly metanarrative, containing event s-i in a subordinated clause and incorporating event s-iii: 1a.

I

n G

When all [the Persian troops] had crossed, while they were moving on their way, a great prodigy [τερας µεγα] appeared to them of which Xerxes took no account, though it was υµβλητον]: easy to interpret [ευσ for [γα ρ] a mare gave birth to a hare. This was easy to interpret because Xerxes was about to lead an army against Greece with the greatest pomp and magnificence but would come back to the same place running for his life.

10. See my introduction, n. 44 and corresponding text. All narratives are of course “narrated” to different degrees, and we could discuss the internal signs of narration in each case. Here I am concerned with making a basic distinction. 11. For the combination of an introductory and a concluding statement framing a narrative in Herodotus, see especially Immerwahr 1966, 12, 52– 58. Cf. Pohlenz 1937, 72, 208– 10; Beck 1971, 11– 17, 57– 59; Muller ¨ 1980, 79– 80. On the concept of metanarrative, I am applying very broadly the definition by Prince (1977, 1– 2): “Chaque fois que le discours narratif (au sens large) renvoie au code qui le sous-tend ou, plus sp´ecifiquement chaque fois qu’il accomplˆıt (paraˆıt accomplir) une function de glose par rapport a` l’un de ses propres e´ l´ements, nous avons affaire a` des signes m´etanarratifs.” See also Prince 1980; 1982, 115– 28. The “shifters” discussed by Barthes (1986, 128– 30) and Fowler’s “markers of the historian’s voice” (1996, 69– 76), including, among others, all statements in the narrator’s first person (for which see Dewald 1987), are all part of the metanarrative as I define it.

Narrative and Metanarrative

23

It will become clear later why I identify statement I as an introduction and statement G as a concluding gloss rather than as a conclusion. What matters now is that both statements I and G are predominantly at a different level of discourse with respect to the central narrative sentence. Their main function is to “read,” summarize, or explain. They perform, in other words, some of the operations a reader/listener might perform, and they do so from a perspective that, like that of the recipient, is not an integral part of the action narrated. This commentary, moreover, leads the narrator to postpone s-iv after he has attached to this story the narrative of a chronologically anterior omen, similar to the one just narrated. The result is a narrative preceded by its own summarizing introduction (7.57.2), which in the present context represents a gloss to the preceding narrative of the mare/hare omen. This is followed by a sentence (CC) that both concludes preceding narratives and narrates story function s-iv. 1b. ⫽ G

I n

CC n


 Also another prodigy [ετερον . . . τερας] occurred for him when he was still in Sardis: for a mule gave birth to another mule with a double set of genitals, male and female, the male on top. Taking neither of these two into account, Xerxes moved forward. (7.57.2– 7.58.1).

A contrasting example to this set of metanarrative interferences is provided, for example, by a minimally narrated narrative reporting what Astyages learned about the meaning of his daughter’s two successive dreams and how he reacted to the information (1.107– 108.3). Astyages is the embedded focalizer of the events; whoever is telling this story (Herodotus or one of the sources mentioned at 1.95.1) is almost invisible.12 In the case of the hare giving birth to the mare, in contrast, while 12. On focalization, see Genette 1980, 185– 210; Bal 1985, 100– 118; de Jong 1987, 29– 36 (in the Iliad); Hornblower 1994a (in Thucydides); Dewald 1999 (in Herodotus and Thucydides). The narrator is first and foremost the one “who speaks”; the focalizer, the one “who sees.” While narrating always entails focalizing (so that the narrator is by definition the primary focalizer), the narrator may report the focalization of someone else (the embedded focalizer); or he may report the character’s act of seeing as a pure event, as Herodotus does when he says that the hare omen “appeared to them.” The distinction between narrative and metanarrative in the history can be described in terms of different focalization when the illusion that there is no narrator is achieved not by means of objective recording of

24

Telling Wonders

the agent in the narrative marches on, the narrator, Herodotus (and this time we are sure it is he), comes forward to communicate his perception. He and his audience come to share an understanding about the discrepan υµβλητον)  cies between the clearness of divine communication (ευσ and men’s failure to respond appropriately and between the initial magnificence of Xerxes’ expedition and its anticipated outcome. The stories of Astyages and Mandane, on the one hand, and that of the omens during Xerxes’ march, on the other, illustrate different discourse possibilities in the Histories. Metanarrative introductions or conclusions may subdivide the narrative at any point; the resulting narrative sections may be theoretically as extensive as the entire work, as small as the smallest segment, or of any extent in between. Introductions (most fre give a preliminary summary that identifies quently with continuative δε) a section of the following narrative as a unit. Conclusions summarize in some way what has been narrated, identifying it as a unit that has ended. Rather than being connected with δε to what precedes, most of these  (or και . . . µεν,  µεν  νυν, µ εν  δη),  to conclusions have anticipatory µεν enhance the mechanical connection of the passage that has just ended with what follows.13 Introductions and conclusions contribute to clarifying the subdivisions of a complex work, but their purely organizational function is secondary to my analysis. Especially interesting, however, is how their form, force, and interpretive potential indicate a more self-consciously didactic undertaking than that performed, for example, by Homeric poetry. Just as the histor is personally involved in investigating his subject in a way that the Muse-inspired bard is not, so he is also in close contact with his public. The way in which he speaks to them and guides their listening, however, is often ambiguous and reflects the complexity of his message. Types of Introductions and Conclusions I begin this discussion of metanarrative by treating introductory and concluding statements because in Herodotus, they are particularly numerous, discrete, and visible. They represent in themselves glosses to the text and thereby attract the presence of other glosses of various types, which can be external events but by means of a narrative focalized through a character. In metanarrative statements, by contrast, we always perceive the presence of the narrator-focalizer. See Marincola 1997b, 9. 13. See Frankel ¨ 1924, 83; Muller ¨ 1980, 77– 78; Immerwahr 1966, 46– 58.

Narrative and Metanarrative

25

found scattered along the narrative (e.g., in parenthetical statements or at the end of a sequence) or within it (in qualifiers). Introductions and conclusions are, in other words, privileged pockets of metanarrative communication. I will briefly survey their basic forms before discussing their general effect on the recipient of the narrative. All introductions and conclusions contain a summary of the narrative they identify, but what I call a summary conclusion is just that—an autonomous plain restatement of the whole or of parts of the preceding account, with no other fixed characteristics.14 For example, the sentence  ο Περσ  δ η υπ  ησι ε δεδουλωντο  2. Λυδο ι µ εν [The Lydians, then, were definitively enslaved by the Persians] (1.94.7) does not mention a new event in the narrative sequence but rather recaps the earlier account of Croesus’ war against Cyrus by rephrasing its result, which has already been recorded (though in different words) along with  δη anticipate a all the other stages of the action. The particles µ εν continuative δε in the introduction to the narrative that follows (1.95.1). The pluperfect tense of the summarizing verb marks the point at which the narrative had arrived before the intervening Lydian ethnography (1.92– 94)—Where were we? Ah, yes: the Lydians had lost their freedom. When an element of summarization on which the emphasis of the sentence lies is either replaced or accompanied by a backward-looking    demonstrative—a form of ουτος, τοιουτος, τοσουτος, or, less often,
οδε—the conclusion is no longer a plain summary. I call it a retrospective sentence. An example is    διε␸θαρησαν.  τουτω   τω 3. κα ι ουτοι µ εν   µορω [and these, then, were killed in this way.] or [and this is how these were killed.] (5.21.1) Here the demonstrative refers back to the unfolding of the action itself in the preceding narrative. “In this way” means “as it has been narrated.”15 14. On the terminology I use here to distinguish different types of introductory and concluding statements, see Munson 1983, 28. 15. I use the terms retrospective and prospective in a more restricted way than does Van Groningen (1958, 42– 43; see also Beck 1971, 7– 10). All else being equal, conclusions in
which the demonstrative is adverbial, especially ουτω(ς), tend to lean back less heavily than those where the demonstrative is subject or object. Elements of summarization that do not

26

Telling Wonders

Like retrospectives, the third and last type of conclusions in Herodotus consists of a nonautonomous, backward-looking sentence whose metanarrative status is formally identifiable. I call it the programmatic conclusion because it makes reference to the narrator’s compositional plan by expressing the idea that the preceding narrative has been narrated and ends at this point. It may or may not include the appearance of the grammatical first person referring to the narrator. Examples follow.   νυν περι  πεπαυµαι.  4. Ροδωπιος µεν [I am through [talking] about Rhodopis] (2.135.6)  των τοσαυτα  αναθηµα

 ε ι ρησθω  5. κα ι περ ι µ εν [And about offerings let this much be said] (1.92.4) Among opening statements, programmatic introductions incorporate a reference to an act of narration that is about to occur. Herodotus’ introduction to his description of Assyrian boats (1.194.1, discussed in the introduction) belongs to this type, 6. But the greatest wonder of all for me . . . I am going to describe [ε#ρχοµαι ␸ρα σων]. The introductory counterpart of retrospective conclusions are prospective sentences, where the primary element of summarization is similarly represented or accompanied by a deictic that points to the narrative or narrative segment that the statement identifies as a unit. In a prospective sentence, the deictic is a forward-looking demonstrative implicitly signify
or τοιοσδε, ing “as it will be narrated” (it is usually a form of οδε but 16  ουτος is also found). An example is 

ι σι ο
ιδε κατεστεασι.  7. νοµοι δ ε αυτο [Their customs are the following.] (1.196.1) bear the main emphasis of the sentence can at any rate be replaced or accompanied by a backward-looking demonstrative pronoun without the conclusion being necessarily “retro  νυν δεδεµενοι  spective” according to my definition. E.g., at 5.72.4, ουτοι µεν ε τελε υτησαν, translated “these [i.e., the Cilonians, just mentioned, then, died in chains” (not “these then were the men who died in chains,” and unlike “these were the only peoples in the cavalry”), is a plain summary conclusion.
16. The prospective value of the adverb ουτω(ς) is sometimes weaker than that of deictic pronouns used as subject or object, in which case the introduction is almost the equivalent of a summary (see, e.g., 1.7.1). A prospective, however, is never as weak as the weakest retrospectives. See n. 15 in the present chapter.

Narrative and Metanarrative

27

Finally, unlike prospective sentences and programmatic introductions, the plain summary introduction does not formally look forward to anything. It consists of a statement grammatically and logically autonomous from the report that follows. If taken out of context, it gives no indication of its introductory function. For example, the sentence 8. There are many other offerings of Croesus in Greece beside those mentioned (1.92.1) happens to represent the heading for a subsequent discussion of specific items. However, the very similar sentence at 1.183.3 (“there are also many private offerings”) does not. Plain summary introductions to narratives may be, in other words, formally identical to summary narratives.17 In fact, another way to analyze summary introductions, especially when the narrative segment they identify is short and connected with γαρ, is to regard the summary introductions as being the narrative and take the following segment as an explanatory gloss that provides further details.18 What identifies a sentence as a summary introduction is the fact that it is more abstract and “processed” than what follows; for example, it may contain broad categorizations or other interpretive elements (see the word prodigy in statement I of passages 1a and 1b quoted earlier). In undecidable cases, the only principle that matters is that when the text contains more than one statement of the type “X happened” in reference to something that happens once in the story, the excess of discourse constitutes a metanarrative phenomenon. The Rhetorical Value of Introductions and Conclusions All introductory and concluding statements in the Histories either can be assigned to one of the three basic types I have described for each or consist of a mixture or series of these. They provide “reading” directions at least by virtue of the fact that they intervene at a certain point to summarize the narrative in a certain way. Statement 2 quoted earlier, 17. With the term summary narrative, I am adapting the concept of summary that Genette (1980, 35– 85, especially 40) develops in reference to novelistic narrative and that Richardson (1990, 35, especially 31– 35) applies to Homeric narrative. My metanarrative summary statements (introductions and conclusions) have much in common with Richardson’s “appositive summaries,” forward- or backward-looking. 18. See, e.g., the narrative at 7.125, analyzed in chapter 4, “Wondering Why.” On explanatory glosses, see “Referential Glosses” later in the present chapter.

28

Telling Wonders

“The Lydians, then, were definitively enslaved by the Persians,” contains no additional glosses but fulfills a function of gloss by bringing out the meaning of the narrative according to the monarchical code. The verb  “enslave” (used metaphorically) is a particularly strong term in δουλοω, this code. It has appeared only once so far in the Histories (1.27.4) but becomes more frequent in subsequent narratives of conquest, especially Persian.19 If we think in terms of “performance,” the sentence seems to require a moment of silence as it underlines a milestone in the story and a major break in the narrative. Following upon the Lydian ethnography and at some distance from the preceding historical account, it concludes the entire Lydian narrative by reminding the audience that the actions of rulers affect communities. Croesus’ defeat by the Persians has caused the “enslavement” of an ethnos whose contributions to culture and initial resourcefulness have just been described (1.94.1– 7). If narrative always entails interpretation, a preliminary summary or a restatement of part of the narrative represents an additional opportunity to interpret, whether by attributing the narrated event to a general class, by privileging a single moment or feature, or by referring to one or more of the cultural codes according to which the narrative can be read.20 Moreover, introductions and conclusions, including those that seem expendable from the point of view of what they actually say, scan and pause the narrative, endow it with a certain rhythm and tone, and perform a “phatic” function vis-a-vis ` the audience.21 Some retrospectives, for example, are equivalent to mere verbalized punctuation marks that leave the listener time to react. This is especially true when both the primary and the secondary elements of summarization are replaced with  19. Similarly, the word δουλος appears in the sense of “subject” beginning with the rule of Cyrus (1.89.1). For cultural codes and the monarchical code in particular, see my introduction, text and n. 28. 20. Cf. Prince’s definition of metanarrative quoted in n. 11 in this chapter. The introduction (1.7.1) and conclusion (1.14.1) framing the Gyges-Candaules episode, e.g., bring out the political code in a story of love, betrayal, and revenge by emphasizing the resulting change of dynasty. They therefore draw attention to the interface between the public and the private spheres in the actions of individuals in power, a major tenet of Herodotean thought. 21. Of six constitutive functions of language that Jacobson (1960, 353– 56) distinguishes in any speech event, the “phatic” focuses on the contact between speaker and addressee through messages “primarily serving to establish, to prolong, or to discontinue the communication, to check whether the channel works (‘Hello, do you hear me?’), to attract the attention of the interlocutor or to confirm his continued attention (‘Are you listening?’ or in Shakespearean diction, ‘Lend me your ears! . . .’).”

Narrative and Metanarrative

29

backward-looking demonstratives. At the end of a riveting narrative of how Alexander of Macedon had an entire Persian delegation killed over dinner and managed to cover up the murder, we find,  µεν  νυν ουτω
 9. ταυτα κ η ε γενετο. [these things, then, happened approximately in this way.] (5.22.2) So, this is it. Period, end of story. What do you know? Prospective sentences, which constitute an implicit promise of something to come, may be almost as inexpressive in their substance as the retrospective sentence I have quoted. When a prospective sentence mentions a set of facts, it serves as the chapter heading for the narrative of those facts (as in “Their [i.e., the Babylonians’] customs are the following” in statement 7). Prospective sentences, however, also point at short range to an individual element about which they anticipate very little information. Thus, forward pointers of the type “he did/devised the following thing” frequently appear to create a tiny moment of suspense before the narrative of clever, unexpected, or outrageous actions. No matter how colorless, a prospective sentence stimulates the recipient to formulate a question (“The following thing happened.” What happened? “He saw the following dream.” What dream? “They sacrifice in the following way.” What way?), and marking a pause before the beginning of the narrative, it signals, “Listen! and you will know.”22 The narrative has then the chance for a new start.23 While the deictic allows the prospective to withhold information momentarily at the same time as it announces a topic, summary introductions must anticipate a complete thought about a topic before the narrative 22. Each logical pause in a narrative implies a question about what will come next, and forward-looking introductions provide the terms in which the narrator wants the recipient to ask the question. In one case (3.6.2), an introductory open-ended question encoding the recipient explicitly formulates a riddle that the recipient may not have thought about and is in turn followed by a programmatic introduction: “Where on earth, one might ask, are the empty jars used? I will explain this too.” All questions in the text that are not in speeches belong to the metanarrative and encode the addressee, but different types fulfill different roles: celebratory (7.21.1), interpretive (1.75.6), or introductory, as here (cf. Iliad 1.8 and 5.703; see Richardson 1990, 179; de Jong 1987, 49). See Lang 1984, 38– 41; Lateiner 1989, 72– 73. For a more detailed discussion of Herodotus’ use prospectives, see Munson 1993a. 23. Examples are “Regarding Croesus himself, this is what happened. He had a son” (1.85.1), and “Smerdis was unmasked in the following way. Otanes was the son of Prexaspes . . .” (3.68.1).

30

Telling Wonders

begins, often with γα ρ, to substantiate and expand on the summary. For this reason, summary introductions tend to be a more suitable tool for interpretation and evaluation than are “strong” prospectives (i.e., prospectives that lean heavily forward because the deictic fulfills most of the summarizing function). See, for example, the introduction to the narrative of the death of Croesus’ son. 10. After Solon left, a great nemesis from god overtook Croesus, one can imagine, because he thought he was the happiest of all men. (1.34.1) To call what happened to Croesus a “great nemesis” and to make a statement on the causality of the event on that basis is a striking interpre ε ι κα σαι] tive maneuver, and the self-referential “one can imagine” [ως identifies it as such. In programmatic introductions and conclusions, the summary that announces the topic of the following narrative or restates some aspect of the preceding narrative is by definition joined to a self-referential gloss of narration, by which Herodotus comes into the open as the one who is ultimately verbalizing the story and putting it together. The narrator displays his control over his material simply by announcing what he is going to say, by cutting a story short, by explaining his reasons for narrating or not narrating something, or by expressing which criteria govern his whole work. Not all programmatic statements are showstoppers. Nevertheless, the introductions or conclusions that bear the greatest rhetorical force and are most expressive about the substance and point of the narrative tend to include a programmatic element. Thus, Herodotus’ first sentence is a mixed summary-prospective-programmatic introduction; it identifies the entire work as a single, though diversified, narrative. This statement signals at the outset the tensions and complications of the Histories themselves, torn between unity and dispersion, fact and meaning, diachrony and synchrony, syntaxis and parataxis.    µητε 

οδεξις
 Ηροδοτου Αλικαρνησσεος ι στορι ης απ ηδε, ως τα    ε ξι τηλα γενηται,

   γενοµενα ε ξ ανθρ ωπων τω   χρονω µητε ε#ργα 
Ελλησι, τα δ ε βαρβαροισι µεγα λα τε κα ι θωµαστα , τα µ εν



 # + α ι τ ι ην ε ποαποδεχθ εντα, ακλε α γενηται, τα τε αλλα κα ι δι ην 

ηλοισι.  λεµησαν αλλ

Narrative and Metanarrative

31

[This is the demonstration of the research of Herodotus of Halicarnassus, lest the events of men may become evanescent with time, or great and wonderful actions, some performed by Greeks, some by non-Greeks, may become inglorious, both the other things and also for what cause they came to war with one another.] The words summarizing the narrative generically and at long range are represented by the subjects of the double purpose clause that gives the reason for narrating. “Great and wonderful actions, some performed by  Greeks, some by non-Greeks” [ε#ργα µεγαλα τε κα ι θωµαστα , τα µ εν   


Ελλησι, τα δ ε βαρβαροισι αποδεχθεντα] are the terms of the narrator’s code of celebration, here advertising the narratability of all the “events of 

 men” [τα γενοµενα ε ξ ανθρ ωπων] treated in the Histories, be they historical or ethnographic. A more specific summary is provided by the final # colon of ambiguous grammatical status, where anticipatory αλλος in τα τε # αλλα κα ι [both the other things and also] serves as a bridge between the + α ι τι ην ε πολεµησαν  broader subject matter and the narrower topic δι ην

ηλοισι  αλλ [for what cause they came to war with one another].24 The word α ι τι η means “cause,” “grievance” (of an offended party), or “responsibility” (of an offender). As a historical, juridical, and ethical term, it anticipates the combination of these three codes in the immediately following narratives about the remote origin of the East-West conflict and, at long range, in the narratives of many other wars both between Greeks and nonGreeks and between members of each group.25 The first sentence also notably names the real author, Herodotus. This person will henceforth become the “I” of the text, the histor. He will appear again in his historical dimension, as he does here, in a particular type of metanarrative statement that I call glosses of historie, that is, 24. The best analysis of the first sentence is by Erbse (1956, especially 217– 19 for this point). Cf. also Krischer 1965; Drexler 1972, 6– 9; Moles 1993, 92– 94. Initial sentences of # other fifth-century prose works are discussed by Fowler (1996, 69– 70). Anticipatory αλλος (as opposed to “additional” as in statement 8 earlier in the present chapter) has the function of narrowing the focus of the narrative that follows by postponing (or excluding) certain aspects of the topic being treated. Wood (1972, 14) rightly regards it as one of the signs of “a perspective . . . which views discrete events as parts of a whole, which sees always the meaning of events.” 25. See α ι τι ους at 1.1, where it means “those at fault,” “those responsible,” and  . . . υπα   ρξαντα αδ

ι κων therefore also the “causes” and is eventually picked up by τον ε#ργων [the one who initiated unjust actions] at 1.5.3. See Pagel 1927, 8; Immerwahr 1956, 245. As Payen argues (1997, 88– 91), the narrative of the Histories imposes a broad interpretation of the phrase “they came to war with one another.”

32

Telling Wonders

references to Herodotus’ fact-finding process.26 The gloss of historie in  

οδεξις.

οδεξις this sentence is in the expression ι στορι ης απ The word απ constitutes the programmatic element (or gloss of narration) of the sentence. It means “performance” in the sense that the product of historie becomes a display of narrative; but it also points to the disclosure of the preliminary discovery process, in a sort of contrapuntal narrative of the actions and reasonings of the histor as researcher—where he went, where he learned things, how he put together what he saw and heard, and so

on.27 But another type of gloss is embedded here: since the root αποδεκin this same sentence also advertises the narrated (“great and wonderful 



οδεξις deeds performed” [αποδεχθ εντα]), the term απ is attracted into the celebratory code as well. The narrator, Herodotus, in other words, presents his own speech act as a performance in some way comparable to the great and wonderful performances of the characters of his narrative.28 Self-Referential Glosses As my preceding analysis of the first sentence of Herodotus’ Histories shows, an introductory or concluding statement is a gloss for the way in which it summarizes the narrative, but at the same time, it may also contain a number of other glosses. There are two major categories of glosses, selfreferential and referential. A self-referential gloss belongs to the level of metanarrative that is most distant from the story level, because it defines and qualifies—“talks about”—another piece of the text (i.e., a narrative, part of it, or another gloss) as a verbal product, rather than focalizing through the narrator a referent in the world of the narrated. Glosses of narration. We have already encountered glosses of narration: they identify a narrative, a portion of a narrative, or another metanarrative statement as something that the narrator, Herodotus, has said or will say, or they postpone or recall narratives (“as I will narrate later,” “as has been said before”). In the negative form, glosses of narration describe the text by default; they state that it lacks certain features by decision of the narrator. 26. See Dewald 1999, 232, 236. 27. See Dewald 1987, 167; Marincola 1987, 127.  28. See Erbse 1956, 209– 11; Nagy 1990, 218– 21 with nn. 24 and 35. Αποδεξις is

also a part of a semic code and connotes wisdom because another use of αποδε ι κνυµαι has to do with the display of advice, opinion, or sophie (cf., e.g., 2.146.1, 7.139.1, 8.8.3: 

 γνωµην αποδ εξασθαι, applied to the narrator).

Narrative and Metanarrative

33

Herodotus identifies the unfinished product of his narration as his  λογος. He also applies this word to some of his individual narratives or parts of the whole.29 Herodotus denotes his act of narration by both verbs of speaking and verbs of writing; he thereby shows his adherence to the two modes and his double intention to communicate in the present and to create a lasting record.30 Several of Herodotus’ terms for narrating participate in the metaphor of a journey to the places that will become part of the narrated.31 The narrator as such is a traveler who may or may not decide to embark on the roads of other people’s logoi.32 The itinerary is his work in progress, his logos, which “seeks” or “wants” (verb ε πιδι ζοµαι or δι ζοµαι) more or less predictable destinations or topics.33 This metaphor of the journey creates the illusion of an overlap between the code of narration and the code of historie (where verbs of motion are in most cases used literally). Thus it reinforces our sense of the identity between the narrator and the researcher.34 29. See, e.g., 1.5.3, 1.140.3, 1.184. See also Payen 1997, 63– 66. 30. Nagy (1990, 219) writes, “Saying something is in the case of Herodotus the equivalent of writing something because it is ultimately being written down in the Histories.” Cf. Hartog 1988, 276– 89. According to Immerwahr (1966, 15), γρα␸ω emphasizes exactitude (1.95.1, 2.70.1, 2.123.1) or the notoriety of a person or event (2.123.3, 7.214.3). Cf. Hartog 1988, 285. Herodotus’ vocabulary of narration is vast. Verbs with the root µνη (ε πιµνησθεω, 2.3.2; etc.) may suggest the idea of memorializing through words (cf. the memorial of words at 7.226.2; see Immerwahr 1960, 267). Applicable to both oral or written speech are verbs of mentioning, indicating, explaining, or displaying, including the particularly fluid σηµαι νω (on which see Hartog 1988, 366; Nagy 1990, 165). Other  performative verbs of narrating include “go on at length” (µηκυνω, 2.35.1, 3.60.1), “use”  in the sense of mentioning information he knows (χρωµαι, 8.85.2), “refrain from saying”  

ερχοµαι,  (ε πεχω, 7.139.1), “stop” (παυοµαι, 2.135.6), “revert” (αν 1.140.3, 7.239.1),

ι ηµι, 3.95.2), “forget about” (ε πιλανθανοµαι, 4.43.7). “omit” (απ  31. The act of narrating is often expressed with verbs of going: e.g., ε#ρχοµαι ε ρεων # λεξων  (1.5.3, 2.99.1), ηια (4.82). At 1.5.3, the narrator will proceed forward in the logos     (προβησοµαι ε ς το προσω του λογου), going through (ε πεξι ων) cities great and small. Bidding goodbye to a subject is equivalent to leaving a place (see 2.117, 4.96.2). 32. At 2.20.1 (where metaphorical and literal terms of the code of narration mix in a striking way), the narrator states that some Greeks wanting to be distinguished (or perhaps  “visible”: ε πι σηµοι) for their cleverness, “said three roads [οδους]” (cf. 2.22.1), two of which he does not deem worth mentioning, except that he wants only to point to them   about Cyrus (σηµηναι). At 1.95.1, he knows three other “roads of logoi” [λογων οδους] besides the logos according to which he will write. For the journey metaphor applied in the narrative to the speeches of characters, see 1.117.2, 2.115.3. 33. See 1.95.1, 4.30.1. The verb δι ζοµαι is often found in the narrative in conjunction with verbs of motion, as when Heracles goes all over Scythia looking for his mares (see 4.9.1). 34. On Herodotus as alter ego of the Homeric Odysseus, see Nagy 1990; Marincola 1997, 1– 3.

34

Telling Wonders

Narrating is only one of the activities the text attributes to the authorial first person, but one in which Herodotus displays unparalleled selfassurance.35 Occasional apologies compensate for his expressions of control over what he narrates,36 and these phrases sometimes take the form of an appeal to the narrator’s obligation or lack of obligation to mention something or to include certain types of material. The verb δει appears frequently in this context and bears little emphasis (“And yes, I must still explain . . . where was the dirt from the ditch utilized,” 1.179.1). It suggests “a sense of strain inherent in the problem of composition,”37 the need to be detailed and exact, perhaps the fear to go on for too long and appear trivial. Other times, we find stronger phrases expressing a graver commitment to the integrity of his logos. The logos that seeks/wants (his logos, always with a dative of possession) is his narrative agenda. This is never sharply defined for the recipient, but it imposes on the narrator a duty that overrides personal inclinations, other moral issues, or external pressures.38 Glosses of source. Complementary to glosses of narration are glosses of source, which identify a narrative or statement as originating from other narrators outside the narrative (“so and so say” or “it is said”).39 The logoi of others have the important function of giving a voice to the other, who challenges the subjectivity of the audience.40 At the same time, 35. See Dewald 1987, 164. The problems connected with the first-person plural are discussed by Chamberlain, forthcoming. 36. At 2.45.3, the narrator exceptionally apologizes to the gods and heroes, as if they were a special part of his audience. 37. Dewald 1987, 165.

38. On the narrator’s obligation with αναγκ-, see 7.139.1, 2.3.2, 2.65.2, 7.99.1. Cf.  οκειται),  2.123.1 (ε µο ι . . . υπ 7.152.3 (ο␸ει λω, ου . . . ο␸ει λω). The idea of obligation is discussed further in chap. 4. 39. See Dewald 1987, 153; Jacoby 1913, 398– 99; Darbo-Peschanski 1987, 91– 97. Sometimes sources do more than “say”: they demonstrate (see, e.g., 5.45.1, 5.45.2), know exactly, calculate (see 2.145.3), agree or disagree with one another (see, e.g., 1.23.1), swear (see 4.105.2), accuse (see 6.14.1), have nothing to say (see 3.111.1), and so on. Though Herodotus refers to the provenance of his evidence throughout his work, he does so irregularly, and many evidently received narratives are not marked by “they say” or any such gloss or slide from the second to the first narrative level of the narrator’s own voice. See Darbo-Peschanski 1987, 113– 18, 124– 25; Hartog 1988, 291– 94. Dewald (1999) discusses the problems of the focalization, and therefore the status, of unattributed narratives. 40. Glosses of sources identifying foreign logoi resemble cognitive statements of the type “the Persians say.” These, however, record a people’s cultural beliefs in an ethnographic context and are therefore just as much a part of the narrative as are statements of the type “the Persians do.” An example is “The Taurians say that the divinity to whom they sacrifice is Iphigeneia, the daughter of Agamemnon” (4.103.2).

Narrative and Metanarrative

35

the logoi reported in the Histories also expose the self-interest or subjectivity of the speakers themselves.41 More often than not, they provide unreliable evidence, contradict one another, and lead to contention. If Herodotus’ first sentence puts the narrator in charge, his second sentence (“The Persian logioi say that the Phoenicians were the ones responsible for the conflict,” 1.1.1) introduces a verbal quarrel that remains unresolved.42 This is only the first of many quarrels in the metanarrative of the text that echo, reflect, and metaphorically represent the disputes and struggles among the characters in the narrative and in the real world. Glosses of historie. When a source “says” something in the past tense and directly to the researcher, the gloss of source is colliminal with a gloss of historie, which identifies a part of the text as containing the results of Herodotus’ inquiry.43 No longer a passive recipient of information, the histor travels, “wants to know,” participates in interviews, collects hearsay, verifies by autopsy. The more extended passages that place him at the scene are small narratives in their own right, though at a different level, and represent the closest thing in Herodotus to what modern ethnographers call “personal narrative.”44 To put it in literary terms, the outside narrator enters the narrative and almost becomes a character.45 Evaluations of accuracy, glosses of evidence, knowledge, and ignorance. Other than underlining the provenance of a statement or a narrative in the text, self-referential signs give indications about their reliability. In a text mostly composed of the received logoi of others, these are important directions. Evaluations of accuracy46 corroborate or decline to 41. See Dewald 1987, 168. 42. See Dewald 1999, especially 236. For an inventory of Herodotean alternative versions, see Lateiner 1989, 104– 8. The interface between metanarrative and narrative quarrels is discussed in chap. 3, “Disputes, Arbitration, and the Subjectivity of Opinions.” 43. Jacoby 1913, 247– 76, 395– 400; Macan 1895, 1:lxxxi– lxxxii; How and Wells 1928, 1:16– 20; Marincola 1987; Dewald 1987, 155– 59; Hartog 1988, 261– 73; DarboPeschanski 1987, 84– 87. 44. Pratt 1986; Geertz 1988. 45. See Marincola 1987, especially 127– 28. Marincola gives a complete survey of Herodotus’ statements of inquiry and shows that the extended type only occurs in book 2, where the statements represent implicit polemics against previous accounts. In these passages, we find verbs of setting forth, traveling, going, arriving, sailing (see ε#πλευσα at 2.44.1), being in  someone’s company (συνεγενοµην, 3.55.2), and so on. Occasionally the histor’s sources become part of the narrative as well, as do the Egyptian priests in the famous piromis scene (2.143) and the disingenuous scribe of the temple of Athena on Sais (2.28.2). 46. Accuracy is not a very satisfactory term here, but I am reluctant to use truth, a word that Herodotus himself avoids in most contexts. See chap. 3, “Specific Glosses of Interpretation.”

36

Telling Wonders

corroborate, refute or reject tout court.47 Herodotus guarantees his own statements with glosses of evidence (“it is evident that such and such is the case”)48 or of knowledge.49 When he says he does not know (or he “cannot tell exactly”), he is either unable to corroborate someone else’s logos or acknowledging the incompleteness of his own.50 Glosses of opinion. Among the most pervasive self-referential metanarrative signs are those that identify a referential gloss or a statement of fact (e.g., the number of troops at 7.184.1) as representing the result of Herodotus’ mental activity—estimate, reasoning, conjecture, judgment, and so on. To these I apply the blanket term opinion.51 Glosses of opinion both weaken and enhance the authority of the text. When they accompany a referential gloss that proclaims how great the significance of a fact is (celebration), discloses what its significance is (interpretation), or evaluates its worth, they underline Herodotus’ own subjectivity, separate from that of other voices in and outside the text.52 They constitute the most forceful markers of Herodotus’ own ideological and philosophical position. For example, 47. Rejections qualify received information as not trustworthy. A refutation is an explanatory gloss attached to a rejection (called an ε#λεγχος at 2.23; cf. 2.22.4). See, e.g., 3.45.3. 48. When Herodotus is confronted with controversial issues, the gloss of evidence often introduces a referential gloss that explains the evidentiary basis for something (see, e.g., 5.22) See Darbo-Peschanski 1987, 131– 47; Thomas 1997 and 2000, 168– 200. 49. With such generalized expressions as “I learn” [πυνθανοµαι] (e.g., 7.114.2), the category of glosses of historie overlaps with that of glosses of knowledge, which emphasize the results rather than the process of the research. An example occurs at 1.20: “I know, having learned it from the Delphians.” See Lang 1984, 11– 17. Glosses of knowledge   include the limitative phrase “the first [or greatest, etc.] we know about” (πρωτος των  ις #ιδµεν, etc.). See Shimron 1973; Darbo-Peschanski 1987, 113; Hartog 1988, 289– 90. ηµε 50. See, e.g., 4.25.1 (no one knows), 4.180.4, 6.14.1. See the lists “Ignorance Universal” and “Certainty Impossible” in Lateiner 1989, 69– 71. See also Flory 1987, 49. 51. In the programmatic statement with gloss of historie at 2.99.1, Herodotus mentions  the role of his own γνωµη, “judgment,” in processing the data of eyewitness and oral report. Besides verbs of seeming, thinking, and conjecturing, Herodotus’ vocabulary of opinion also  (3.111.1, 3.38.2, 4.195.4, 5.97.2, 7.167.1, includes expressions with ο-ικα and ο ι κ ως 7.239.2), the dative of reference µοι (see passage 17 later in the present chapter), and certain first-person verbs in self-referential or referential glosses (e.g., πει θοµαι, “I am persuaded”;  “I laugh”; α ι νεω,  “I praise”: see passage 15 in the present chapter). See Beck 1971, γελω, 70– 72; Dewald 1987, 162; Lateiner 1989, 98; Hohti 1977; Darbo-Peschanski 1987, 164– 89, especially 184. 52. Darbo-Peschanski (1987, 186) writes ‘Lorsque l’enquˆeteur consent a` e´ valuer ses propres discours comme ceux de ses informateurs . . . il donne . . . a` son jugement le caract`ere relatif d’un avis susc´eptible d’ˆetre discut´e.” See also Dewald forthcoming a.

Narrative and Metanarrative

37

  η µετερην]  11. In my judgment [κατα γνωµην τ ην the following is the wisest custom, which I learn that the Illyrian Eneti also have. (1.196.1) 12. . . . since I believe [νοµι ζων] that all men know equally about the gods . . . (2.3.2) To the extent, however, that opinion identifies information that may not be accurate, it is a compromise that replaces the vacuum of “being unable to tell” but falls short of “knowing.”53 Various glosses encoding the addressee. What the narrator Herodotus knows, corroborates, and believes (in the sense of nomizein) provides reference points for the understanding of the world he wants to communicate. Glosses that express his ignorance or uncertainty are the explicit marks of an interrogative text. They put the recipient of the narrative in charge. We perceive the presence of the recipient in the text especially through the mediation of self-referential metanarrative. He is the anony
mous οστις who is asked by a gloss of noncorroboration at 2.123.1 to “use” the reported logoi if he finds them credible, and he is the τις who is invited to choose between different versions at 5.45.2. This “whoever” or “someone” is also enlisted in glosses of historie as a potential histor, traveler, observer, or eyewitness (“it is evident even for one who has not heard about this but just sees, at least if he has some intelligence . . . ,” 2.5.1).54 Direct addresses in the second-person singular coopt the listener to the inquiry.55 Yet the narrator also expresses some doubt that his audience will be up to the task with which he entrusts them. When glosses of narration explicitly refer to their likely reaction to what is or could be narrated, they 53. I am paraphrasing Romm (1989, 100 n. 12). The nonauthoritative aspect of Herodotus’ opinion is emphasized by Darbo-Peschanski (1987, 146– 47, 184– 89). 54. The attributive participle also encodes the recipient in this capacity, e.g., at 2.135.3:   ] can see . . .”; cf. 1.105.4, 2.31. In some cases, mostly “anyone who wants [τω   βουλοµενω in referential metanarrative (2.154.4, 3.116.3, etc), the first-person plural encodes the audience because it means “we the Greeks.” On the role of the narratee, see Prince 1973; 1982, 16– 24; 1987, s.v. “narratee.”  55. An example occurs at 1.139: “if you look into this, you will find . . .” [ε ς τουτο  ησεις   διζηµενος ευρ . . .); see also 3.6.2 (the recipient as interviewer); 1.199.4, 4.28.1, 3.12.1 (the recipient as prospective or hypothetical experimenter); 2.5.2, 2.29.5, 2.30.1, 2.97.2 (the recipient as prospective traveler). See Dewald 1987, 155; 1990, 220. De Jong (1998) notices the similarity with the use of the second person in the Hippocratics (e.g., Airs, Waters, Places 8.10).

38

Telling Wonders

especially mention disbelief.56 One famous passage (7.139.1), which I examine later (chapter 3.2), attributes to them hostility. Referential Glosses A referential gloss provides directions on how to receive the narrative by commenting not on the narrative itself but on the narrated. Such glosses often represent the propositional content of glosses of the self-referential group—in other words, they correspond to item Y in a statement of the type “X is evident because Y” or “it is my opinion that Y.” But referential metanarrative can stand on its own. Referential glosses constitute the level of metanarrative that is closest to the narrative. For this reason, they fulfill their function indirectly and often in a subtler way than statements that identify a piece of text as coming from a certain source, as being Herodotus’ opinion, or as representing—or not—an accurate report. Consider, for example, what I call glosses of testimony. These consist of references to poetic and other written testimony of narrated events or to tangible vestiges of the past that are generally well known, verifiable by Herodotus’ contemporary audience, or allegedly verified by the narrator/researcher. Glosses of testimony sometimes appear to be the referential content of an implicit gloss of corroboration or evidence. An example is the item y in “X is evident/ proven/accurate because Y,” whatever X in the narrative may be in the particular case. Thus, the Spartans were defeated by the Arcadians (X), and their chains were visible “still in my time” in the temple of Tegea (Y): here Y memorializes event x and confirms its gravity (1.66.4).57 In many cases, however, what notations of this sort contribute to Herodotus’ account, what they confirm or go to prove, is not entirely transparent. At 5.77.3, for example, we encounter another mention of chains: the chains of Chalcidians and Boeotians, crushingly defeated by the Athenians in 56. Glosses of narration automatically encode a recipient of the narrative—“I say/ narrate/write (for the benefit of someone)”—but they are also a type of gloss where the recipient is likely to be explicitly mentioned. At 1.193.4, the disbelief of the audience motivates the narrator’s negative program. Disbelief is attributed to “some Greeks” (i.e., some of the audience) in a gloss of narration at 6.43.3 and in an implicit gloss of corroboration at 3.80.1. See Hartog 1988, 289– 90; Packman 1991, especially 406. Other glosses of narration refer to the audience’s cultural knowledge as the basis for narrating or explaining something: see 3.37.2, 3.103, 4.81.4, 4.99.5. 57. On these glosses providing evidence for the “greatness” of an event, see Immerwahr 1966, 269.

Narrative and Metanarrative

39

their first battle after the liberation of Athens from the tyrannical regime, still hang at the Athenian Acropolis on the wall “half-burned by the Medes.” This notice juxtaposes the distant past, the more recent past, and the present of narration in an allusive way that requires decoding. It goes beyond a testimonial function in the most obvious sense.58 Explanatory glosses. Questions about the function of metanarrative intrusions sometimes emerge within the broad referential category of explanatory glosses (to which glosses of testimony also sometimes belong). These provide new factual information apparently designed to clarify some element of the context in which they occur, but their explanatory value is not always clear. At the end of the narrative of the murder of the Persian ambassadors at the hands of Alexander of Macedonia during a banquet, for example, we find a gloss (itself emphasized by selfreferential glosses of knowledge and source and generating a gloss of evidence with its explanation) stating that the kings of Macedonia are Greek (5.22). Is this information designed for the sake of apologia or irony? Or has the narrator simply taken the opportunity to insert information that will be useful later on?59 Explanatory glosses in general occupy an intermediate position between the metanarrative and the narrative, because along the main narrative line that proceeds in chronological order from Croesus to Xerxes, insertions marking a change of time always more or less begin as explanatory glosses. Thus, the long flashback on Cyrus’ antecedents is introduced by the summarizing statement “who was this Cyrus who conquered Croesus” (1.95.1). It interrupts the chronological narrative, in other words, like the delayed and much expanded gloss of identification that typically accompanies the entrance of a character in the history to give a few facts about his family, his position, and his accomplishments.60 Similarly, even the most structurally autonomous ethnographic description can be regarded as a gloss explaining the background of a people that has played a role in the narrative or is about to do so. In comparison with inserted semiautonomous narratives, the metanarrative status of an 58. For other glosses of testimony, see, e.g., 1.12.2, 1.24.8, 2.123.3, 2.131.3, 4.11.4, 4.12.1, 4.166.2, 5.58.3, 6.14.3, 7.167.2, 7.178.2. At 3.38.4, the Pindaric quotation is used somewhat differently, to corroborate a gloss of interpretation. 59. See Badian 1994, especially 121, on the ambivalence of Herodotus’ whole Macedonian narrative. 60. See, e.g., 1.6, 1.23, 5.32, 6.35.1, 6.131, 8.79.1 (cited as passage 16 in the present chapter). Cf. the discussion of Homeric character introductions in Richardson 1990, 36– 51.

40

Telling Wonders

explanatory gloss is based on its brevity. Consequently, it is, or we expect it to be, immediately functional as a short-range explanation and not autonomous from the narrative to which it is attached. An isolated ethnographic gloss within the Gyges-Candaules episode, for example, is clearly designed to help a Greek audience to evaluate Candaules’ behavior in displaying his wife naked. 13. Among the Lydians, as also among almost all other nonGreeks, to be seen naked even for a man brings great shame. (1.10.3) While historical glosses bring out the historical code in an ethnographic description (e.g., by explaining the origin of certain customs or monuments), ethnographic glosses underline the “code of customs” in the history. Both testify to the mutual interdependence of the two genres in Herodotus’ work. Sometimes, we need to know about culture to understand history, or we need to know about historical occurrences to understand culture. But again, specific cases raise different issues: not only whether and how an ethnographic gloss actually contributes to the context where it appears, but also whether it is complementary or contradictory in relation to the ethnographic information about the same people given elsewhere, especially in autonomous ethnographic narratives; or how these isolated bits of ethnographic knowledge work as a group when they can be classified cross-culturally, according to the areas of culture they discuss.61 Glosses of comparison. An explanatory intrusion in the text also occurs every time the narrator brings into the narrative an extraneous referent for the purpose of affirming its resemblance to or difference from something in the story. The resulting gloss of comparison often explains an unfamiliar phenomenon through one that is, from the point of view of the audience, more familiar. But this is clearly not the case, for example, with the numerous comparisons proclaiming the similarity or equivalence among customs and beliefs of different peoples (see the earlier examples 11 and 13). The function that these and many other comparative glosses perform individually and the cumulative effect of 61. See, e.g., the collections of ethnographic glosses concerning oath taking or purification rituals around the world (1.74.1, 3.8.1) and the metalinguistic glosses (translations of terms) scattered along the narrative. On the latter group, see Hartog 1988, 237– 48; Harrison 1998; Chamberlain 1999.

Narrative and Metanarrative

41

explicit comparison in the text are, as I shall show in the next chapter, more profoundly interpretive. Glosses of interpretation. Interpretation is explanation at a higher conceptual level. Interpretive glosses occur when the voice of the narrator comments on some of the more covert or questionable aspects of the narrated. Herodotus’ decoding of dreams, omens, symbolic objects, oracles, logoi, and other utterances belongs at the more explicit end of the explanatory/interpretive spectrum. At the other end, interpretation is communicated by the summarizing elements of introductory and concluding statements and, closer still to the edge of “pure narrative,” even by the words and codes with which the text verbalizes the story in the narrative itself. Between these two poles, we find a number of statements that discuss, more or less conspicuously or problematically, why an event is important; its value, meaning, or origin; and its less obvious motive or result. In most cases, interpretive glosses do not add new facts, as pure and simple explanations tend to do, but rather process those given in the narrative. The operation is often made more visible by a gloss of opinion or by some other self-referential gloss—for example, one of knowledge or evidence. An example is 14. And it is clear to me [δηλοι τε µοι] that the whole situation on the barbarian side depended on the Persians, since also these fled even before they engaged with the enemy, just because they saw the Persians flee. (9.68) Other metanarrative signs that alert us to the presence of an interpretive gloss are a generalized form of discourse (e.g., “all men know equally about the gods” in example 12), certain types of negations and questions, hypothetical constructions, particles signifying “perhaps” or “somehow” (especially κως), or a listing of alternative choices about what may have happened or why.62 62. For multiple choices, see, e.g., 1.86.2 (motives of Cyrus), 7.54.3 (motives of Xerxes), 7.239.2 (motives of Demaratus), 8.87.3 (motives of Artemisia). For an interpretive negation, see, e.g., 6.61.1. Negative statements are always a part of the metanarrative, because what we call the story does not include nonevents. See Prince 1982, 18– 19; de Jong 1987, 61– 68, especially 67; Hornblower 1994a, 152– 53. The same is true for hypothetical constructions (see, e.g., 7.139.2– 4). Important interpretive glosses with κως and κου appear at, e.g., 3.108.2, 7.191.2, and 6.98.1. Cf. Lateiner 1989, 31– 32. Interpretive glosses in the form of a question (often in combination with a hypothetical construction) appear at

42

Telling Wonders

In the absence of any marker of interpretation whatsoever, statements that attribute undisclosed motives to characters can present special problems.63 In other cases, the perception that we are in the presence of a gloss depends first and foremost on the level at which the text processes the raw external data. No self-referential sign of interpretation marks the statement that the Athenian war against Aegina, by forcing the Athenians to build ships, proved to be the salvation of Greece at the time of the Persian invasion (7.144.2). Yet the causal connection it establishes a posteriori between two otherwise unrelated occurrences clearly reveals the mental process and deliberate intervention of the narrator. By creating a thoughtprovoking paradox, he goads the reader to search for further meaning.64 Evaluations of worth. Embedded in many narratives is a judgment that some of the actions narrated are “Bad” or “Good” on either moral or intellectual/strategic grounds or by both standards at once.65 In other cases, the narrator himself, explicitly and in his own voice, makes evaluations of worth either by using evaluative words in the course of the narrative (e.g., 9.78.1: “he uttered a most impious speech”) or by using glosses of praise or blame.66 In the following retrospective/prospective that marks the transition between two items of the Persian ethnography, the evaluative verb incorporates a self-referential gloss. “I praise” is more or less equivalent to “I record/believe that [gloss of narration and opinion] it is good [evaluation].”   νοµον,    µεν  νυν τονδε  δ ε κα ι τονδε 15. α ι ν εω τον α ι ν εω [I praise that custom [reported earlier], and I also praise the following one.] (1.137.1) 2.11.4 (twice), 2.125.7, 4.46.3, 1.75.6, 2.15.2, 2.22.2, 2.45.2, 2.45.3, 2.57.2 (the last six occur in refutations). Generalizations are discussed in chap. 3. 63. This is also due to the fact that the narrator of the Histories swings back and forth between the positions of omniscient narrator—that is, with “the privilege . . . of obtaining an inside view of another character” (Booth 1983, 160; cf. Chatman 1978, 212, 215– 16)— and researcher. In the latter stance, he distinguishes seen from unseen and marks attributions of motives by a self-referential gloss. Dewald (1987, 161 n. 161) counts twenty-two  and δοκεειν  cases in which δοκεω ε µοι have this function. 64. Interpretive glosses also include glosses of anticipation of doom, which underline the decisive role of a functional event in the plot in triggering an overdetermined negative outcome. An example occurs at 1.8.2: “After not much time, since Candaules was bound to end up badly, he said to Gyges the following.” See also 2.161.3, 5.92δ1, 6.135.3, 9.109.2, 4.79.1. For discussion of these passages, see Hohti 1975; Gould 1989, 72– 78; Munson forthcoming. 65. I borrow these deliberately vague expressions from Asad 1986. 66. See Prince 1982, 11.

Narrative and Metanarrative

43

Praise (reinforced by glosses of opinion and historie and by a pun with the proper name) is the most emphatic component in the identification of

Aristides (8.79.1). The gloss is designed to underline the element of αρετ η (moral excellence) in the narrative of Salamis, otherwise dominated by an ethically more ambiguous σο␸ ι η (cleverness) of Themistoclean stamp.67 16. When the generals were gathered together, there crossed over from Aegina Aristides, the son of Lysimachus, an Athenian who had been ostracized by the people and about whom I have come to believe, when I was inquiring about his character, that  ε γω he was the best man in Athens and the most just [τον    τροπον, 

υ τον # # νενοµικα, πυνθανοµενος αυτο αριστον ανδρα    γενεσθαι ε ν Αθηνησι κα ι δικαιοτατον]. Advertisements of narratability. In Herodotus, evaluation is intimately joined with explanation and interpretation because bad or good behavior and foolish or wise actions determine the course of history. At the same time, Herodotus’ positive evaluation of a fact in the world of the narrated sometimes also coincides with a different metanarrative function: the celebration of his subject as such. Thus, a conclusion/introduction system within the narrative about Sperthias and Boulis at the Persian court represents both an evaluation of worth and what I call an advertisement of narratability (or celebratory gloss). It is phrased in terms designed to recall the text’s initial advertisement in the promise to preserve the renown of “great and wonderful deeds.” 
τε η τολµα   ανδρ

 θωµατος 

ι η κα ι ταδε 17. αυτη τουτων των ων αξ  τουτοισι  προς τα ε#πεα. [This boldness [i.e., that reported earlier] from the part of these men is worthy of wonder, and in addition also the words they said, as follows.] (7.135.1) Here and in other cases where the narrative is about brave deeds, “goodness” of conduct and what is deserving of mention are one and the same. This attitude conforms to the Homeric tradition, later pursued by 67. Cf., e.g., 8.124.2. In Herodotus, when the moral and the strategic standards of evaluation are separate—which is by no means always the case—the split between the two centers especially around the fluid term σο␸ ι η, which can be used to mean “intelligence” in a narrower sense and does not necessarily convey moral approval.

44

Telling Wonders

the praise poetry of Pindar.68 In Herodotus, however, what I have called evaluation of worth is more specifically moralistic than is praise of greatness of the Homeric type; and conversely, his field of celebration (“great and wonderful deeds”) is both different and broader than the range of what Homer, Pindar, or (in his own way) Thucydides claim is worth preserving.69 Perhaps because Herodotus’ concept of narratability is so unpredictable, the advertisement we find in the first sentence needs continuous tending along the logos. Terms of the celebratory code in fact recur throughout the Histories, often joined to programmatic statements and glosses of opinion. These express the narrator’s authority in determining what he will tell for no other reason than that it is, or he considers it to be, worthy of being told.70 Phenomena of very different types and magnitude, historical or ethnographic, are emphasized in this manner. Explanation in a broad sense, including interpretation and evaluation, purports to indicate why something in the world of the narrated occurs and what constitutes its importance, meaning, or worth. Celebration, in contrast, is simply designed to signal that a feature in the world of the narrated possesses some sort of importance, meaning, or worth. Earlier in this chapter, I showed that in many cases, self-referential glosses put the reader in charge of filling in the blanks left in the text and of interpreting the reality the text represents. So far as referential glosses have an unstable or multiple role, they achieve a similar effect, albeit even more implicitly. Referential metanarrative represents the main focus of my analysis. In the next two chapters, I examine how Herodotus explains through explanation. In chapter 4, I explore, albeit through the study of a single term, how he directs us to explain (if that is what he does), or what else he does, through celebration. 68. On Pindar’s praise poetry in relation to the Homeric tradition, see Nagy 1990,  150– 207. For Homeric praise of αρετ η in Herodotus, see my chap. 3, “Interpretation in the History.” 69. Thucydides magnifies his subject as the greatest war and the most worthy of report (1.1– 19, especially 1– 2; cf. 7.87). Cf. Herodotus at 7.20.2. The tradition of ancient historians’ magnification of their subject is broadly surveyed by Marincola (1997a, 34– 43). On praise and Homeric glory in Thucydides, see Immerwahr 1966, 177– 279.  70. These celebratory terms include words of the θωµα family; the phrases “great 

οδεξις, deed” or “great work,” “display of deeds” (ε#ργων απ 2.101.1), “deeds greater than   human” (2.148.6); the noun λαµπροτης (“brilliance” (2.101.1)); the adjective µεγας,  µεγιστος, as well as other superlatives; words and expressions indicating fame (verb

 ευδοκιµ εω), originality (“the first we know about” to do something), primacy (“the first in his time”), or uniqueness; various expressions equivalent to “worthy of being told”  

ολογος)  (αξι or “greater than words” (λογου µεζω, 2.148.1).

Chapter 2

Comparison

Comparison is an interpretive operation that “puts together” two facts for the purpose of explaining one on the basis of its similarity to or difference from the other.1 That which sparks the comparison is an element of the story. The second term can be also drawn from within the narrative, or it may come from outside of it, such as a past or present fact that belongs to the “real world” of the narration or to another “text” familiar to the narrator and his audience.2 One of the peculiarities of the Histories is that the boundaries of the narrative are especially fuzzy. By virtue of the contract that Herodotus establishes with his audience, everything is at least potentially part of the story he has to tell. In the actual telling, a fact that is brought in incidentally for the sake of comparison or some other reason may become the object of a narrative within the logos in a way that is hard to predict. Nevertheless, Herodotus’ references to events after 479, for example, coupled with his evident reluctance to include such references, demonstrate the existence of boundaries as well as their provisional nature.3 We expect a continuity between the logos and the “real world” of the narration in Herodotus that we do not expect from Thucydides, but at the same time, we acknowledge an inside and an outside and, between the two, a necessary break. Comparison may be implicit or explicit. It is explicit when the narrator directs the recipient of the narrative to consider a fact of the narrative in reference to some other fact by means of a gloss of comparison indicating similarity, analogy, or difference. It is implicit when the recipient of the narrative perceives on his or her own that a fact of the narrative wants to 1. On συµβαλλειν (put together, compare), see “The Texture of the Earth” later in the present chapter. 2. On the relation between a story and its extratextual context or subtext, see, e.g., Bal 1985, 81. 3. Cf. my introduction.

45

46

Telling Wonders

be considered in light of another and that a conceptual relation of analogy or (theoretically) difference links the two. In and by itself, difference establishes no relation at all. Even employing apples and oranges to denote items that cannot be “put together” at all is somewhat misleading. When two items are mutually related through difference, this can only be because they are similar in other respects. Resemblance lurks in the background every time comparison is an issue. If resemblance overcomes fundamental differences and makes them appear circumstantial, it constitutes analogy.4 Comparison and analogy may be activated “horizontally” to bind overlapping, concentric, or parallel classes of similar objects. But they also work “vertically,” through indices and symbols across different levels of reality, as in inductive prophecy. In the Iliad, the nine sparrows devoured by a red serpent somehow resemble and therefore represent the nine years of the Trojan War.5 In Thucydides, extraordinary natural cataclysms, though not ominous, are nevertheless analogous to and symbolic of the upheavals that the Peloponnesian War has produced (Thuc. 1.23.1– 3). In Herodotus, the world of animals mirrors the human world, while concrete actions and objects are indices for something different or more intangible. Whereas horizontal analogy is based on the notion that phenomena recur with variations, vertical analogy brings out the similarity of situations on different planes, so that one becomes a sign for the other. Whether explicit or implicit (and often simultaneously activated along the horizontal and the vertical planes), comparison and analogy are fundamental strategies by which the text of the Histories organizes its material. Because the logos contains so many story elements that escape the network of causal connections of the plot, classification and the comparative approach that classification entails provide a powerful glue; this in turn also acts on causally connected facts. In the historical narrative, it is most frequently facts belonging to different points of the chronological continuum that are compared to one another (diachronic comparison). The comparative field created by descriptions in the present tense extends not in time but in space (synchronic comparison). 4. Analogy is defined by Lloyd (1966, 175) as “any mode of reasoning in which an object or complex of objects is likened or assimilated to another.” 5. Il. 2.308– 32, especially 326– 29, cited by Corcella (1984, 33– 34). See also Lloyd 1966, 180– 85. The language of many oracles reported in the Histories also relies on symbolic analogy.

Comparison

47

Diachronic and synchronic comparison are largely distinct operations, partly because the material that is being compared in history and ethnography, respectively, belongs to different classes (“apples” and “oranges” in the proverbial sense). It just so happens, moreover, that in the two types of discourse, comparison mobilizes the narrator’s presence to a widely different degree. To put it simply, qualitative diachronic comparison largely imposes itself on the consciousness of the listener through the implicit similarity of the events narrated; synchronic comparison, by contrast, gains momentum by frequent explicit and far-ranging glosses that advertise the notion of qualitative similarity or difference by saying “X is like Y” or “X is different from Y.” What is the meaning of this discrepancy? To what extent are diachronic comparison and synchronic comparison mutually related and complementary strategies? I will begin dealing with these questions by considering comparison in the history. Comparison in Time Analogy as an Interpretive Tool Before I examine the forms and the contexts of glosses, I will briefly recall the role of implicit comparison. Like all the εργα µεγαλα τε κα ι θωµαστα [great and wonderful achievements] promised in the programmatic first sentence, each past event of Herodotus’ narrative is strictly speaking a particular and unique occurrence. In general, however, the exceptional historical event that deserves a place in the Histories is different mainly for its magnitude rather than for its quality.6 Superlative evaluations, for example, often serve to advertise narratability (the axion logou) by stating that certain story elements possess the greatest amount of a certain quality among others of the same class.7 Since they underline what is exceptional in a quantitative sense, these markers also indirectly identify classes of qualitatively similar phenomena. Thus Pisistratus’ “most simple-minded device” (1.60.3) and Hermotimus’ “greatest revenge” (8.105.1) make reference to other acts of deception and retribution across the broad range of 6. Thoma, “wonder,” the strongest marker for a qualitatively unique phenomenon, is  discussed in chap. 4. In past-tense narrative, µουνος more frequently singles out people who behaved differently from the others in the same situation (see, e.g., 1.147.2, 3.55.1, 7.107.1) rather than “the only time X ever happened” or “the only one who ever did X” in an absolute sense (see, e.g., 1.25.2, 9.35.1). 7. See Bloomer 1993.

48

Telling Wonders

the work and outside of it. Through classification, analogy wins the contest with difference. The proem of the Histories implies that several occurrences are similar enough to be grouped under the generalized headings “small city becomes great” and “large city becomes small” (1.5.4). Similarities among different actions, their motives, and their outcomes may emerge from the recurrence of words and concepts within different contexts.8 Occasionally, speakers compare and contrast. When they discuss circumstances of their present in light of events of their past, they encourage the audience to perform the same sort of analogical operation and apply to their own present the same or other parts of Herodotus’ logos.9 Helped by these clues and by the exceptional diachronic range of the Histories, we register the uniqueness of the events narrated, but at the same time, we also overcome it; we regroup facts in different ways on the basis of their mutual resemblance. In the theoretically endless variety of particulars, “X is like Y” over and over again; together these similar facts insistently recall elements of what lies outside of the logos and belongs to the real world of narrator and audience. Herodotus’ original listeners had far more practice than the modern reader for receiving the work in this manner.10 Assiduously studied, analogy in the Histories has provided an indispensable tool for interpreting the work.11 The repetition of similar events results in a number of patterns or—at the level of discourse—cultural codes that recur throughout the work.12 Some of the most conspicuous of these patterns have been identified, described, and named. They now represent canonical terms for speaking about the Histories: the crossing of geographical boundaries for the purpose of conquest;13 the “rise and 8. Cf., e.g., the first mentions of Deioces (1.96.1) and Themistocles (7.143.1): both are individuals rising in power. See Wood 1972, 17. 9. On the comparison between Sparta and Athens partially focalized through Croesus at 1.56– 69, see Gray 1997. Artabanus compares Xerxes’ prospective expedition against Greece to the earlier expeditions at 7.10α– γ. 10. See my introduction, n. 10 and corresponding text. For a parallel between tragedy and the work of Herodotus from the point of view of the audience’s receptiveness to analogy, see Raaflaub 1987, 231– 32. On the didactic and political dimension of tragedy, see, e.g., all the essays in Goff 1995; Sa¨ıd 1998. 11. See Immerwahr 1966; Wood 1972 (especially 17– 20); Lateiner 1984 and 1989; Corcella 1984. Cf. Waters 1971, but see Lateiner 1989, 165– 67. 12. To the extent that a code is the language the text uses to speak about something, patterns correspond to codes, though the notion of codes is broader. See my introduction, n. 25. 13. See my introduction, n. 26 and corresponding text.

Comparison

49

fall of the ruler”;14 the expedition of a superpower against a tough and poor nation, the so-called primitive opponent;15 the wise adviser or “tragic warner” mostly unheeded by the recipient of the advice, who rushes to his ruin;16 the pattern of imperialism;17 the exile who seeks refuge at the king’s court.18 An important recent addition is the kinginquirer, a figure of metahistorical significance who by analogy or opposition illuminates the purposes and methods of the histor of the Histories and his counterpart outside the text, Herodotus himself.19 Among the various concentric or overlapping patterns in the logos, the monarchical model is especially pervasive, since it tends to subsume many others to itself. This model is constituted by all the specific actions and features that serve as indices of actual or potential autocracy—of an individual’s attempt or ability to rise above a community, his own or someone else’s, and impose his will on it while living himself by different rules. Such actions and features are attributed especially to those who hold royal power or aspire to do so in a literal sense (though rarely, if ever, does a single character exhibit the full repertoire).20 They may be historically consequential (e.g., political manipulations; subversions of the social order; punishments; conquests; victories and defeats in war), or they may be mostly symbolic or connotative of the abnormal position of an individual within the state (e.g., crossings, mutilations, exceptional marriages, gift giving, athletic victories, trickster actions).21 An exceptional feature confirms the centrality of monarchy/tyranny as a predominantly negative paradigm in the Histories: one of the characters describes the phenomenon, and the components of his theoretical descriptions are 14. See Immerwahr 1966, especially 149– 98. 15. See Hellman 1934, 77– 98; developed by Cobet 1971, especially 172– 76. See also Flory 1987, 81– 118. 16. See Bischoff 1932; Lattimore 1939; Dewald 1985. Consider also the pattern of unheeded or misunderstood dreams and oracles (see Corcella 1984, 160). 17. See Evans 1991, 86– 87. 18. See n. 74 and corresponding text later in this chapter. 19. See Christ 1994. 20. See Immerwahr 1966; Lateiner 1984; Lateiner 1989, 163– 86; Corcella 1984, 163– 77; Hartog 1988, 331– 34; Gammie 1986; Dewald forthcoming b. For the pattern of monarchy as an archaic and classical Greek cultural stereotype, see McGlew 1993, 24– 35. 21. Thus, my monarchical model differs from what Dewald (forthcoming b) calls “the despotic template,” which she defines as “Herodotus’ description of the evils of autocratic rule.” The code of kingship is the broader discourse on kings that shapes and delimits the monarchical model. The subtlety of such discourse has led Flory (1987, 119– 49) to detect (wrongly, I think) advocacy in favor of a one-man rule in the text. Gray (1997) rightly emphasizes the variety of internal patterns.

50

Telling Wonders

verified by the actions of specific historical rulers throughout the narrative.22 The problematic of monarchy represents in fact much of what Herodotus’ work is about. It provides terms to be taken symbolically as  η),  imperialism, well as literally for presenting the issues of rule (αρχ transgression, oppression, and the competitive self-aggrandizing drives of both individuals and states.23 The narrative explores monarchy as a historical phenomenon and as a plausible prospect for the future of Greece in the literal sense (we tend to forget, for example, the extent to which Herodotus, with his long-range view of patterns, unwittingly predicted fourth-century monarchical outcomes). This is an especially conspicuous case, however, where the analogical system works not only horizontally but also vertically, that is, metaphorically, by assimilating to monarchy other manifestations of power, leadership, and alienation from the commonwealth.24 The bilateral equivalence between cities and kings, like the translation of rivers into moral boundaries, also largely depends on the symbolic and semic work of vertical analogy. The implicit interaction between horizontal and vertical analogy conveys the sameness of apparently different kinds of experience through a covert mode of communication akin to that of the ainos. Explicit Analogy While resemblance plays such a large role in the historical narrative, much rarer are metanarrative glosses of comparison that place two specific facts side by side and expressly say that one is somewhat like the other or, for that matter, different from the other. Yet there are other metanarrative clues. Additional allos (other) and adverbial kai (also) serve as markers of horizontal assimilation. The qualification of events in terms of their position in a series of similar items—for example, first, second, third conquests of Ionia (1.92.1, 1.169.2, 6.32)—brings out the continuity of a historical process potentially down to the time of narration (what about, for example, a fourth conquest of Ionia?). Coincidences 22. The character is Otanes in the Constitutional Debate (3.80). See especially Lateiner 1989, 167– 81. The other generalized descriptions in the debate are not equally related to the appearance of patterns in the narrative. 23. The topic of empire in the Histories has been discussed by Stadter (1992). 24. See discussion of horizontal and vertical analogy earlier in this chapter. The participation of the monarchical model in the symbolic code of the Histories is indicated, e.g., by Croesus’ words to Cyrus that Solon spoke for all men and especially for those who deem themselves happy (1.86.5).

Comparison

51

among mutually autonomous occurrences point to a unitarian historical movement and mysterious interconnections.25 Two items may follow one another in close narrative sequence on the basis of their similarity. Thus, the narrator consecutively mentions two queens of Babylon (the second “more clever than the one who reigned before,” in an example of quantitative difference highlighting likeness); another time, he recalls two Persian governors in Thrace, Mascames and Boges, both distinguished for bravery.26 A queen pattern, in the first instance, and a Persian cultural  definition of αρετ η (valor), in the second, emerge from the juxtaposition of similar historical cases. Occasionally, through a gloss of analogy, the discourse tampers with the story by bringing in a more or less extraneous element, as when the narrator compares different stages of a war.27 The comparison may attach a new small narrative to one that has just ended: thus, the report of Xerxes’ decision to send the Greek spies home unharmed (7.146.2– 7.147.1) attracts the announcement of the narrative of another, similar gnome about his releasing enemy ships captured at the Pontus (οικε . . . 28 By this compositional principle, the narrative accumulates  τ ηδε αλλη). evidence of a certain type and thereby trains the listener to generalize and interpret. A more drastic manipulation of the story occurs when the second term of comparison belongs to a different historical context than the first. The name of this woman who ruled [in Egypt] was the same as that of the Babylonian [queen] [το περ τ η Βαβυλωνι  η], Nitocris. (2.100.2) Trivial as it may seem, this comparative back reference provides a significant reading direction, one the text has already suggested by the simple juxtaposition of the two Babylonian queens in the earlier narrative 25. E.g., Cambyses is wounded in the same part of his body where he has wounded Apis (3.64.3); the battles of Himera and Salamis occur on the same day (7.166), and so do those of Plataea and Mycale (9.100.2). 26. 1.184– 85.1, 7.106– 7. In a continuous dynastic line, a king is frequently compared  . . . δε from eunomie to kakotes), to his predecessor. See, e.g., 2.124.1 (a reversal with µεν 2.127.1 (similarity), 2.134.1, 1.103.1, and the more elaborate 2.110.1– 3. 27. See 9.98.4: the strategy of Leotychides at Mycale was the same as that of Themistocles at Artemisium (see Immerwahr 1966, 256). The battle of Salamis is constantly compared to the battle of Artemisium: see 8.42 (same commander and larger fleet), 8.66.1 (number of enemy not inferior). See Immerwahr 1966, 268. Cf. 1.191.3: to capture Babylon, Cyrus did that which (τα περ) the Babylonian queen had done; cf. 3.152. 28. Cf. 7.144.1, 7.57, 7.114.2, 4.78.1.

52

Telling Wonders

(1.184– 185.1). Here Herodotus explicitly adds the Egyptian Nitocris to the pattern, as if the homonymy were the sign of other analogies—two women, both Eastern, both rulers, both shrewd, and so on. Despite certain striking differences between them, the narrator establishes the continuity of a type across time and space.29 Within the small group of glosses comparing contextually distant items, two are of particular interest. On the formal level, both transform analogy into historical action by assuming mimesis. In other words, the narrator says not that “X is like Y,” but rather that “X imitates Y.” This formulation agrees with his general reluctance elsewhere to exercise authority over diachronic analogy. These glosses almost suggest that the responsibility for bringing two disparate actions together lies not with the collector of the logoi but with the historical agents who chose to play someone else’s role.30 More importantly, both glosses belong to the same broad analogical field whose overarching prominence in the logos I have already mentioned—the monarchical model. Though they do not in themselves fulfill an indispensable function, they confirm the symbolic aspect of this analogical network in the Histories. Both glosses of comparison and the narratives they respectively introduce serve to justify the analogy between the citizen of a Greek polis and a monarchical ruler. The Monarchical Model in Athens The first gloss leads to the abrupt narrative juxtaposition of Cleisthenes, the tyrant of Sicyon, to the Athenian democratic reformer of the same name. Cleisthenes of Athens enters the logos because he belongs to the story entitled “Athenian Ordeals and Achievements after the Fall of Tyranny” (see the introductions at 5.65.5– 5.66.1). The section of this narrative where Cleisthenes plays an active role is deceptively dry. After prevailing over his political rival Isagoras and obtaining the support of the people, he replaced the four old Ionian tribes with ten new ones; he named these after ten local eponymous heroes and discarded the old denominations (5.66). 29. The queen pattern is rich in symbolic associations. For related studies, see Dewald 1981; Tourraix 1976; Flory 1987, 23– 48; Munson 1988; Dillery 1992.  30. In Herodotus, the verb µιµεοµαι refers to animate imitators in the sense of “doing as someone else does or did” and does not necessarily imply awareness or intention to imitate (see, e.g., the analogy drawn by Cambyses’ wife at 3.32.4), but in practice it almost always refers to a derivative imitation (see, e.g., 2.104.4 in an ethnographic context and 4.166.1 in a historical one).

Comparison

53

This account of the foundation of the new Athenian democracy, however, bristles with negative signs.31 It identifies Cleisthenes and Isagoras, respectively, by referring to the former’s bribery of the Delphic oracle and to the latter’s foreign origins (5.66.1). It denotes their political prominence with a verb (ε δυναστευον, “held sway”) that is appropriate to narrow oligarchic circles. Herodotus elsewhere applies it to aspiring tyrants and corrupt potentates.32 The two men “engage in stasis for the sake of power” [ε στασι ασαν περ ι δυναµιος],33 and Cleisthenes coopts  δηµον  the demos at large to his hetaireia (τον προσεταιρι ζεται). Because, in the time of Herodotus, at least, Athenian hetaireiai were exclusive aristocratic clubs formed with the aim of forwarding the political advancement of their members, the term here suggests that Cleisthenes’ democratic stance was a self-serving maneuver in the context of an aristocratic power struggle.34 A verbal parallelism confirms Cleisthenes’ resemblance to Isagoras, who aimed at establishing himself tyrant in Athens:  first Cleisthenes, when he is “losing ground” (ε σσουµενος, 5.66.2; cf.  5.69.2), and then Isagoras, when he is “losing ground” (ε σσουµενος,  5.70.1), come up with a special device (see ε πιτεχναται at 5.70.1). Just as the one elicits the friendship of the Athenian people, the other appeals to the Spartan king, his guest-friend. Once Cleisthenes, banished by Isagoras and the Spartans, fades from the story and from the narrative, the next section reports the resistance of the Athenians to the Spartan intervention in favorable terms (5.72.1– 2), and a conspicuous interpretive gloss praises the resulting democratic order (5.78). The Cleisthenes passage, however, emphasizes the reformer’s bid for personal power while representing the reform itself in such reductive terms that its democratic implications either in a practical or in an ideological sense appear unintelligible: “He gave the Athenians ten tribes, whereas they used to have four, discarding the names of the children of 31. See Strasburger 1955, 15; Bornitz 1968, 49– 50; Fornara 1971, 55. Cf. Myres 1953, 180– 82. 32. On dunasteia, see Ostwald 1969, 113, 116– 17, citing Thuc. 3.62.3– 4. Dunasteuein appears in a negative context in the Histories at 6.39.2, 6.66.2, and 9.2.3. The verb is used of Athens with similar effect (5.97.1). 33. 5.66.2 Stasis (civil struggle) is another oligarchic term (see 3.82.3) and a key negative concept in Herodotus: cf. 8.3.1, quoted and discussed in chap. 3, “The Evils of War.” 34. Georges (1994, 160) dubs this expression a “conscious paradox.” Cf. the less derogatory formulation of Arist. Ath Pol. 22.1. Herodotus’ term is anachronistic for the beginning of the fifth century. See Ostwald 1969, 142– 43. Ober (1993, 227) renders the tense and the middle form with “Kleisthenes embarked in the process of becoming the demos’ trusted comrade”; but the reference to hetairiai cannot possibly have benign connotations.

54

Telling Wonders

Ion, Geleon, Aegicores, Argades, and Hoples, and coming up with  the names of another group of heroes, all native of the land [ε ξευρων] except for Ajax” (5.66.2). The verb ε ξευρι σκειν here connotes a contrivance, something new and invented for the sake of expediency but at the same time ostensibly recovered from the traditional past.35 After the narrative has done its work through connotation, the coup de grace ˆ arrives in the form of a gloss of comparison that introduces the account of a reform by the tyrant of Sicyon by stating that it provided the model for that of the Athenian Cleisthenes. In this [i.e., the replacement of the old tribes in Athens], this   Cleisthenes it seems to me [δοκεειν ε µοι ] was imitating [ε µιµεετο] his maternal grandfather Cleisthenes the tyrant of Sicyon. (5.67.1)  The self-referential δοκεειν ε µοι identifies both Cleisthenes’ mimesis and the resulting similarity as interpretations of the narrator. The family relation and the homonymy (which we will never be allowed to forget; cf. 6.131.1) serve as corroborating evidence for the one and the other, respectively. Herodotus often attaches one story to another incidentally, by exploiting an intrinsic story connection; he adds the pretext of the axion logou, narratability, as he lets the thematic correspondences take care of themselves.36 Here, however, he openly says that the tyrant of Sicyon is brought into the logos not because the narrative about Cleisthenes of Athens provides an opportunity for talking about his famous grandfather but because of the tyrant’s thematic relevance to the present context of the democratic reform. The juxtaposition would have been enough to trigger the analogy, and the narrator did not have to be and normally is not as explicit as he chooses to be in this case. The analeptic narrative introduced by the gloss of comparison compounds the damage for the Athenian Cleisthenes, whose policy it purports to explain by analogy and derivation. The section on the reorganization of the tribes in Sicyon, which is the essential point of the similarity between the Athenian democratic reformer and his grandfather, is pre ι σκειν are often used positively in reference to a 35. The verbs ε ξευρι σκειν and ευρ people’s inventions (see 4.46.2, 8.98.1, 2.92.1) but are also applied to constructs (see, e.g., 2.23, 4.79.3). See especially Cambyses’ far-fetched “discovery” of a nomos overriding another Persian law that he does not want to obey (3.31.4). 36. For the ways in which narratives are brought into the logos, see especially Jacoby 1913, 344– 50, 383– 92; Pagel 1927, 41– 61; Erbse 1961, 243– 57; Munson 1986.

Comparison

55

ceded by another account. This relates how the older Cleisthenes outlawed Homeric recitations (the Homeric poems being full of the glory of the Argives) and managed to expel the Homeric hero Adrastus, who happened to be buried in Sicyon. Consulted on the matter, the Pythia forbade it and declared that whereas Adrastus had been the king of the Sicyonians, Cleisthenes was their lapidator (λευστηρα, 5.67.1– 2). Reference to the typical despotic action of “killing men without trial” is thus, through the voice of the oracle, inserted into an account of violations of a religious nomos.37 To circumvent the Delphic interdiction, Cleisthenes “invented” a device for making Adrastus leave on his own. The verb ε ξευρι σκω (5.67.2) connotes once again subversive manipulation of tradition, as in the case of the new heroic names of the younger Cleisthenes’ Athenian tribes. The two actions are in fact parallel, since the Athenian Cleisthenes, according to Herodotus, abolished the four heroes as namesakes of the city tribes, while his grandfather found a way to get rid of one hero as the recipient of cult in the city. The “device” of Cleisthenes   the Elder (µηχανην; cf. µηχανωται applied to the religious ruse of Pisistratus at 1.60.3) consisted in importing to Sicyon from Thebes the hero Melanippus, who had been Adrastus’ worst enemy, and in transferring Adrastus’ solemn festivals, sacrifices to him, and his choruses to the god Dionysus (5.67.2– 5). The survey of these blatant violations of the city’s time-honored traditions (cf. 5.67.4– 5) is followed by the account of how the tyrant changed the names of the Dorian tribes in Sicyon “in order that Sicyonians and  Argives would not have the same ones.” He mocked (κατεγελασε, a term crucially related to violations of customs at 3.38.1– 2) the Dorians of Sicyon, by choosing for them new denominations derived from the words for “pig,” “swine,” and “ass.” For his own (non-Dorian) tribe, however, he reserved a dignified name, “Archelaioi,” referring to his own arche, “rule” (5.68.1). This reform outlived Cleisthenes only by sixty years (5.68.2). After this pointed reference to the tyrant’s fall, the narrator reemerges to bring us back to the younger Cleisthenes in Athens with a resumptive gloss that reiterates almost obsessively analogy and its supportive story elements—imitation, kinship, and homonymy.  see Elayi 1979, 224– 27; Ogden 1993. McGlew 37. See 3.80.5. On the term λευστηρ, (1993, 16) notices that Cleisthenes’ expulsion of Adrastus constitutes a reversal of the expulsion of the deceased tyrant’s bones extra fines. In Herodotus’ narrative, it also constitutes a reversal of the expulsion of the Alcmaeonids (5.72.1: same verb ε κβαλλειν) as a result of their ancestral curse. On the political dimension of hero cults, see Boedeker 1993.

56

Telling Wonders

This is what Cleisthenes of Sicyon had done [retrospective conclusion]; and, indeed, with regard to the Athenian Cleisthenes, who was the son of the daughter of this one from Sicyon and was named  after him, it seems to me [δοκεειν ε µοι ] that he also despising the Ionians and so that the Athenians would not have the same tribes,  imitated [ε µιµησατο] his namesake Cleisthenes. (5.69.1) This second gloss of comparison further enhances the relevance of the preceding narrative about Cleisthenes of Sicyon to the younger Cleisthenes. The tyrant, for example, made sure to reserve a special place in the city for his own tribe (5.68.1): this casts doubt on the impartiality of the tribal reform of the democratic leader.38 Herodotus’ attribution of antiIonian bias to the Athenian Cleisthenes (parallel to the anti-Dorian intent of Cleisthenes of Sicyon) makes reference to the hostility among different groups of Greeks and to later hegemonic struggles.39 I have stopped to discuss this explicit comparison because the analogy it brings into the open affects Herodotus’ overall representation of modern democratic Athens—born at this historical point after the fall of tyranny and risen to greatness through the Persian Wars and beyond. Tyranny, as it has been often repeated, is viewed by fifth-century Greek thought as antithetical to Greece.40 In Herodotus, two elements favor the assimilation between Greek sixth-century tyrants and Eastern monarchs: a deliberate confusion in his use of the terms turannos, basileus, and mounarchos;41 and his attribution of similar features to kings and tyrants, despite all the variety of his representations in individual cases. The analogy between the two Cleisthenes, however, undermines the notion that the “tyrannical” stage in the political development of the Greeks has come to an end. As the creator of Athenian democracy is moved closer to the opposite tyrannical/barbarian camp, the monarchical model begins vertically to affect the new Athens as well, firstly through its leadership and secondly through the demos, both of them “tyrannical” in a broad metaphorical sense. After representing Cleisthenes’ bid for primacy as analogous to the 38. Some modern historians have tried to explain the apparent territorial oddities of Cleisthenes’ reorganization by speculating that the Alcmaeonids may have received special treatment. See Fornara and Samons 1991, 39– 56, for a discussion of the problem. 39. Here Herodotus’ analogy may not be as misleading as Macan argues (1895, 1:211). See Hall 1997, 53– 54. 40. See, e.g., Georges 1994, 37– 46. 41. See Ferrill 1978; Dewald forthcoming b.

Comparison

57

factionalism of the time of Pisistratus, the narrative partially recruits to the monarchical model subsequent Athenian democrats who seek or exercise power with high-handed methods, albeit in a constitutional setting. Secrecy and self-interest mark the essential alienation of each from the commonwealth of citizens. At the same time, however, the demos, whom Cleisthenes has coopted to his ambitions, has inherited power in internal and external affairs, including, at the time of narration, his hostile policy toward the Ionians.42 With democracy, king and city have to some extent become one, because while separate from the city and a threat to its institutional integrity, each “tyrannical” or potentially tyrannical leader at the same time often expresses the will of the demos. The city is both the victim of an individual’s schemes and his willing accomplice—both subject and ruler, in a metaphorical sense. This contradiction emerges from the cases of Miltiades and Themistocles. Miltiades plans an imperialistic venture against Paros on his own by keeping the Athenians in the dark as to its destination and deceiving them about its goals. The sovereign democratic assembly, however, decides on the expedition (6.132). When the expedition fails, they assert their authority over the leader by bringing him to trial. Miltiades’ conviction by the city and a death that appears as the result of divine punishment coincide to mark the “tyrant’s” downfall.43 Themistocles, like the Pisistratids, establishes a personal connection with Xerxes (8.109.5) and has tyrannical features of his own.44 But he also embodies the enterprising and opportunistic tendencies of the city as a whole and in most cases acts as its executive.45 The city’s own monarchical image catches up with the representation of the individual leader seeking power for himself in the state. At the end of the Histories, individuals are but pale reflections of the policy of the city. It is “the Athenians” who 42. The Athenian prejudice against the Ionians is referred to the present of narration by  See Fornara and Samons 1991, 106– 9, on the the gloss of testimony at 1.143.3 (κα ι νυν). Athenian contempt of the Greeks of Asia as a psychological justification for the aggressive policy of Athens against its allies after the Persian Wars. 43. In his expedition against Paros, which results in impiety and death, Miltiades imitates in reduced format the career of the arch despot Cambyses (see 6.132– 36; cf. 3.37, 3.16, 3.27– 29, 3.64.3, 3.66.2). See Immerwahr 1966, 191– 92; Corcella 1984, 137; Flory 1987, 115. 44. Themistocles imposes his own strategy on the Greeks at Salamis (see 8.62.2– 63, 75– 76) and compulsion on other Greeks after the battle (see 8.111– 12). See Munson 1988, 101. Themistocles’ self-interest and deception (see, e.g., 8.4– 5, 8.109– 10.1) are characteristic of the rising tyrant: see “The Monarchical Model in Sparta” later in the present chapter. 45. For the affinity between Themistocles and Athens, see Immerwahr 1966, 223– 25; Wood 1972, 185– 86; Raaflaub 1987, 227; Munson 1988, 100.

58

Telling Wonders

begin the offensive stage of the war against Persia by crossing over to the Chersonese (9.114.2). The commanders follow the instructions of the Athenian community at home (9.117). Xanthippus is not prominent in the overall narrative, so the crucifixion of Artayctes, carried out by his orders, tends to convey the barbaro-monarchic mentality of postwar Athens as a whole.46 But Xanthippus’ son Pericles, chronologically the last Athenian leader in the Histories, is significantly introduced as a descendant of Cleisthenes, the one “who established the tribes and the democracy and who bore the name of his maternal grandfather from Sicyon” (6.131.1). This gloss of identification inevitably recalls the earlier comparison between the democratic reformer and the tyrant precisely at the moment when Pericles flashes through the narrative of the Histories in the guise of a lion, symbolically completing the identification between leader and city.47 To some of Pericles’ contemporaries, the arche of Pericles in Athens resembled a tyranny.48 But Pericles himself, according to Thucydides, declared that the arche of the Athenians in Greece was “like a tyranny.” A tyrant is a “lion, mighty, ravenous” among his fellow citizens, as the Pythia says of Cypselus, (5.92β3). Pericles is the leader-turannos of the city-turannos, the representative of the sovereign demos, a lion at home and to the rest of the Hellenic world. The fluid naming of political realities in contemporary discourse represents the linguistic foundation of the analogical impact of Herodotus’ text. By glossing over substantial distinctions, the monarchic analogy implicates the Athenian demos as it does no other Greek community in the Histories and in a way that directly resonates with the fifth-century notion of the rise of a polis turannos in Greece.49 In this overarching perspective, the explicit comparison of the two Cleisthenes is a reading 46. The juxtaposition of the sacrifice of Eobazus by a Thracian tribe (9.119) and the Athenian execution of Artayctes (9.120) encourages comparison between the two actions. For the ambivalence of Herodotus’ representation of Athenian action at the end of the Histories, see Dewald 1997. 47. 6.131.2. For lion symbolism, see my chap. 4, “Wondering Why” and “Wonder and Disbelief.” For glosses of identification, see my chap. 1, “Referential Glosses.” 48. For Pericles as tyrant, see Plut. Per. 7.1– 2, 12, 16, Cratinus (frag. 240 Kock) quoted in Plut. Per. 3. Cf. Thuc. 2.65.9. For the Athenian empire as tyranny, see Thuc. 2.63.2 (words of Pericles); cf. 3.37.2. 49. For the idea of Athens as the polis turannos, see especially Thuc. 1.122.3     (τυραννον . . . πολιν), 1.124.3 (πολιν τυραννον), 6.85.1; Aristoph. Knights 1111– 12; Knox 1971, 53– 106; Raaflaub 1979; Raaflaub 1987, especially 223– 25, 241– 46. For the idea of the demos as tyrant at home, see Kallet 1998, 52– 54.

Comparison

59

direction whose importance is proportionate to the political change Cleisthenes initiated in Athens. The Monarchical Model in Sparta Elsewhere in Herodotus’ Greek narratives, the monarchical model more straightforwardly serves to dramatize an internal dilemma—the tension that exists between individual and state, even within a constitutional order as free as possible from the danger of true tyranny. Sparta not only represents the paradigmatic setting for this sort of confrontation but also provides the context for the second gloss of historical analogy, which I am now going to discuss. The seer Tisamenus of Elis performed the sacrifice for the Greek army before the battle of Plataea. The mention of his role provides the occasion for a narrative on how he and his brother had just recently managed to become Spartan citizens (analepsis: 9.33) and on the victories the Spartans subsequently achieved after he had taken permanent service with them (prolepsis: 9.35). I will focus on the first (analeptic) part, which triggers the comparison and is introduced, in terms of narratability, by the rather unusual marker of absolute uniqueness: “These were the only men in the world who became Spartan citizens.”50 Tisamenus belonged to the family of the Iamidae of Elis. On receiving a prophecy from Delphi that he was destined to win five victories of the greatest importance, he started training to become a champion in the pentathlon (the five agones of the pentathlon would give him five victories in one). This was not what the oracle meant, however, and Tisamenus did not attain the athletic success he expected. The Spartans understood the oracle and offered Tisamenus a post in their army. But when Tisamenus demanded as compensation that they make him citizen “with full privileges,” the Spartans were indignant and let him go. Subsequently, however, fearing the Persian invasion, they sought him out again and even yielded to his demand that his brother as well as himself be awarded citizenship (9.33.1– 5). The flashback I have just summarized belongs to the first of three narratives about Greek seers that interrupt the account of the battle of Plataea and mark a pause in the “real” action.51 The proleptic section of 50. 9.35.1. How and Wells (1928, 2:302) point out that this is an exaggeration in light  of 4.145, 4.149, and 7.134. On Herodotus’ use of µουνος, see n. 6 in the present chapter. 51. See Masaracchia 1975, 153.

60

Telling Wonders

the Tisamenus story serves to anticipate prophetically the outcome of the battle, because it mentions Plataea as the first of the five victories Tisamenus was destined to win for the Spartans (9.35.2). Seers are important indices on the battlefield of Plataea, since here more than ever we feel the presence of the divine.52 The stories of Tisamenus and the other seers are not, however, centered on the role of their protagonists as intermediaries between the divine world and human actions; rather, they seem to pursue a different set of themes. The first and third narratives are expanded glosses of identification for Tisamenus and Hegesistratus, the seers on the Greek and Persian sides, respectively. The second and apparently most accessory of the seer narratives brings in the mythical code, with the heroic diviner Melampus, and gives the key for interpreting the other two.53 It is introduced by a gloss of analogy that counterbalances the uniqueness of Tisamenus’ achievement of Spartan citizenship by pointing out its structural similarity with an earlier event of a different order.  In making these demands, [Tisamenus] was imitating [ε µιµεετο] Melampus, if one takes the liberty to compare/imagine a bid for  ε ι κα σαι βασιληι ην τε κα ι citizenship and one for kingship [ως  πολιτηι ην α ι τεοµενους]. (9.34.1)54 As in the glosses on the two Cleisthenes discussed earlier (5.67.1, 5.69.1), this statement expresses the analogy in story terms, although the 52. See Immerwahr 1966, 294. 53. For the famous Melampus (mentioned at 2.49.1 and 7.221) and his alter ego and brother, see especially Od. 11.281– 97 (where Melampus is not mentioned by name), 15.225– 46; Hes. Ehoiai frags. 37, 129, 130, 131 MW; Pherecydes FGrHist 3 F 33, 114, 115; Bacch. frag. 4.50– 51 SM; Pind. Paean 4.24– 26; Diod. 4.68.4– 5; Apollod. 1.9.11– 12. See also Lloyd 1976, 224– 25.  54. Α ι τεοµενους is a correction by Stein. Masaracchia (1975, 154 n. 173; 1978, 169–  70) stands by the manuscript reading α ι τεοµενος, referring it to Tisamenus and interpreting  ε ι κα σαι absolutely, in the sense of “suppose” (as at 1.34.1). He argues that with τε και , ως one must refer both βασιληι ην and πολιτηι ην to the same subject(s), with the translation “In saying this, he [i.e., Tisamenus] was imitating Melampus—so one may suppose—by asking for both the kingship and the citizenship.” My translation follows Stein and most other critics, but Masaracchia’s reading, which brings out the ambiguity of the gloss, (1) is consistent with the fact that nowhere else does the narrator apply the verb ε ι κα ζω in the sense of “liken,” to himself and (2) renders the connective force of the phrase τε και , which joins together the two terms more strictly than is expressed in the translation “if we compare people asking for citizenship and [people asking for] kingship.” I agree with Masaracchia that the narrator wants the listener somehow to envision Tisamenus as asking for the kingship.

Comparison

61

contextual link (both Tisamenus and Melampus are seers from Elis) this  ε ι κα σαι marks the comparitime remains implicit. The self-referential ως    son as an interpretation, like δοκεειν ε µοι in the Cleisthenes passage. The narrative records that when the Argives wanted Melampus to cure their women who had been seized by Dionysiac madness, he requested onehalf of the kingdom as compensation. The Argives at first refused. Since the women’s madness increased, they eventually decided to give him what he wanted. Realizing that the Argives were desperate, Melampus raised his price and demanded not only half the kingdom for himself but also another third for his brother. Reduced to a tight spot, the Argives agreed to both requests (9.34.1– 2). The concluding statement repeats the comparison: “In the same way the Spartans too, since they desperately needed Tisamenus, yielded to him in everything” (9.35.1). Mythical history is never a focus of Herodotus’ exposition; the myth of Melampus here provides an archetype to the modern story of Tisamenus and lifts it from the realm of the literal.55 An emergency places both individuals in a de facto position of power so that they obtain an exceptional political advantage that violates the city’s integrity.56 By using Melampus to interpret Tisamenus, the text emphasizes the invasive character of Tisamenus’ request and paradoxically transforms his achievement of citizenship into a metaphor for the acquisition of kingly power. As in the case of Cleisthenes of Athens and Cleisthenes of Sicyon, we are here in the presence of vertical analogy, in which the juxtaposition depends on the symbolic code. The narrative tries hard on its own to convey the idea that the position Tisamenus acquires in Sparta signifies something beyond what the story allows. Tisamenus enters the Plataea narrative as a seer, but the analeptic passage de-emphasizes that fact, even as it tells the story of how he became a seer. It does not remind the audience that the genos of Tisamenus, the Iamidae, is a famous family of seers, nor does it attempt to justify the strange fact that despite his background, Tisamenus misinterprets a prophecy, especially one concerning something as predictable as his future as seer.57 Tisamenus proceeds to train in athletics—an index 55. On Herodotus’ distinction between historical and prehistorical space, see especially 1.5.3 and 3.122.2. See von Leyden 1949– 50; Shimron 1973; Hunter 1982, 105– 6; Lateiner 1989, 118– 23. For the symbolic function of heroes in Herodotus, see Vandiver 1991, especially 68– 69, on Melampus. 56. Cf. Cobet 1971, 71. 57. The Iamidae are descended from Apollo; see Pind. Ol. 6.35– 72; Paus. 6.2.5. A “divine seer” plays a role in Simonides (frag. 11W2, line 42). It is tempting to speculate that

62

Telling Wonders

of political ambition—and not even his last-minute defeat at Olympia induces him to reconsider the meaning of the oracle.58 When the Spartans correctly interpret the oracle, the narrative implies that they understand the five victories prophesied by Delphi to be military victories to which Tisamenus would contribute as diviner. No one says so in so many words, however: “The Lacedaemonians, having realized that the proph ι ους αγ  ωνας],  ecy referred not to athletic but to martial contests [αρη tried to persuade Tisamenus to become, for a fee, leader in their wars   πολεµων],  [η γεµονα των together with those of the descendants from Heracles who were kings” (9.33.3). This job description forgets to limit Tisamenus’ new assignment to the field of religion, just as the preceding section of narrative has underplayed Tisamenus’ specialized vocation. Though the duties of the diarchs included priestly functions (see Xen. Lak. Pol. 15.2, 13) and though the vagueness of the Spartans is plausible, their reported expression nevertheless resembles a proposal that Tisamenus fulfill the same role as the Spartan kings. Neither a proven diviner from the start nor an athletic victor, Tisamenus is an ordinary man who obtains something unique and hard to achieve. Added to Tisamenus’ kingly position as “war leader,” the attainment of citizenship for himself and his brother is all that is missing for the metaphor of a second diarchy.59 Inserting the Melampus story through explicit comparison establishes that the audience should imagine (ε ι κα σαι) a vertical analogy between one who became king in a literal sense and one who became “king” in a figurative sense, because the narrator himself reads his material in this way. This gloss is therefore of some importance in endorsing the application of the symbolic code to the analogical interpretation of the Histories. As Tisamenus the citizen is like Melampus the king, so the reverse is true. Literal kingship is in Herodotus the paradigmatic manifestation of the abstract concept of personal power. Within Herodotus’ complex exploration of the ways and means of Simonides’ narrative of Plataea was somehow the source for Herodotus’ Tisamenus/ Melampus analogy, which has a poetic cast. 58. For the tyrannical connotations of athletic victories, see 5.71 (Cylon); cf. 5.47, 6.36, 6.103, 6.122, 6.125– 26. See Nagy 1990, 186– 87. On athletic victories by seers, see Pritchett 1979, 55, citing Pausanias (6.4.5, 6.17.5– 6) and Pindar (Ol. 6). 59. The brother’s name is Ηγι ας (9.33.5), which according to Macan (1908, 1.2.668) “is Ionic (and Attic) for Αγι ας or Αγι ας, a name perhaps identical with #Αγις.” The parallel with Melampus marks the end of the analeptic part of the insertion about Tisamenus. In the proleptic continuation, which has a different function, Tisamenus’ role as seer becomes again central and is explicitly mentioned (9.35.1).

Comparison

63

kingship (and metaphorical kingship), we recognize two partially distinct but overlapping patterns. The first, theorized through the words of Otanes in the Constitutional Debate and embodied in the figures of specific autocrats in the narrative, concerns the way in which the king exercises his power.60 More pertinent to the Tisamenus/Melampus sequence, however, is the second pattern, which focuses on how men who count as outsiders to the system attain kingly status. This pattern is especially important in the Histories because it provides a bridge between the inaccessible hereditary monarch figures and “every man”: any individual may upset the existing political structure by rising in status on account of chance or skill, often without the backing of force and even with the consent of those on whom he imposes himself. The royal bodyguard Gyges stumbles into the kingship under undesirable circumstances and makes the best of things (1.8– 13). Amasis, a man from the people, suddenly becomes king by the decision of the army and then proceeds to reconcile the rest of the Egyptians to his rule by means of sophie (2.162.1– 2, 2.172.2). Darius, an Achemaenid close to the throne, is nevertheless treated as a more or less ordinary Persian noble who takes advantage of a situation of crisis: he must first reinstitute the monarchy and then obtain the post for himself with a ruse (3.82– 83.1; 3.85– 87).61 The aristocrat Pisistratus drives around in a cart drawn by mules (an index of inferior status) and gains power by his tricks (1.59– 63, especially 1.59.4). By illustrating the accessibility of monarchical rule to the clever or lucky, these stories emphasize its essential illegitimacy. Thus, the founder of the Persian monarchy is the mule Cyrus, a mixed breed, the grandson of a king and a social outcast.62 Deioces, an especially insidious upstart, begins his career in a society that has achieved autonomie and eleutherie (freedom), and he leads it to permanent doulosune (slavery) and tyrannis all on his own (1.95.2,  and lusts for absolute power. Already 1.96.1). Deioces is clever (σο␸ος) highly regarded, he becomes even more so by practicing justice at a time of great lawlessness in Media. But when all depend on him for arbitration, 60. See discussion of the monarchical model earlier in this chapter. 61. Darius was the last of the seven to join the conspiracy (3.70) and was “a spearbearer of Cambyses and still a man of no great account” (3.139.2). 62. For Cyrus as mule, see 1.55.2, 1.91.5– 6. See also Nagy 1990, 335– 37. Persecuted babies who grow up to become rulers, such as Cyrus and Cypselus (5.92γ2– 4), illustrate the unthreatening beginnings of monarchical power. A marginal instance of the pattern of the rise to kingly status is the rascally thief who becomes the son-in-law of Rhampsinitus (see 2.121).

64

Telling Wonders

Deioces becomes unwilling to continue as judge; he claims that it is not expedient for him to neglect his own affairs and tend to those of his neighbors, settling their controversies day in and day out. Given the present state of lawlessness, the people decide to give themselves a king, and their choice falls on Deioces (1.97– 98.1). His actions from now on create the precondition for the exercise of absolute power enhancing the separation of the monarch from the community and his existence beyond and above the law.63 The end of the story of Deioces almost rejoins the other pattern that focuses on full-blown monarchical rule. The first part, however, describes the mechanisms of his coming to the throne and is strikingly similar to the story of Melampus. Deioces refuses to be judge among people who need his arbitrations, until they make him king. Melampus refuses to rescue the beleaguered city of Argos by his divination, unless he and his brother receive nothing less than the kingship. The narrative about Melampus reproduces the pattern of the Median Deioces in an ancient Greek setting, namely, Argos in the heroic age. But Melampus is brought in as the analogue of Tisamenus, who in turn transfers the pattern to modern Sparta. Each of these three men—Deioces, Melampus, and Tisamenus— possesses a special skill that the community needs, and each obtains a contract of power that satisfies his ambitions. The dual citizenship of Tisamenus and his brother is a term of the symbolic code of kingship as it emerges in the Histories from narratives about Sparta. Far removed from autocratic monarchy in the Eastern sense, Sparta is also historically immune from tyranny as is no other city in Greece.64 At the same time, Sparta provides monarchical paradigms through those who occupy or are close to occupying the peculiar office of the dual kingship. Thus, one issue that symbolizes the ambivalence of the Spartan kings as individuals vis-a-vis ` the constitutionality of their office is whether Demaratus is the legitimate heir to the throne or a more or less illustrious (mule-related) interloper.65 Many Spartans of royal blood in 63. Deioces’ literal isolation in the fortress of Ecbatana (see 1.98.2– 100) symbolizes his autonomy and unaccountability, such as, in the words of Otanes, “would place even the best of all men who occupies this office outside of customary ways of thinking” (3.80.3). This condition is a source of abuses, if not by Deioces himself, by his successors. 64. See 5.92α1– 2; Thuc. 1.18.1. 65. If he is not the son of his predecessor Ariston, he was fathered either by the hero Astrabacus (“the one with the mule saddle”) or by one of the household servants, the “guardian of donkeys” (6.68.2, 6.69.4). See Boedeker 1987a. For mules and kings, cf. my n. 62 in the present chapter.

Comparison

65

the Histories ardently desire this position or plot to obtain it,66 contract irregular marriages,67 or are guilty of treason for the sake of power or gain.68 The Spartan diarchy is literally different from other royal systems, but its significance in the logos rests largely on the fact that at Sparta, the men who make trouble truly bear the title of king.69 Through the particulars of Spartan history and prehistory, Herodotus’ narrative is able to shape the monarchical model in a special way that applies to the problem of leadership in the Greek states. In Herodotus, diverse pressures tear the Spartan diarchy. Given the nonindividualistic ideals that Sparta stands for (see, e.g., 9.71.3), the requirement to adhere to the ethos of the city as a whole is greater for these kings/nonkings than for the politically prominent elsewhere in Greece. At the same time, the inherited privilege of their office separates the diarchs from the rest of the city; they lie close to the dangerous sphere of autocratic transgression. The narrative connects the tension inherent to the status of the Spartan kings to the ambivalent legacy of their heroic ancestry, since the hero is both capable of the highest display of excellence and unfit to live with his peers.70 Thus, the closest Greek analogue to the despot Cambyses is the Heraclid furens Cleomenes. At the other end of the spectrum, we find the citizen-king Leonidas, in whose case the code of kingship exceptionally denies the monarchical model, just as the heroic code is able to transfer Homeric kleos to a hoplitic ethical context.71 Heroic antecedents within the monarchical model—emerge again with the introduction of Melampus as the archetype for Tisamenus (who through his own family of the Iamidae has a heroic connection of his 66. Examples are Theras (see 4.147.3), Dorieus (see 5.42.2), and Leotychides (see 6.65). 67. Examples are Anaxandridas (see 5.40.2), Ariston (see 6.62), and Demaratus (see 6.6.2). Eros for turannis (see, e.g., 1.96.2, 3.53.4) and unbridled sexual desire go together. See Hartog 1988, 330. “Doing violence to women” is one of the typical monarchical actions in the words of Otanes (3.80.5), amply illustrated in the narrative (see 1.8– 13, 1.61.1, 3.50.1, 5.92.η1– 3, 9.108– 13). 68. E.g., Leotychides takes bribes (see 6.72); Demaratus turns East (see 6.70.1– 2). 69. The issue of Spartan kingship in Herodotus is related to that of the paradoxical “foreignness” of this city. See “Implicit and Explicit Difference in the Ethnographies,” and “The Texture of Nomos” later in this chapter. 70. This is arguably one of the issues of Sophocles’ Ajax. See Knox 1961, 144– 48; Rose 1995. 71. The codes operative in Herodotus’ representation of the Spartan kingship are discussed in Munson 1993b. See also Boedeker 1987a. For the parallels between Cleomenes and the insane Cambyses, see Griffith 1988, 70– 71; Munson 1993b, 45 with n. 32. For the kleos of Leonidas, see chap. 3, “Specific Glosses of Interpretation,” in the present book.

66

Telling Wonders

own). In a minor counterpoint, Herodotus’ diviners and their heroic ancestors chime in with the kings to symbolize the relation of individual and state. This representation exploits the shared knowledge of a culture in which prophets and seers are close to the sources of power, predict and assist the rise of despots (1.62.4), influence policies (7.6.3– 4), compete with political leaders in deliberations (7.142.3– 7.143.1), and invent war strategies (8.27; cf. Thuc. 3.20.1).72 Like kings, seers, by definition, stand out from the citizen body. Normally outsiders, they become a part of the polis only to fulfill a leading public role. They are suited, therefore, to be taken as the doubles of kings, who also tend to claim special authority over prophecies.73 At the side of Leonidas at Thermopylae, we find as his equivalent Megistias the seer, who imitates the selfless action of the king by choosing to die with the three hundred; beside Leonidas, he alone is represented as making the choice (7.221). Just as Leonidas is the offspring of Heracles, so we are told that Megistias is from the stock of Melampus (7.221), that same Melampus who forces himself on Argos as king in the digressive narrative in the account of Plataea. A conflicting heroic heritage lies behind the present-day seers just as it lies behind the Spartan kings. The archetype Melampus produces two antithetical modern metaphors. One is Megistias, the citizen-seer, his descendant; the other is Tisamenus, who by becoming a citizen “imitates” Melampus, the king-seer. Implicit Analogy The Seers of Plataea In the Plataea narrative, the explicit analogy between Tisamenus and Melampus combines with an implicit but inevitable thematic comparison between the Tisamenus/Melampus doublet and a third Elean seer, eloquently named Hegesistratus, also connected with Sparta. While Tisamenus enters the logos of Plataea as the mantis of the Spartans and of the whole Greek side, Hegesistratus, an old enemy of the Spartans, now performs the sacrifices for the Persians. Hegesistratus’ appearance, like that of Tisamenus, gives rise to an expanded gloss of identification that explains how he came to fulfill his present role (9.37). Just as the unique72. Evidence for the importance of Greek seers in politics and war and for the sometimes problematic power relations between strategos and mantis is collected by Pritchett (1979, 46– 90; see 65– 70 for Sparta). 73. These include the Spartan kings: see 6.57.2, 5.90.2; cf. 5.93.2.

Comparison

67

ness of Tisamenus’ case justifies the inclusion of his story, so here superlatives advertise the narratability of the story of Hegesistratus. He belonged to another heroic family of seers, the Telliadae, and was a man “of the highest consequence” (9.37.1); at a time previous to the current events of Plataea, he performed “a deed that defies description” and “devised the boldest thing of all we know about” (9.37.2). Like the story of Tisamenus, that of Hegesistratus has an analeptic/ proleptic movement, since it begins at a time before the battle of Plataea and ends several years after it. Having suffered from this man many wrongs that remain unspecified, the Spartans had once captured and imprisoned him and were about to put him to death (9.37.1). But Hegesistratus cut off the instep of his foot to free himself from the stocks, escaped through a hole he had made in the wall of the prison, and by a painful march reached anti-Spartan Tegea. After his wound had healed,  he made himself a wooden foot and openly became an enemy (πολεµιος) of the Spartans. Hegesistratus’ hatred (ε χθος) against the Spartans did not in the end turn to his advantage, since the Spartans eventually captured him as he was prophesying in Zacynthus and put him to death (9.37.2– 4). The conclusion underlines the crucial elements of self-interest and hatred: “But the death of Hegesistratus occurred after the events of Plataea; at the time, he was sacrificing for Mardonius for no small compensation, and he did so with great eagerness on account both of his hatred [ε χθος] for the Spartans and his gain” (9.38.1). Tisamenus and Hegesistratus—the benefactor of Sparta on the Greek side at Plataea and the public enemy of Sparta serving the Persians—are antithetical figures. To the extent, however, that both dramatize friction between individual and community, they are also part of the same phenomenon. As a fugitive to the Persian side, Hegesistratus belongs to the group, well represented in the Histories, of those exiled or alienated Greeks who seek refuge in the East and in some cases “instigate incursions against their fellow citizens.”74 This type of actant includes, among others, the deposed Spartan king Demaratus (6.70, 7.3), the Pisistratids (6.94.1, 6.102, 6.107.1, 7.6.2– 5), and Themistocles (8.109.5)—all individuals who try to impose their will on a Greek community and are therefore, in this respect at least, also Tisamenus’ analogues. The Spartans capture Hegesistratus in Zacynthus, where they catch up with the fugitive Demaratus. Hegesistratus’ self-mutilation parallels the action of 74. Boedeker 1987a, especially 191– 92.

68

Telling Wonders

another royal enemy of Sparta, Cleomenes, who is also put in the stocks, obtains a knife, and proceeds to cut himself, though in a more selfdestructive way.75 In the intersection of the patterns and in the partial overlap between the otherwise opposite Tisamenus and Hegesistratus, the idea emerges that the enemy of a city will also potentially try to become its ruler; conversely, an individual who rises to power and benefits the city may become its ruler or its enemy. The last royal Spartan of the Histories to be implicated in these different stages of the monarchical model is the king-regent Pausanias. In the extratextual aftermath of Plataea, Pausanias became, like Hegesistratus at Plataea, a wanted public enemy of the Spartans, eventually suffering death at their hands (Thuc. 1.131.1, 1.134). A fifth-century Greek audience would have regarded Pausanias as a tyrannical type without any help from Herodotus. His had been a cause c´el`ebre, in which Athens and Sparta, for different reasons, had found themselves in mutual agreement in condemning the hero of Plataea.76 The result was probably a basically familiar story (though embodied in different versions somewhat varied in intensity and detail), about a “good” Pausanias before and a “bad” Pausanias after Plataea.77 Thucydides speaks of Pausanias’ eagerness to dominate Greece (Thuc. 1.128.3); he quotes a letter where Pausanias promises to Xerxes to make Greece subject to him and proposes to marry Xerxes’ daughter (Thuc. 1.128.7). The story evokes and perpetuates a special model of degeneration within the broader monarchical model.  Just as Herodotus’ Cleisthenes “imitates” (ε µιµεετο) the tyrant Cleisthenes, so in Thucydides the generalship of Pausanias after Plataea is “an imitation of tyranny” [τυραννι δος µι µησις] (Thuc. 1.94.3). Thucydides’ description of Pausanias’ brief despotic tenure at Byzantium includes predictable indices, what he calls “small matters that displayed what he wanted to do in the future on a larger scale”: inability to live in the 75. The parallels between Hegesistratus, on the one hand, and Cleomenes (see 6.75.2– 3) and Demaratus (see 6.70.2), on the other, are noticed by Macan (1908, 1.2.673, 675). 76. Fornara (1966, 266) argues that for the Athenians, the alleged aberrancies of Pausanias served to rationalize their taking over the leadership of the Greek allies; the Spartans, for their part, had put Pausanias to death (perhaps because they had evidence of his plotting with the helots; see Thuc. 1.132.4) and needed to justify that action. 77. Evans (1988) reconstructs two versions of the story of Pausanias, one corresponding to Thucydides’ account, and the other, somewhat less unfavorable, reflected in Herodotus. There is no evidence, however, of a favorable version of the story of Pausanias’ behavior after Plataea.

Comparison

69

established style, Median dress, a Persian table, foreign bodyguards, and an unapproachable temper (Thuc. 1.130.1– 2; cf. Thuc. 1.95.1). The narrator of the Histories communicates the relevance of the Pausanias story—and implicitly testifies to his audience’s familiarity with it—in two explanatory glosses in different contexts. Once, he offhandedly mentions “the hubris of Pausanias” in his interpretation of what happened after the Persian invasion (8.3.2). Another time, he brings up Pausanias’ lust for power and his Eastern marriage (though a different marriage than the one in Thucydides’ reported letter), in a gloss of identification for Megabates, “a Persian of the Achaemenid family, cousin [of Artaphrenes] and of Darius, the one to whose daughter at a later time, at least if the story is true, the Spartan Pausanias, son of Cleombrotus, arranged his betrothal, having conceived a passion [ε ρωτα] to become the tyrant of Greece” (5.32).78 Since Megabates commands the expedition against Naxos that Aristagoras has persuaded the Persian king to undertake for reasons of his own, the mention of Pausanias’ ties to the Persians via this same man reinforces the message about disloyal Greeks. In the narrative of Plataea, however, Herodotus evaluates Pausanias as the author of the “fairest victory we have ever known” (9.64.1) and surprises his audience by his apparent determination not to mention subsequent events. But the “Pausanias before” protests too much here and in terms better suited to remind everyone of the “Pausanias after” than to correct public fame.79 Confronted with three monarchic temptations of the type that Thucydides says he yielded to shortly afterward— mutilation, luxury, sexual indulgence (9.78– 79, 82, 76)—Pausanis selfrighteously rejects them all and declares himself at one with his fellow citizens (9.79.2). Though his refusal to defile the body of Mardonius is laudable, his contemptuous distinction between what a barbarian and a Greek would do contradicts the ideological thrust of Herodotus’ text and sounds excessively self-assured.80 Pausanias’ alleged contemplation of Persian riches and his experimental comparison between a Persian and a 78. See n. 67 in the present chapter: the two forms of the tyrant’s eros here become confused. The joke is noticed by Macan (1895, 1:176) and Shimron (1989, 66 n. 35). 79. Fornara (1971a, 62– 66, especially 64) calls Herodotus’ portrayal of Pausanias “a masterpiece of irony and a harbinger of tragedy.” Cf. Evans 1988, 4– 8. 80. See the narrator’s gloss declaring that Xerxes’ mutilation of the body of Leonidas was against Persian custom (7.238.2). See also Payen 1997, 174– 75. In his narratorial voice, Herodotus avoids explicit comparisons between Greeks and barbarians of the type made by his Pausanias.

70

Telling Wonders

Spartan banquet occur in a context of fascinated curiosity, ill-concealed greed, and ambiguous laughter.81 All this is not merely a favorable portrayal but rather a pretend inversion of the monarchical model that has been applied to all other Spartan kings in the Histories except for Leonidas. Like his seer Tisamenus, Pausanias gains a splendid victory for Greece. Like the seer Hegesistratus, however, he is already poised to become the accomplice of Persia against Sparta and all the Greeks. Thus, in antithesis to the pair Leonidas and Megistias at Thermoplyae, the diviners of Plataea on either side synthesize the ambivalence of Pausanias. As they do so, they symbolically carry forward and broaden Herodotus’ messages on the disharmonious combination of leadership and citizenship and on the dangers of prominent individuals for the city-states of Greece. The Last of the Seers Two more diviners appear at the end of the Histories in similar fashion as Tisamenus and Hegesistratus and demand association to the same analogical network: De¨ıphonus, who performs the sacrifice for the Greeks at Mycale; and especially Euenius, his father, “to whom the following thing happened” (9.92.2). At the end of the story of Euenius, the resumptive summarizing statement is followed by a gloss that undermines even this secondary connection to the main narrative: “but now I have also heard this: that De¨ıphonus . . . was not the son of Euenius” (9.95).  Just as Tisamenus and Melampus demand a special reward (µισθος, 9.33.3, 4; 9.34.1) and Hegesistratus sacrifices for no small salary  ο λι γου, 9.38.1), so De¨ıphonus plies his trade all (µεµισθωµενος ουκ  Ελλαδα ε ργα, 9.95). The narrative over Greece (ε ξελαµβανε ε π ι τ ην concerning Euenius separates the issue of compensation from the profession of seer, but it maintains the theme of the individual’s blackmail of the city by translating it into an ethical and juridical question of dike (justice). Euenius was one of the prominent citizens of Apollonia appointed for a year to guard at night a special flock that by daytime grazes along the banks of the local river. The flock is sacred to the Sun, and the people of Apollonia, who greatly revere it “on account of some prophecy,” give it 81. Fornara (1971, 62– 63) notices Pausanias’ interest in the spoils (at 9.80.1 and 9.82.1), and see 9.82.2. The reference to the helots at 9.80.1 may be an allusion. Pausanias’ laughter (9.82.3) is a negative index. See Lateiner 1977, especially 177.

Comparison

71

shelter at night in a cave. One night, during his turn as guard, Euenius fell asleep and about sixty of the sheep fell prey to the wolves. The citizens of Apollonia brought Euenius to trial and condemned him to be deprived of his sight. After the sentence was carried out, however, both the animals and the land became barren, so the Apolloniates consulted the oracles of Delphi and Dodona. Here they learned the reason for their calamity: they ι κως) blinded the guardian Euenius. The gods themselves had unjustly (αδ had sent the wolves; they would not, therefore, cease avenging Euenius until the people of Apollonia would make reparation by giving him any  thing he would choose and consider fair (ου προτερον παυσασθαι           αν  τιµωρεοντες ε κεινω  πρ ιν η δικας δωσι των ε ποιησαν ταυτας τας  ος εληται κα ι δικαιοι , 9.93.4). The gods, for their part, would grant αυτ Euenius a gift that would make many men regard him as fortunate. The people of Apollonia concealed this response and entrusted the matter to a group of fellow citizens, who approached Euenius. Sitting next to him on a bench, these men started speaking of various things until the opportunity came for casually asking him what reparation he would choose if the people of Apollonia should be willing to compensate him for  ελοιτο, ε ι ε θελοιεν  what they had done (τι να δι κην αν Απολλωνιηται    ε ποι ησαν, 9.94.1). Euenius, who knew δι κας υποστηναι δωσειν των nothing of the oracle, said that the gift of two fine estates and the finest house in town would constitute for him adequate compensation (δι κην   γενοµενην,  ο ι ταυτην αποχρ αν 9.94.2). As soon as he had finished saying that, those sitting at his side replied: “Euenius, the people of Apollonia pay you this compensation [δι κην] in accordance with the oracles.” When he learned about the response, Euenius made a great fuss because he felt he had been deceived. But the Apolloniates bought the properties he had chosen from their owners and gave them to him, and from that day on he also had prophetic powers and became famous (9.93– 94). I have paraphrased this story rather fully to convey its remote and idealized setting. The flock sacred to the Sun, grazing on the shores of the nameless river during the day and sheltered at night in a cave, evokes a primordial and mythical atmosphere of a community close to the gods. As with the cattle of the Sun in the Odyssey (12.127– 33),82 the violation of this herd creates a crisis. The wolves attacking the cattle, the trial, and the elders on the bench next to Euenius seem to have come out of the 82. See Vernant 1989b.

72

Telling Wonders

Homeric ecphrasis of the shield of Achilles, where two lions devour two oxen from the herd that pastures by the river, and where the elders sit in a circle on polished stones in the judgment scene (Il. 18.573– 81, 503– 4). The punishment sent by the gods to the Apolloniates for their unjust treatment of Euenius recalls the calamities by which Zeus avenges transgressions of Dike in Hesiod’s Works and Days: “great suffering, famine, and plague at once; the people perish away, women do not give birth, and households are diminished” (242– 44). In this archaic paradigm of the polis, where good government and communal deliberation prevail, a situation arises that, as in the stories of Tisamenus and Melampus, places an individual in a position of posing a threat. Here prophecy itself is a gift and a compensation from the gods, not a service in exchange for which compensation is due. But just as Tisamenus demands Spartan citizenship in exchange for his needed services as seer, Euenius, supported by prophecy of the oracle, could have demanded from the Apolloniates excessive compensation for his blindness. His reaction after learning about the oracle shows that he would have indeed done so. Kingship is not mentioned as a possibility for this price, but the earlier case of Melampus and the gift that Euenius actually requests (a choice portion of land) point in that direction.83 Unlike Tisamenus and Melampus, however, Euenius is never allowed to exceed his political status. A group of citizens delegated by the people keeps control of the negotiations and manages to correct the earlier miscarriage of justice according to divine injunction; at the same time, they preserve the city. The gods apparently do not object to the deceit, and Euenius settles down with their priceless gift of divination and with a reasonable human prize, neither of which violates the city’s institutions. He remains an ordinary citizen but is famous for his prophecy and richer than before; his new material possessions are bought at public expense, and the transaction wrongs no one. The issue of reward in this story, corresponding to the misthos (salary) of the seer in the other narratives, centers around the juridical, political, and religious notion of rightful balance, a Dike of Hesiodean stamp. 83. Cf. the “many gifts of land” obtained by Callias of Elis, another seer of the family of the Iamidae, as a reward for helping the Crotoniates in their war against Sybaris (5.45.2). Since the only other outsider who is supposed to have helped the Crotoniates in this war (though they deny it) is Dorieus, a royal Spartan, this points again to the Greek cultural notion of a metaphorical equivalence between seer and king.

Comparison

73

Unlike the former stories of seers, moreover, this one is pervaded with the words and actions of the gods—the prophecy regarding the cattle, the divinely induced curse on the land, the prophecy about Euenius’ compensation, and the gods’ gift of prophecy to Euenius. The gods supervise the human legislators and judges but leave them to their own devices in managing themselves politically. This archaic morality tale about remote Apollonia offers a vision of the righteous city and a hypothetical solution to the problem of the individual’s personal power and privileged status in the Greek polis. I have ranged far and wide on the basis of Herodotus’ two explicit analogies. In the historical narrative, glosses of comparison constitute the exception, and the conveyance of meaning does not depend on their presence. Both the explicit analogy of the two Cleisthenes and the Tisamenus/Melampus parallel, out of a mere handful of this type of interventions, confront a central issue in the Histories, that of kingship, and point to its broad metaphorical application. These rare visible stitches in a far broader analogical weave reveal the thought processes of the narrator and confirm a host of implicit analogical associations, horizontal and vertical, that the audience would have made elsewhere without prompting. Comparison in Space Although both history and ethnography advertise themselves as the report of extraordinary facts, the discrepancy between the two forms of discourse is perhaps nowhere more conspicuous than in the way in which they approach qualitative comparison. In the account of past events, particulars become mutually linked through implicit analogy; but comparison operates predominantly several levels below the surface, and the reconstruction of analogical networks is almost entirely dependent on the interpretive operations of the listener. The report of existents in faraway lands, by contrast, is from the start based on the qualitative comparison between what is to be found “over there” and what belongs to the familiar horizon of narrator and audience. In ethnography, therefore, and to a lesser extent also in geography, comparison is very much on the surface of the text even when it is implicit. Frequently it is explicitly the object of discussion.

74

Telling Wonders

Implicit and Explicit Difference in the Ethnographies Ethnographic description is designed to communicate and enhance the idea of difference.84 History follows a certain chronological and causal development, but not every historical narrative statement places the uniqueness of the fact it records at the center of attention. In ethnographic description, by contrast, relating a fact is generally a declaration of difference with respect to some other fact that the audience assumes to be in the normal order of things. They hunt locusts, and after they catch them they dry them in the sun and grind them, then they sprinkle the powder on milk and drink it. (4.172.1) Because ethnography is primarily a description of difference, it frequently expresses itself in negative statements that contradict the cultural or geographic norms to which the audience is accustomed.85 Similarly, advertisements of narratability in an ethnographic description tend to reinforce the narrative’s built-in raison d’ˆetre of reporting differences from anything Greek. These then are the customs [of the Thracians] that are most noteworthy. (5.6.2)  The customs qualified as ε πι␸ανεστατοι in the conclusion just quoted include almost complete inversions with respect to the Greek norm. For example, “the Thracians do not guard their young girls [notice the negative form] and allow them to have intercourse with any men they wish, while they strictly confine married women; they acquire the women they marry from their parents at the cost of great riches” (5.6.1); “Tattoos are considered a sign of nobility, not to be tattooed a sign of low birth” (5.6.2).86 84. See, e.g., Od. 6.119– 20, 9.105– 15. Redfield (1985, 99– 100) describes Herodotus as a tourist avid for difference. 85. For more on ethnographic negations, see chap. 3, “Cognitive Statements and Polemical Negations.” In the historical narrative, the statement that someone did not do something is an interpretive gloss that more generally contradicts the audience’s expectations according to their contextual or extratextual knowledge. 86. Inversion as a way of explaining the world “over there” is discussed by Rosellini and Sa¨ıd (1978, 985– 91) and Hartog (1988, 212– 16).

Comparison

75

Only exceptionally does ethnographic description need to state difference from Greece by explicit means.87 More frequent are Herodotus’ statements of absolute uniqueness: he refers to “the only men in the world,” “the only region on earth,” and a climate “not as among other men.”88 The incomparable natural phenomenon par excellence, both in a qualitative and in a quantitative sense, is the Nile. By virtue of some special power of its own, the Nile “is naturally opposite” (τα ε µπαλιν  πε␸υκεναι) from other rivers, since it spontaneously floods in summer when all other rivers are dry, making the Egyptians into a people who “do not plow.”89 The Nile’s unparalleled nature causes it to be something other than a river, more similar to a sea, with cities emerging on the  surface, “very similar in a way” (µαλιστα κ η ε µ␸ερεες) to the islands on the Aegean (2.97.1). Here the Egyptians sailing across the flooded plain and the Scythians driving their cart on the frozen sea (4.28.1) are complementary visions of the extraordinary.90 Implicitly but unmistakably, Scythia is the polar opposite of Egypt; and each of the two is not only different from Greece but utterly unique.91 In correspondence with the uniqueness of the nature of Egypt and its river, Herodotus declares the absolute uniqueness of the Egyptians in the programmatic introduction to the ethnographic part of the logos. The statement begins with a celebratory gloss that proclaims that Egypt is more worthy of narration than any other region because it has “the most wonders and works superior to description” (2.35.1). The comparative glosses in the next sentence explicitly connect this pure advertisement of narratability to the fact of Egyptian qualitative differences in the two 87. It does, e.g., at 1.131.1, with polemic force (see chap. 3, “Cognitive Statements and Polemic Negations”). 88. See 3.107.1 (Arabia is the only land to produce incense, etc.); 4.184.1 (the Atarantes are the only men we know who have no individual names); 2.104.2 (the Colchians, Egyptians, and Ethiopians are the only men who practice circumcision); 2.68.3 (uniqueness of the crocodile); 3.113.1 (unique sheep); 3.104.2– 3, 4.28.2 (uniqueness of climate; see Hartog 1988, 29– 30). 89. 2.19.2– 3 (cf. 2.24.1; 2.14.2; 2.12.2). Also, with respect to the abundance of its resources and volume, “no other river can be compared with it” (4.53.1; cf. 2.10.1– 3). See also 4.50.1, comparing the size of the Ister with that of the Nile. 90. Cf. 2.92.2. See Corcella 1984, 71. 91. On the implicit Egypt-Scythia polarity in Herodotus, see Redfield 1985, 106– 7. Its principal terms are hot/cold, no rain/rain in summer, changeable Nile/unchanging Ister, one river/multiple rivers, old ethnos/young ethnos, culture/nature, complex customs/simple customs, many wonders/few wonders, immobility/nomadism, centralized monarchy/multiple kings. The pairing of Egypt and the North occurs already in Pind. Isthm. 6.23.

76

Telling Wonders

specific areas of nature of the land and culture of the people. They establish, in fact, a double polarity. The Egyptians, at the same time as they have a CLIMATE that is different [ε τεροι ω  ] and the RIVER, which has a NATURE unlike  [␸ υσιν αλλο ι ην] that of the other rivers, [so] for the most part established LAWS AND CUSTOMS that are the complete opposite [πα ντα ε µπαλιν] to those of the rest of mankind. (2.35.2) The preliminary list of about twenty-three customs that immediately follows (2.35.2– 2.36.4) is designed to illustrate what the Egyptians customarily do παντα ε µπαλιν (completely opposite) not just from the Greeks (as is implied of the Thracians at 5.6, cited earlier) but from the rest of # mankind, ο ι αλλοι ανθρωποι (or ο ι αλλοι, ωλλοι). This declaration of utter Egyptian difference gives the audience a jolt by establishing a new subdivision of the world, in which the Greeks become marginalized. The opposition between the Egyptians and everybody else—rather than, in the usual way, between the Greeks and the non-Greek world, Egyptians included—is a part of Herodotus’ polemic against both the Greeks’ sense of being a special nation (even special in the sense of their being the exclusive representatives of normalcy) and the dismissive attitude they affect toward Egypt in particular.92 No gloss in the Histories proclaims the uniqueness of the Greeks, and only two passages attribute to all barbarians a nomos that the Greeks do not have.93 In a rare instance where Herodotus attributes the same nomos to all barbarians, the statement highlights the similarity between the Spartans and the barbarians.94 The Histories both presuppose as a given and discourage the commonplace notion of a Greek/barbarian polarity. Other glosses expressing the cultural uniqueness of a people vis-a-vis ` ο ι αλλοι ανθρωποι almost exclude a few remote ethnea from the mainstream of 92. Lloyd 1976, 310. 93. At 8.105.2, the barbarians of Asia are meant. The gloss at 1.10.3 (“Among the Lydians, and among almost all barbarians, even for a man to be seen naked brings great shame”) is in its context designed to emphasize similarity a fortiori rather than difference between barbarians and Greeks. On the quantitative evaluative comparison at 1.60.3 (Greeks more intelligent than barbarians), see chap. 3, “Explicit Evaluation.”  ος]  as 94. “The custom of the Lacedaemonias at the death of their kings is the same [ωυτ that among the barbarians of Asia. For, as a matter of fact, most of the barbarians follow the same custom at the death of their kings” (6.58.2). Cf. n. 161 and corresponding text in the present chapter.

Comparison

77

humanity, be it Greek or non-Greek: “their language is like no other because they squeak like bats” (4.183.4); certain peoples of the Caucasus and India “couple in the open like cattle.”95 Scholars have identified several systems that underlie Herodotus’ overarching conception of world differentiation. These schemes reflect our modern understanding of partially inherited forms of classification, which enabled the Greeks to define their own identity by supplementing and interpreting what they knew about exotic cultures in an unconsciously organized way. Thus, Rosellini and Sa¨ıd have traced the structural correspondences among the sexual, dietary, and religious customs that Herodotus attributes to the most remote and primitive peoples he describes. This demonstrates the consistency with which these three areas of culture are made to diverge from the Greek norm of monogamous marriage, on the one hand, and the consumption of cooked and cultivated cereals and boiled or roasted meat in the ritual context of sacrifice, on the other. The three marginal and mutually distant peoples said to “couple like cattle,” an index of extreme primitivity, also have alimentary practices that are equally abnormal in each case. Their sexual promiscuity goes together with a diet that diverges either on the side of omophagy (consumption of raw meat or fish, cannibalism) or on the side of vegetarianism (uncultivated, uncooked herbs and cereals).96 Redfield formulates a somewhat different set of principles by subdividing Herodotus’ peoples into “hard” and “soft” cultures. The first are represented by Scythians, Massagetae, Thracians, Ethiopians, and, relatively speaking, Greeks; the second by Egyptians, Babylonians, Lydians, and, relatively speaking, Persians. Soft peoples are characterized by luxury, the division of labor, and complexity of nomoi, especially in the sphere of religion; hard peoples are simple, harsh and fierce. Among soft peoples market exchange proliferates; hard peoples rely on gift and theft, the heroic mode of exchange. Soft peoples centralize resources through 95. 1.203.2 (κατα περ τοισι προβατοισι), 3.101.1; cf. 4.180.5. See Hartog 1988, 226. In two cases, glosses of absolute uniqueness draw attention to the individuality of the Greeks’ close neighbors, Caunians and Lycians (1.172.1, 1.173.4). Uniqueness is also attributed to the Ethiopians, in the same terms as to the Egyptians (3.20.2) and, within Persian society, to the Magi (1.140.2). 96. See Rosellini and Sa¨ıd 1978, especially 955– 60. On the Greek dietary norm, see Detienne and Vernant 1989, especially the essays by Detienne, Vernant, and Durand.

78

Telling Wonders

taxation, build monuments, are literate and organized; their politics tend toward tyranny. Hard peoples have relatively weak political organizations and tend toward anarchy. Soft peoples tend to acculturate the dead, hard peoples to naturalize them; among hard peoples women are treated as an abundant natural resource, more or less freely available, whereas among soft peoples women tend to become a commodity, disposed of by sale, through prostitution, or otherwise. Hard cultures fall short of civility; they are unwelcoming and difficult to visit. Soft cultures are confusing and seductive, difficult to leave once visited.97 This brilliant synthesis sheds light on the links between ethnography and history by refining Hellmann’s pattern of the “primitive opponent.” In Herodotus, in fact, no “soft” people conquers a “hard” people; “hard” people remain free or even conquer their “softer” aggressors.98 Other implicit systems have been detected as determining Herodotus’ description of the world. According to the conceptual map devised by K. Muller, ¨ for example, high cultures are placed in the center of the oikoumene (Egyptians, Babylonians, Lydians, Persians, Greeks), and primitivity increases as one proceeds outward, with builders, cattle raisers, and hunters/gatherers distributed in concentric circles toward the edges.99 These conceptual reconstructions provide us with a vocabulary and fundamental frameworks for discussing Herodotus’ ethnographic material. We should recognize, however, that they sometimes constitute devices by which we attempt to make sense of the apparent disorder of Herodotus’ description, more than actual distillations of the way in which Herodotus represents the world. Structuralist studies, such as those of Rosellini and Sa¨ıd, are especially prone to shift from an analysis of the discourse of Herodotus to a discussion of the mythical forms of thought that inform his material.100 The different schemes that have been proposed, moreover, cohere with one another only up to a point. They sometimes break down altogether and fail to account for numerous narrative details and metanarrative interventions, because Herodotus both is subject to cultural ways of thinking and rebels against them, is both a 97. 98. 99. 100.

Redfield 1985, 109– 10. See Redfield 1985, 112– 13; Hellman 1934, 80– 89; Cobet 1971, 104– 20. Muller ¨ 1972, 121– 22. See also Romm 1992, 46– 47. See, e.g., Mora’s criticism (1985, 60– 72) of Rosellini and Sa¨ıd 1978, 962– 66.

Comparison

79

lover of symmetry (as Redfield and others have maintained) and contemptuous of it. Differentiating from Within When glosses express dissimilarities between two foreign peoples, rather than uniqueness in an absolute sense, they emphasize what a complicated and irregular place the world is. When they mark differences between ethnea that live in the same general area or tribes belonging to the same ethnic group, they seem especially designed to discourage schematization. Based on the ethnographer’s detailed knowledge of people and places, the narrative reveals to the audience that difference manifests itself in ways they might not expect. The common stereotype of rudeness, primitivity, and poverty the Greeks frequently attached to the words Scythian or Thracian (names for “hard” cultures according to the modern critical notion) is, for instance, shown to be inadequate in the case of the Agathyrsi, who are neighbors of the Scythians but “extremely luxurious and wearers of gold” (4.104). The Agathyrsi also “practice the community of women, so that they may be brothers of one another and, being all related, not have mutual envy or hatred.” “In their other customs, however,” notes Herodotus, “they come close to   the Thracians [Θρηιξι προσκεχωρηκασι]” (4.104). The portrayal of the Agathyrsi forces us to a further distinction. In the poetic and ethnographic traditions about remote peoples, Lovejoy and Boas have long ago identified “hard primitives” and “soft primitives,” a categorization that Romm has recently expanded.101 The first group includes the tough and relatively poor Scythians, Thracians, most Libyans, and the extra-Herodotean Arimaspi. The second—Herodotus’ Argippaeans, or Bald Men; the Ethiopians; the Hyperboreans of tradition—are the simple but joyful cultures blessed with Golden Age abundance and a natural state of peace and justice.102 Presenting an idealized alternative to the phthonos (envy) among citizens in a Greek polis, the communal life of the Agathyrsi fits into this traditional type.103 But the expression that    οτατοι describes their prosperity— αβρ . . . κα ι χρυσο␸οροι [extremely 101. Lovejoy and Boas 1935, 287– 90; Romm 1992, 47– 81. 102. See Lovejoy and Boas 1935, 1– 15; Rosellini and Sa¨ıd 1978, 962– 66. 103. On the community of women and children in Greek utopia, see Aristoph. Eccl. 613– 35; Plato Rep. 463c– 465b, criticized by Arist. Pol. 2.1.4– 18 (1261b16– 1262b37). For Greek traditions about exotic family customs, see Pembroke 1967.

80

Telling Wonders

luxurious and wearers of gold]—does not suit idealized Golden Age  connotes cultural refinement, even a degree  ος savages. The adjective αβρ of effeminacy; it is only used to qualify advanced cultures (“soft” in Redfield’s sense of the word) and in particular the notoriously delicate Lydians.104 The resulting combination of three different stereotypes (hard primitives, soft primitives, and soft civilized people) creates a sense of unexpected and asymmetrical difference. In Libya, the nomads east of Lake Tritonis and the nonnomads west of it “do not have the same customs” (4.1871). Beyond the borders of Scythia, the rule of extremes at the edges accounts for the presence of the most just and peaceful of men (4.23) and, to the southwest of these, of the Androphagoi, a “peculiar people” that ignores the very notion of justice and “the only ones . . . who feed on human flesh” (4.18.3, 106). At the opposite extremity of the earth, the Libyan “Country of Wild Beasts” contains “the wild men and the wild women” mentioned at the end of the list of monstrous animals (4.191.2– 4). Yet this is also the home of the Maxyes, who wear their hair long on the right and shaven on the left and paint their bodies red: they, at least, practice agriculture, “own houses by custom,” and “say that they are descendants from the men who came from Troy” (4.191.1). The civilized features of this ethnos and their link to the heroic splendor and the past of the Greeks throws somewhat out of kilter the symmetrical correspondence with the Androphagoi on the basis of bestiality.105 Herodotus’ explicit comparisons reinforce the message of irregular differentiation both with and without the presentation of differentiating material. Though neighbors of the Scythians, the Melanchlainai are αλλο  [a different people, not Scythian].106 The narra εθνος κα ι ου Σκυθικον tive does not substantiate the gloss in this case, and elsewhere another gloss assimilates the customs of the Melanchlainai to those of the Scythians, aside from the former’s black cloaks (4.107). Also, the uniqueness of  “of the delicate feet,” when she advises him to 104. The Pythia calls Croesus ποδαβρε,  or good”  ον flee the Persians (1.55.2), who are still a “hard people” and “have nothing αβρ (1.71.4). For habrosune outside Herodotus, see Nagy 1990, 282– 85. 105. Pace Rosellini and Sa¨ıd 1978, 960. Herodotus records, without explaining them, connections between the primitives of central Libya and Greek traditions (4.178, 179, 189, 190). At 4.170 and 4.180, he speaks of customs borrowed from the Greeks, in the first passage specifically from Cyrene. 106. 4.20.2 (contra Hecataeus, FGrHist 1 F 185, who apparently defined them as  ε θνος Σκυθικον). The general tendency to group all northern barbarians in the category “Scythians” is noted by Strabo (11.6.2; cf. 7.3.9, quoting Ephorus, FGrHist 70 F 42).

Comparison

81

the Tissagetae, “a numerous and peculiar” people who live by hunting, remains elusive (4.22.1). Fully documented is the distinction between Budini and Geloni, who live in the same area but “do not at all have the η γλωσσ  η . . . ουδ ε same material culture or the same language [ου τ η αυτ η]”  and are “not at all similar [ οµοιοι]” in appearance and διαι τα η αυτ coloring (4.109.1). The red-haired and blue-eyed Budini are autochthonous nomads, while the Geloni (“which the Greeks wrongly call Budini”) are settled agriculturists with language and customs that correspond to their Greek origins.107 But an unpredictable peculiarity of the Budini   ταυτ  η] arises here: they are “the only ones in that region” [µουνοι των 108 who eat lice. Another explicit reference to the misconceptions of the general public concerns the distinction between Scythians and Massagetae, “which some say is a Scythian ethnos” (1.201). The Massagetae “wear the same dress and have the same material culture as the Scythians” (1.215.1), but in the area of social organization, the custom of having wives in common is only their own, “for what the Greeks say that the Scythians do, it is not the Scythians who do it but the Massagetae” (1.216.1). Isolated differences of various kinds also exist among tribes that belong to the same large ethnos.109 Glosses of similarity of the type “[The Gilgamae] have about the same customs as the others” (4.169.2) are not merely fillers that “make it possible to extend knowledge by moving from one group to its closest neighbors.”110 Rather, the ethnographer intervenes to say that neighbors share the same customs because he does not allow his audience 107. 4.109.1. Language is a frequent criterion of internal differentiation in this and  other parts of the world (see 4.106: γλωσσαν . . . ι δι ην; 4.23.2; 4.108.2; 4.117; 3.98.3; 1.57.3; 2.42.4; 2.105; 6.119.4; 7.70.1; 7.85.1). 108. 4.109.1. The statement has the same effect if we accept the interpretation of Stein,  who, on the basis of Photius’ definition of ␸θει ρ, understands ␸θειροτραγεουσι as “they eat pinecones.” But see, e.g., Macan 1895, 1:78, and cf. Strabo 11.2.1, 14, and 19. The case (mentioned again shortly) of the Libyan women “biting lice off themselves” (Hdt. 4.168.1) not only is decisive for the meaning of the word in this passage but also represents a pendant to this passage according to Herodotus’ procedure of balancing differences and similarities: some people eat lice; some people only bite them off. 109. They exist, e.g., among the Thracians: see 5.3.2– 5.4.1. In Herodotus, the word ethnos is used both in this sense (see, e.g., 4.197.2) and to denote smaller subdivisions within the same ethnic group. See Jones 1996; Hall 1997, 34– 40. The differentiation of nomoi within a single unitarian people (e.g., not all the Egyptians venerate the same gods in the same way; see 2.42.1) or within a system (e.g., there are different Egyptian practices concerning extraction of entrails depending on the festival; see 2.40.1) gauge the internal complexity of the system. 110. Hartog 1988, 226.

82

Telling Wonders

simply to assume that this is the case. Concerning the Adyrmachidai, who are the first group of Libyans one meets proceeding westward from Egypt, he relates that they (1) “generally follow Egyptian customs” but (2) dress “like the other Libyans” and (3) have a few customs all their own: they are “the only ones of the Libyans” whose women get rid of lice by biting them off themselves and throwing them away and “the only ones” who present to their kings the young girls to be married (4.168.1– 2). This passage illustrates Herodotus’ painstaking recording of differences and congruences among tribes that live in the same general area and his individualized portrayals of less well known ethnic groups.111 Explicit Sameness and Analogy The Texture of the Earth If ethnography and geography are first and foremost the representation of what is different for the listener, and if Herodotus seems particularly concerned with internal differentiation as well, what role does similarity play in his account of foreign lands and peoples? “The experience of difference,” says Corcella, “if it is to be a real experience with its level of intelligibility, cannot do without a certain recognition of similarities to the world of habitual experience. It is precisely such recognition which makes comparisons and translations possible.”112 In the case of Herodotus, however, we should go one step further and even talk of an active pursuit of the similar, which counterbalances his observation of difference in all areas of his ethnographic and geographic research.113 Explicit comparisons that establish that something is like something else are frequent and of many kinds, and they employ an extensive metanarrative vocabulary of similarity. The most interpretive or speculative ethnographic comparisons are occasionally accompanied by self-referential 111. Cf., e.g., 4.180.1 (“While the Machlyes grow their hair at the back of their heads, the Aseans grow it in front”), 4.178 (“Next to the Lotophagi . . . are the Machlyes, who also consume lotus, but less at any rate than those mentioned above”), 1.173.4, 4.17, 5.4.1. See also the catalogue of Xerxes’ forces (7.61– 95) in reference to the equipment of the various national contingents (7.62.1, 7.63, etc.), where the balance is between what is  ο (the same) or belongs properly to a different ε πιχ ωριον (strictly local) and what is το αυτ people. 112. Corcella 1984, 74. Herodotus’ practice of noticing correspondences is also emphasized by Muller ¨ (1972, 116). 113. Gould (1989, 11– 13) notices the physical dimension of this pursuit: the histor  travels to Thebes to verify whether the traditions there would correspond (συµβησονται) with those in Memphis (see 2.3.1; cf. 2.44.1).

Comparison

83

signs, especially the verb συµβα λλω.114 Two things that belong to different real or narrative worlds can be “put together” with respect to size or the degree to which they possess a certain feature, if their similarities in other ways furnish grounds for the comparison. Or two things can be compared qualitatively, with the reservation that excessive quantitative discrepancy in size may make the comparison invalid.115 If one is unable to “put together” (sumballein) something unfamiliar with something already known, the foreign object may appear strange indeed and aporia ensues.116 The ethnographer Herodotus is, however, a master both at recognizing difference and finding likeness. Excessive difference in nature, especially if not properly corroborated by autopsy or reliable verbal testimony, is rejected out of hand.117 For things that can be verified, geography in particular preserves the balance between the amazing individuality of a particular phenomenon and the need to integrate it into a unitarian world ruled by uniform natural laws. While Herodotus overstates his presentation of the Nile as the incomparable river/nonriver with a different phusis (nature), he nevertheless attributes its uniqueness to natural factors susceptible to observation, such that they either actually affect this or that other river as well or would do so if they occurred elsewhere.118 Herodotus’ assertions that the Nile “cannot be 114. Ε ι κα ζω as a self-referential metanarrative term in the sense of “liken” only occurs in the history at 9.34.1, where, if the reading of the passage is correct, it signals a vertical analogy that has almost the boldness of a metaphor. See n. 54 and corresponding text in the present chapter. In the narrative, ε ι κα ζω designates Gelon’s metaphor of the spring that has been taken away from the year (7.162.2), the assimilation of snow with feathers made by the Scythians (4.31.2), and the Ionian assimilation of Egyptian crocodiles with lizards (2.69.3). The imprecision of the operation indicated by ε ι κα ζω in the sense of “liken” is also present in the speculative ε ι κα ζω in the sense of “guess,” “imagine,” or “suppose,” which is a gloss of opinion used to mark a gloss of interpretation. 115. The narrator apologizes twice for putting together (συµβαλλειν) qualitatively things with widely discrepant size (2.10.1, 4.99.5). 116. For putting things together (συµβαλλειν) as a source of understanding (with συµβαλλεσθαι, “conjecture,” in glosses of opinion), see Hohti 1977. 117. E.g., the difference of the Arimaspi is rejected at 3.116.2: “Neither this do I   believe, that one-eyed men exist in nature [␸ υονται] with the same nature [␸ υσιν] as other men.” Cf. 4.25.1, 4.105.2, and the cautionary gloss of source at 4.191.4. Herodotus’ rejection of fabulous peoples represent an innovation with respect to his predecessors (see Scylax of Caryanda in Tzetzes Chiliades 7.629– 36, qtd. in Romm 1992, 84– 85). 118. For Herodotus’ scientific outlook on the Nile, see especially 2.20– 23 (criticism of previous theories on the flooding of the Nile), 2.24– 26 (Herodotus’ own explanation), 2.27 (explanation of the peculiarity of the absence of breezes). See Lloyd 1976, 91– 107. Herodotus’ insistence on conformity to physis is analyzed by Corcella (1984, 74– 84) and Thomas (2000, 135– 38). See also Donadoni 1947. For identity of natural processes in different

84

Telling Wonders

compared” (verb συµβα λλειν) in fact appear in contexts where he mentions other rivers that in one way or another resemble this one, at least “if one can put together [συµβαλειν] great things and little.”119 Comparing the Nile to the well-known Ister even leads Herodotus to theorize on the unknown sources of the former, because an observed similarity between the two rivers becomes evidence for conjecturing (verb συµβαλλεσθαι) one that is not visible.120 Both unique and similar to other rivers, Herodotus’ Nile fulfills two contradictory functions. On the one hand, unfathomable and mysterious, its sources lost beyond reach, the Nile is with Egypt as a whole the very symbol of the problems facing human understanding that the know-it-all Greeks, with their usual penchant for simplifying things, do not properly recognize.121 On the other hand, the strangeness of the Nile is itself limited by its participation in the general nature of the rivers of the world, which it cumulatively sums up and represents. Herodotus both seeks symmetry and rejects the assumption of it, because his attempt to make sense of things scientifically goes hand in hand with a fear of the ideological consequences of oversimplification.  His inference that the Nile and the Ister “are equal” [ε ξισουσθαι] (2.34.2) in conformity to a conceptual model appears to contradict his polemic elsewhere against the regularity of the shape of the earth as was represented in the early maps—a perfect circle, surrounded by Ocean and internally subdivided into the three great landmasses of Europe, Asia, and Libya (4.36.2, 4.42).122 The edges of Herodotus’ earth are parts of the world, see the gloss of similarity at 2.25.1. Herodotus’ view that something that occurs in nature is susceptible to reduplication under the same circumstances emerges in the report of the “wonder” of the skulls observed at Pelusion (3.12) and verified elsewhere: “I saw another similar case [κα ι αλλα οµοια] at Papremis” (3.12.4). 119. 2.10.1– 3. Cf. 2.29.3, 4.53.5  ε γω συµβαλλοµαι τοισι ε µ␸ανεσι   τα µ η γινωσκοµενα 120. Herodotus writes, ως  τεκµαιροµενος [as I conjecture, inferring the things that are not known from those that are  apparent] (2.33.2; see also 2.34). Lloyd (1966, 337– 44) cites Anaxagoras’ dictum οψις των  ηλων  αδ τα ␸αινοµενα [things that are apparent are the vision of things that are unclear] (Sext. Emp. VII 140 ⫽ DK 59 B 21a) as the first extant formulation of this use of analogy, widely applied by the Hippocratic writers. 121. See, e.g., the Greeks’ fanciful theories on the flood, by which they want to be “signaled for their cleverness” (2.20.1); their limited conception of the extent of Egypt (2.15– 16); their ethnologically ignorant version of the story of Psammetichus’ experiment (2.2.5) and tale about Heracles (2.45); and their “myth” concerning the Nile’s origin from Ocean (2.21, 23). 122. Cf. 2.23, 4.8.2 (see Lloyd 1966, 342; Gould 1989, 89– 90, Thomas 2000, 75– 101). See also the contradiction between the “symmetrical” assumption at 4.36.1 and the criticism of symmetry at 4.36.2 (see Romm 1989).

Comparison

85

irregular and provisional; the perimeter of the inhabited world is often merely defined by the impossibility for men to reach out and discern what lies beyond a physical obstacle.123 The large-scale subdivisions of the oikoumene are to him mere theoretical constructs with little empirical validity. Herodotus attributes tremendous importance to natural boundaries as symbols of the limits human action must respect. In this context, the separation between Asia and Europe is fundamental; ignoring it for the sake of aggression epitomizes adikie.124 A different set of epistemological and ethical principles, however, is operative in his geographical discussions. This disorganized stade-by-stade traveler refuses to interpret the earth in terms of global canonical subdivisions.125 These lead to unacceptable territorial claims and to the compartmentalization of experience.  The Persians consider as their property [ο ι κηιευνται] Asia and the foreign peoples who live in it, while they regard as something separate [ηγηνται κεχωρι σθαι] Europe and the Greek world (1.4.4; cf. 9.116.3) Herodotus rejects this sort of imperialistic allotment just as he devalues big geographical boundaries.126 Arabia does not end at the Arabian   ] (4.39.1). The notion of the three gulf except “by convention” [νοµω continents, he insists, falsely assumes the similarity of these lands in size  and shape, while “not small are the differences [τα δια␸ εροντα] between them” (4.42.1). The contours of Asia, except on the eastern side, “have υρηται  been found to be similar” [ανε οµοια] to Libya (4.44.3), but Europe is “equal” to both of them in length; in width, Europe “does not ι η]” (4.42.1). appear to me even comparable [συµβα λλειν αξ  4.36.2) for the Asia/Libya subdivision adumHerodotus’ scorn (γελω, brates an objection to continents in general, be they three or two. To him,  η γ η), and the three continents are only the whole earth is one (µι  η ε ουσ 123. There are numerous glosses of “not knowing” in reference to what lies beyond certain points: 2.31, 3.98.2, 4.16.1, 4.17.2, 4.18.3, 4.25.1, 4.27, 4.40.2, 4.45.1, 4.53.5, 4.56. See Romm 1992, 10– 37, especially 32– 37. 124. See my introduction. 125. Contrast the apparently regular percourse of Hecataeus’ Periodos Ges (see Pearson 1939, 30– 96). On the theoretical schematism of Anaximander’s map, see Van Paassen 1957, 57– 61. 126. See Thomas 2000, 98– 100. Cf. chap. 3, n. 176, on how Herodotus’ conception of geographical boundaries must adapt to ideological considerations.

86

Telling Wonders

names.127 They represent such a distorted view of the way in which lands and water are distributed that, among other things, they are even unable to take into account the existence of Egypt or at least its delta (2.16). A country, he interprets, is not a predefined geographical entity surrounded by physical boundaries but rather coincides with the area of habitation of a certain people (2.17.1). Herodotus does not replace the old schemes with a new one of his own making; instead, he uses the traditional names as a matter of convenience (4.45.5). But what travel and opsis empirically teach him of the physical world, and what he wants his audience to visualize, is rather the parceling up of the earth by multiple boundary lines and the assiduous and more or less random repetition of physical patterns and shapes, in various sizes, throughout the unitarian surface of the oikoumene.128 “Comparing large things with little” once again, Herodotus explains that “Scythia borders on the sea, to the east and to the south, just like [κατα περ] Attica,” and that “the Taurians inhabit it  ε ι ] in Attica a people  in about the same way as if [παραπλησια . . . ως different from the Athenians should inhabit Cape Sounion” or “as if in Iapygia a people other than the Iapygians were to begin at the Brentesian harbor and inhabit the cape up to Taras.” These are just two examples,  and there are “many other similar [παροµοια] promontories that Tauris 129 resembles [ο(ικε].” In physical processes, as in shapes, differences and similarities are everywhere mutually balanced. In the land of Assyria, it does not rain and the crops are irrigated by the river through hand-operated machines, “not as [ου κατα περ] in Egypt, where the river rises.” But “canals cut across all the territory of Babylon, just like [κατα περ] Egypt (1.193.1– 2).” In Scythia, the country most antithetical to Egypt, “rivers are not much fewer in number than canals in Egypt” (4.47.1). The nature of both lands is such that “men do not plow.”130 When differences between two regions are most striking, analogy operates across categories (vertical analogy) and by compensation: “Somehow, the extremities of the inhabited earth obtained as their lot the most beautiful things just as [κατα περ] Greece 127. See 4.45.2– 5; Lloyd 1976, 78, 83– 85; Bornitz 1968, 190– 92; Benardete 1969, 111; von Fritz 1936, 320– 30; Thomas 2000, 80– 86. 128. Cf. Immerwahr 1966, 316. 129. 4.99.4– 5. See also 4.156.3, 4.182, 4.183.1. 130. 2.14.2, 4.2.2. See n. 91 in the present chapter; Hartog 1988, 17– 18. See also the parallels between Babylon and Egypt (4.198.1– 3), Libya and Babylon (4.198.1– 3). On Libya and Scythia, see Benardete 1969, 121– 26; Rosellini and Sa¨ıd 1978, 973– 74.

Comparison

87

obtained the most beautifully mixed climate” (3.106.1). In a subsequent description, the eschatiai (edges) are shown to contain both the most beautiful and the ugliest of things, as when the most fragrant of Arabic spices, ledanon, is said to be found in the stinkiest of places (3.112). Yet the beautiful and its opposite are distributed everywhere, and the verbal  correspondence between τα καλλιστα [the most beautiful things] and τας    ωρας πολλον τι κα λλιστα κεκρηµενας [the most beautifully mixed climate] alerts us to a qualitative similarity hidden behind the opposition. Does Climate Determine Culture? The word somehow (κως) at 3.106.1 (just quoted) expresses the idea that the empirically observed equivalence eludes the clear cause-and-effect combination of scientific reasoning. In a similarly inexplicable way, the patterned texture of the physical earth extends to the men who inhabit it. When the programmatic introduction to the Egyptian ethnography, quoted earlier, states that the Egyptians, “together with” [αµα] their different climate and river, also have laws and customs that are the complete opposite of those of the rest of mankind (2.35.2), Herodotus is reformulating the fifth-century medical theory about the influence of climate with drastic changes. The Hippocratic Airs, Waters, Places, though not exempt from inconsistencies and self-corrections, attempts to establish a logical connection between the natural features of a land, the physique of its inhabitants, and even their moral character. The description of the marshy area around the Phasis exemplifies the mixture of the scientific and the ethical codes: here the climate is hot and rainy and the waters turbid; the inhabitants have big, thick bodies, no visible joints or veins, jaundiced complexions, and a natural sluggishness with regard to physical toil (Airs 15). All finer regional distinctions, moreover, are subordinate to the difference that according to the Hippocratic author separates Europeans from Asiatics. In Asia, the stability of the climate and the fertility of the land stunt the courage, industry, spirit, and strength of the people (Airs 13, 16). The Hippocratic writer’s interest in the scientific causes of ethnic differences is proportional to his penchant for evaluating foreignness as a set of pathologies, abnormalities, or diseases. Herodotus’ ethnographic discourse stays clear of both things. He notices that even the inhabitants of the same general area are often different, but he does not say why. Only occasionally and provisionally does the text suggest that climate may

88

Telling Wonders

partly account for the way people are.131 The introduction to the Egyptian ethnography, at any rate, does not scientifically explain the correspondence between land and people but treats it as a sort of mystery: “Egyptian customs differ from those of the rest of mankind not because Egypt’s geographical situation differs from the rest of the world, but just as it is different.”132 A vertical analogy is here established between land and people, by which one level of reality represents the other.133 The physical texture of a unitarian oikoumene finds its symbolic correspondence in the patterns of differences and similarities of nomos. The Texture of Nomos At the intersection between a people’s land and its nomoi, between the physical and the cultural, are natural resources and the manifestations of material culture that these resources determine: foods, fabrics and clothing, buildings and utensils. In this area especially, Herodotus employs the sort of comparative language that emerges in the geographic passage about the Thracian Chersonese cited earlier. Hartog emphasizes its rhetorical value for making the exotic understandable through an appeal to 131. See Lateiner 1986, 16, and now especially Thomas 2000, 103– 14. In the Histories, the idea that an infertile land makes people hardy and a fertile land produces “soft” men appears in the historical narrative, in the mouth of characters: see 7.102.1 (Demaratus) and 9.122.1– 2 (Cyrus), the second particularly close to Airs, Waters, Places 12, but not backed by the narrator’s authority and occurring in a notoriously enigmatic episode (see Dewald 1997; Pelling 1997). The narrator establishes in his own voice a connection between the nature of the environment and the people’s physical characteristics, at 2.22.3 (men south of Egypt are black because of heat) and 2.77.3 (a stable climate promotes good health; seasonal changes bring diseases). For effects of climate on a region’s fauna, see 4.29, 5.10. Elsewhere in the Histories, the influence of environmental determinism is muted. Lateiner (1989, 159) sees an implicit connection between Egypt’s stable climate and the Egyptians’ unwarlikeness, acceptance of despotism, and unchanging institutions (see, e.g., the interpretive gloss at 2.147.2), though he admits that the narrator never formulates that thought. Herodotus says that the Ionians have built their cities in the most beautiful climate of the world (1.142.1), and in some cases, he represents the Ionians as weak (see especially 1.143.2, 6.11– 12). Beautiful climate is, however, elsewhere attributed to Greece as a whole (3.106.1), and the weakness of the Ionians appears to be a matter of history and institutions (1.169.1, 4.137– 39, 4.142). Aside from the general “semiclimatic” and moralistic idea that a tough life makes a people tough, Herodotus’ message on why an ethnos is the way it is proposes a variety of partial or possible reasons, none of them with absolute and general validity. 132. Immerwahr 1956, 279. 133. The Egyptians are like their land somewhat in the same sense as Darius (albeit misguidedly) perceives that he and the river Tearos are alike and belong together, “the best and the most beautiful of all men” and “the best and most beautiful of sources” (4.91.2).

Comparison

89

the familiar.134 But comparisons of this kind are also a manifestation of Herodotus’ ideology of a unitarian world. The world can be explained in this way because it is “same,” with a limited number of possible shapes. When the river rises and the plain is flooded, a great number of lilies, which the Egyptians call lotus, grow from the water. People pick them and dry them in the sun and then ground what they extract  to poppy seed, and make from the middle, which is similar [ε µ␸ερες] from it breads baked on the fire. There is also a root of this lotus, which is edible and rather sweet, of the size of an apple [µεγα θος  to roses, κατα µηλον]. There are also other lilies, similar [ε µ␸ερεα] that grow in the river, the fruit of which is found in a separate calyx that sprouts on the side from the root, very similar in appearance   οµοιοτατον] [ ι δ εην to a wasp’s honeycomb: inside this are several  grains as large as [ οσον τε] an olive, which can be eaten fresh or dry. (2.92.2– 4) Each object that is “different” becomes disassembled into parts—roots, flower, fruit—and explained as a combination of similarities.135 The comparative discourse visually recalls familiar appearances, even in those cases when size is what is actually being compared.136 The second terms of comparison include natural and cultural products from the internally diversified Greek world: Cyrenaic lotus (2.96.1), Boeotian sandals (1.195.1), Lesbian craters (4.61.1). The Egyptian Hephaestus in Memphis is “very  similar [ε µ␸ερεστατον] to the Phoenician Pataiki.” If someone has never seen a Pataiki, another analogue is available from a different region: it is the “imitation” [µι µησις] of a pygmy (3.37.2– 3). Exotic objects that resemble things found in Greece or in other foreign lands are not merely 134. Hartog 1988, 225– 30. See also Corcella 1984, 69. 135. So also are the hippopotamus (2.71), the ibis, and the water snake (2.76.1, 3). The procedure is similar to that for describing hybrids, but the effect is familiarizing rather than the opposite. Cf. 2.68.2– 3, 4.23.3, 4.177. See Hartog 1988, 249– 50. Another type of gloss follows the pattern “similar in everything, except such and such” (see 4.183.3, 4.61.1, 4.23.3).   (4.192.1) 136. In Libya, there are oryxes “as big as oxen” [µεγαθος . . . κατα βουν]  ησι ε µ␸ερεστατοι]  and crocodiles three cubits long and “very similar to lizards” [τ ησι σαυρ (4.192.2); the giant ants in India are smaller than dogs and larger than foxes, and they dig in

αυτ

τροπον]   ον the sand “in the same way” [κατα τον as Greek ants, to which they are “very similar in appearance” (3.102.2). See also 2.67.2, 2.73.2, 3.100, 3.106.2, 4.192.3.

90

Telling Wonders

easier to describe; they reveal the patterned character of cultural artifacts all over the world. Human industry, like nature, is always different but imitates itself incessantly. Reminders of this fact shorten the distance between the rest of the world and Greece in ways that, given the historical framework of Herodotus’ descriptions, may suggest more intangible points of vertical analogy. The simple notice that the outermost wall of Median Ecbatana is

Αθηνεων   “just about the size of that of Athens” [κατα τον κυκλον  µαλιστα κ η το µεγαθος] puts two imperial cities side by side.137 Sparta momentarily intrudes in the Scythian narrative when the narrator stops to notice that the cauldron built by king Aryantas at Exampaios in Scythia is “six times as large as” the crater Pausanias dedicated on the mouth of the Black Sea.138 Just as quantitative comparisons concerning size also suggest analogy in shape, so comparisons between two objects cause an overlap of their respective contexts.139 Foreign objects may also be explained in terms of their similarity to something that appears to lie outside their own category. If we look at the explanatory analogies in texts that are not concerned with the description of the world (e.g., most of the Hippocratic treatises), we notice the special poetic quality of Herodotean comparisons.140 These set side by side war, sailing, hunting, agriculture, and building as branches of a unitarian, device-producing field of human activity. A boat can be compared to an ox to be curbed (2.29.2) or to a shield (1.194.2), because it constitutes in itself the manifestation and symbol of many different sorts of effort. The most famous of these Herodotean transferences, in an ethnographic gloss describing the relay system of Persian postal couriers,141 has an impact not unlike that of a Homeric simile: “the first rider delivers his charge to the second, and the second to the third, and thence it passes from hand to 137. 1.98.5. See Georges 1994, 140. 138. 4.81.3. This comparison occurs in a narrative full of implicit analogies with Spartan realities discussed later in the present chapter; (“The Sameness of the Scythians”). 139. For other comparison of man-made items, see 2.7.1 (the Egyptian road from the sea to Heliopolis is almost equal in length to the road that leads from the altar of the twelve gods in Athens to the temple of Olympian Zeus in Olympia); 2.170 (the pond in the precinct of Athena at Sa¨ıs is about the same size as the Round Pond at Delos); 5.59 (Cadmean letters are similar to Ionian letters); 4.74 (Scythian hemp is very similar to linen). 140. Hippocratic analogies tend to assimilate physiological processes of the human body to technological processes or the life of plants. Lloyd (1966, 345– 60) gives many examples. See also Lateiner 1986, 13. 141. On ethnographic glosses, see chap. 1, “Referential Glosses.”

Comparison

91

hand, just as the torch-race that the Greeks perform in honor of Hephaestus” (8.98.2). The function of technical explanation and illustration cannot be separated from the ideological message it communicates metaphorically: the moral equivalence between two different kinds of performances, with different practical goals, occurring in different cultural contexts. From the experience of “same shape, same size” or “same procedure,” Herodotus’ pursuit of the similar extends to the observation of “same function”: “X is to ‘them’ what Y is to ‘us’.”142 Just as a linguistic translation must overcome a conceptual discrepancy, so the ethnographer translates culture by adjusting the difference to point out equivalence: the Issedones clean out and gild the skull of a dead man and then use it “as a    λµατι], making solemn sacrifices every year: sacred image” [ατε αγα “each son does this for his father, just as the Greeks celebrate the Genesia  [κατα περ Ελληνες τα γενεσια]” (4.26.2). The natural and the literal have a greater role among the Issedones than among the Greeks, but the purpose of foreign and domestic custom is the same.143 Glosses of this type make explicit a procedure Herodotus follows as a matter of course. Every time he describes a practice that occurs with no altar, no fire, no libation, no flute, no fillets, no barley meal (1.132.1), or a ritual in which the victim is tripped from behind and strangled (4.60), and calls each of these a θυσι α (sacrifice), he posits an equivalence that turns otherness into otherwiseness.144 The practice of interpretatio Graeca does not in Herodotus have an ethnocentric impulse; it is rather the foundation of his relativism. I am now ready to discuss the role of comparative glosses noting that two or more peoples are in certain respects similar or follow similar practices. 142. See Hartog 1988, 227. 143. For the Greeks, this degree of contact with the dead body may have been regarded as polluting. See Garland 1985, 41– 47. On the Genesia and other Greek commemorative practices for the dead, see Jacoby 1944; Garland 1985, 104– 8. For other Herodotean equivalences, see 1.202.2 (Caucasian primitives inhabiting the islands on the Araxes River have discovered [verb ε ξευρι σκω] a particular fruit, which they burn and inhale to become inebriated, “as the Greeks with wine,” the local equivalent of a symposium), 3.98.4 (the Indians of the marshes wear a mat made of reed “like a breastplate”). Cf. 4.75.1 (for the equivalence between a Scythian hemp sauna and a Greek steam bath), 1.193.5 (for the equivalence between Assyrian cultivation of palm trees and Greek cultivation of fig trees). 144. Cf. Mora 1985, 57.

92

Telling Wonders

As a rule, similarities between different ethnic groups are taken as a sign of mutual contact,145 common origin,146 or borrowing.147 Herodotus points out the debt of the Greeks toward various barbarian nations much more frequently than the other way around.148 When emphasizing the debt of Greek culture to Egypt in particular, he creates a corollary to the greater antiquity, wisdom, and moral authority of the Egyptians.149 The facts that ritual practices of certain Greek sects “are in agreement”   [ οµολογ εουσι] with Egyptian religious regulations150 and that the Egyptians celebrate some of the festivals in honor of Dionysus “about in the

παντα Ελλησι] (2.48.2;  α σχεδον same way as the Greeks” [κατα ταυτ cf. 2.49.2) are presented as cases of Greek borrowing from Egypt.151 A gloss of this type records that the Egyptians depict Isis with horns 145. In the case of ethnea living close to each other or tribes belonging to the same large ethnos (e.g., Scythians, Thracians, or Libyans), such interventions serve to confirm the assumption of internal differentiation (see discussion of differentiation earlier in this chapter). Proximity is the easily inferred cause of similarity between Lydians and Greeks (1.35.2, 1.74.6, 1.94.1, 7.74.1), Indians near Caspatyrus and Bactrians (3.102.1), Libyan nomads and Egyptians (4.186.1). 146. See 2.104.1– 5 (Colchians and Egyptians), 4.108 (Geloni and Greeks). 147. This is true even when the historical circumstances of the borrowing are unknown, as in the case of the language of Caunians and Carians (1.172.1) and that of the dress of Sigynnae and Medes (5.9). On diffusionism, see Muller ¨ 1972, 19– 20; Lloyd 1975, 150. 148. The Greeks contribute the technique of welding iron to world culture (1.25), pederasty to the Persians (1.135), most of their customs to the Asbystae (4.170), and a ritual for Perseus to the Egyptians of Chemmis (2.91.4). 149. According to Herodotus, the Egyptians invented the year and its subdivisions (2.4.1); geometry, which was then introduced into Greece (2.109); altars, statues, temples, animal sculptures, and use of the names of the twelve gods, the last of which the Greeks took from them (2.4.2; cf. 2.43.2, 2.49, 2.50.1); divination (2.49.2, 2.52.2) and oracular shrines (2.54– 57); festivals and processions (2.58; cf. 171.2– 3); and hemerology (2.82) and the theory of transmigration of souls (2.123.2– 3), these last two followed by some Greeks. The Athenians, thanks to Solon, have adopted in perpetuity what Herodotus evaluates as an “impeccable” Egyptian law (2.177.2; cf. 2.160). Herodotus says that the round shield and the helmet have come to the Greeks from Egypt (4.180.4). 150. 2.81.2. On the relation that Herodotus establishes (here and at 2.123.2– 3) between Egyptian cult and Pythagoreanism and Orphism, see Froidefond 1971, 187– 89. 151. According to Herodotus, Greek borrowings from other peoples include, “interestingly enough” (αρα), from the Libyans, the dresses of the statue of Athena, the aegis, the ritual cry of women at sacrifices, and the practice of yoking four horses (4.189.1), as well as the god Poseidon (2.50.3); games from the Lydians (1.94.1– 2); hoplitic equipment invented by the Carians (1.171.4); the Dioscuri, Hera, Hestia, Themis, the Charites, the nereids, and ithyphallic Hermes from the Pelasgians (2.50.2– 2.51.1); the solar clock, the meridian, and the subdivision of the day into twelve parts from Babylon (2.109.3); writing from the Phoenicians (5.58); the Ionian dress, which is actually Carian (5.88.1).

Comparison

93

“just as the Greeks do” [κατα περ Ελληνες] (2.41.2). The similarity to the Greeks here works as a sort of protective shield under which the text raises for the first time the difficult issue of Egyptian theriomorphy. The identification of gods with animals and the worship of living animals appear to interfere with Herodotus’ agenda of demonstrating to a Greek audience the superiority of Egyptian purity, religiosity, and dedication to the divine.152 His account is fuzzy in this area, for it simultaneously communicates several overlapping and contradictory messages: (1) Egyptian animal worship is “different” and disturbing; (2) it is (a) purely symbolic and therefore (b) similar to some Greek representations of the divine; (3) it is “Good,” a special sign of the Egyptians’ closeness to the gods and of a profoundly instructive view about the mutual relationship among different forms of life in the cosmos.153 This conflicted attitude first comes to the fore in the excess of discourse that accompanies Herodotus’ account of Egyptian representation of the divinity the Greeks call Pan (2.46.2). He begins by underlining similarity (2b). In Egypt, painters and sculptors represent Pan with a goat face and he-goat legs, just as the Greeks do [κατα περ Ελληνες]. Next, he reassures by negation (2a). . . . not because they believe him to be of such form—no, they believe him to be the same as the other gods.    [ουτι τοιουτον νοµι ζοντες ε"ιναι µιν αλλ οµοιον τοι σι αλλοισι  θεοι σι]. Finally, he admits to some uneasiness with this type of thing (1). For what reason they depict him in this form, it is not too pleasant  ε στι λεγειν].  ον  for me to say [ου µοι ηδι 152. Egyptian zoolatry was repulsive or ludicrous to outsiders. See Lloyd 1976, 291– 96, especially 293– 94 for Greek and Roman evaluations. In a fragment of Anaxandrides (39 Edmunds) an Athenian character comically declares the incompatibility between Greeks and Egyptians largely on the basis of the latters’ zoolatric practices. 153. Cf. Froidefond 1971, 202– 3. The narrator’s ambivalence does not at any rate mean that the analysis “the humanity of the Greek gods makes the Greeks superior to the Egyptians” (Benardete 1969, 46) represents the message of the text.

94

Telling Wonders

The negative program in the sentence just quoted is consistent with the general introduction, where the narrator declares himself not eager   . . . προθυµος) (ουκ to give detailed report of “divine things” he learned, on the grounds that what different cultures know about the gods is “equal” [ ισον]; he will therefore avoid the topic unless absolutely com154 Both  ο του λογου   pelled by his argument (υπ ε ξαναγκαζοµενος). times, Herodotus expresses himself subjectively in terms of personal distaste rather than blame and invokes similarity or equivalence. One passage where he breaks his rule of silence, evidently compelled by the logos, produces an apologetic effect: it etiologically explains the Theban prohibition to sacrifice sheep and rams to Zeus and the statues representing Zeus with the face of a ram, with a local tradition that is comfortingly similar to the familiar Greek myth of Zeus and Semele. Heracles wanted to behold Zeus directly, but Zeus did not want to be seen by him. Since Heracles insisted, Zeus skinned and decapitated a ram and then showed himself to Heracles wearing the ram’s hide and holding its head in front of himself (2.42.3– 4). The myth proves appealing because it reduces theriomorphy to a disguise; it communicates the message that the Egyptians do not really identify Zeus with a ram, just as the interpretive gloss at 2.46.2 explicitly says that they do not really believe Pan to be  τοιουτον—a he-goat.155 The reminders that the Greeks also represent Isis with cow’s horns and Pan with the face and legs of a goat similarly have the function of diffusing strangeness. Yet the “difference” represented by the Egyptians’ peculiar relationship to the animal world is also treated as clearly axion logou. In the account of Pan’s cult in the district of Mendes, the explicit denial that Mendesian Pan is a goat competes with the implication that he really is one. The logical discontinuity of the interpretive discourse barely camouflages the idea: “This is why the aforementioned Egyptians do not sacrifice he-goats and she-goats: the Mendesians reckon Pan to be one of the eight gods” (2.46.1). Later on, the narrative makes a revealing transition from goats in general to one special τρα γος, venerated above all the others (2.46.3– 4). A gloss intervenes, to show that the identification between god and animal is inscribed in the language: “In Egyptian, both 154. 2.3.2. The need to protect the relativistic position is clearly one of the issues here, though not the only one. See Linforth 1924; Gould 1994, 92– 93, 103. Cf. the refusal to explain why animals are sacred in Egypt at 2.65.2 and 2.47.2. 155. For ram cults in ancient Egypt, see Lloyd 1976, 190– 94. Herodotus’ reductionism in explaining Egyptian theriomorphy is discussed by Gould (1994, 202– 3).

Comparison

95

the ram and Pan are called Mendes” (2.46.4). Finally, Herodotus provides a terse narrative of what he introduces as a shocking event and apparently interprets as a public ritual. In this district, in my time, there took place a prodigious thing  [τερας]: a he-goat coupled with a woman. This came about as a public exhibition [ε πι δεξιν].156 Among the entries in the preliminary list of Egyptian customs that are “opposite to those of the rest of the world,” one records, “the daily life of   other people is defined as separate [χωρ ι ς . . . αποκ εκριται] from that of beasts, but the Egyptians live theirs together with their animals” (2.36.2). Later on, an evaluative statement introduces a survey of sacred animals (2.65– 76) and directs the listener to interpret it as an illustration of the extraordinary Egyptian piety. The narrative describes the role of animals in religion, the symbiosis of the Egyptians with their animals, and the phusis (physical characteristics) and modes of behavior of the animals themselves. Like human beings, animals perform goal-directed action; at the same time, they seem to enjoy a special connection with the supernatural world. Thus, male cats contrive (σο␸ ι ζονται) to couple with the females depriving them of their young (2.66.2), while on other occasions 157 As the land of  cat behavior is divinely inspired (θεια πρηγµατα). thomata, Egypt once again puts into focus fundamental problems of analogy and difference. Elsewhere in the Histories, the observation of primitive cultures raises the issue of the cosmic relationship between the realm of animals, human custom, and the divine.158 But here is a developed society where the most cultured and pious of all people worship animals and where animals enact those same elementary impulses that 156. 2.46.4. The form of the discourse here (summary introduction, narrative, concluding gloss) resembles 7.57.1, but the interpretive gloss in the conclusion is opaque; see chap. 1, “What Is Metanarrative?” in the present book. On the tradition of bestiality in the Mendesian nome, see Lloyd 1976, 216. 157. 2.66.3. See Smith 1992, 7– 15, 96– 107. Cf. the funeral ritual of the phoenix (2.73) and the ibises defending the pass at Buto against the winged serpents of Arabia (2.75). See also the alleged wolf ritual at 2.122.3 and the cow statue with the attached story of Mycerinus’ daughter at 2.129.3– 2.130.1. 158. The notion of a divinely and naturally determined analogy between the animal and the human world is implied in the section on the extremities of the earth and especially at 3.108; see especially Pagel 1927, 30– 33, on this passage as a description of nature that parallels the historical process.

96

Telling Wonders

among human beings translate into nomos. Herodotus’ view of a vertical correspondence between different levels of reality, which elsewhere expresses itself in animal symbolism and analogy, is susceptible to historie especially in Egypt.159 In book 2, the contradiction between the sameness and otherness of the world emerges, then, as a theoretical principle, which encompasses the cosmological, the geographical, and the ethnological spheres. The initial introduction to Egyptian culture attributes to it customs that are all opposite to those of the rest of the world (2.35). In the course of the narrative, this statement is partially confirmed and partially corrected, as Egypt becomes archetypal in both historical and symbolic senses.160 Other similarities reveal an ethical affinity between Egypt and Greece that cannot be accounted for in terms of contact, influence, or borrowing (see, e.g., 2.92.1). In the section just preceding the narrative on Egyptian animal worship, this affinity between Egypt and Greece in opposition to the rest of the world is surprisingly formulated in terms of a clear-cut separation, first institutionalized by the Egyptians, between animals and humans. Almost all other peoples, except Egyptians and Greeks, have intercourse in temples and go to the temple after intercourse without having washed, considering that humans are like other animals  [κατα περ τα αλλα κτηνεα]. (2.64.1) Sparta, which throughout the Histories represents what is most desirable from the point of view of social and political ideology, is also the city of Greece where Herodotus finds the greatest number of correspondences with non-Greek societies, including the Egyptian society.161 The origins and causes of such correspondences remain indefinable. Similarly, in his discussion of the special honors the Egyptians reserve for their hereditary warrior caste, the Machimoi, the narrator stops again to notice a similarity with Greek, and especially Spartan, values and declines, to interpret it as yet another sign of Egyptian influence. 159. Froidefond (1971, 193– 94) underlines Herodotus’ philosophical interpretation of Egyptian religion as expressing the notion of a solidarity among living beings not without similarities with Orphic and Pythagorean doctrines. 160. See Benardete 1969, 47. 161. See 2.80 (respect for the elders; see Froidefond 1971, 176); 6.60 (hereditary assignment of professional occupations). Sparta is compared to foreign nations also at 6.58.2 and 6.59.

Comparison

97

I cannot specifically determine whether the Greeks have learned this also from the Egyptians, seeing that also Thracians, Scythians, Persians, Lydians, and almost all the barbarians regard as less worthy of honor than the other citizens those who learn a trade and their descendants, considering noble instead those who abstain from manual work and especially those who devote themselves to war. This is something, to be sure, that all the Greeks have learned and especially the Spartans, though the Corinthians are those who despise artisans the least. (2.167.1– 2) The two cultural features involved in this comparison—contempt of manual labor and admiration for the pursuit of war—find specific formulation in the list of Thracian customs (5.6.2). Of the other foreign peoples named in 2.167.1, the Scythians are described as warlike throughout, and the Persians appear even more aristocratically contemptuous of petty trades than do the Greeks themselves (1.153). The inclusion of the Lydians in the list is surprising, however—even more than the fact that the gloss is attached to the description of the unaggressive Egyptians. It shows the extent to which Herodotus wants to stretch the similarity.162 Wherever the nomos comes from, and regardless of its cross-cultural and intracultural variations (e.g., the fact that in Greece it is followed more by the Spartans than by the Corinthians), it represents an area of ideological agreement between and within two internally differentiated groups— “almost all the barbarians” and “all the Greeks.” Many other glosses in the Histories underline resemblances between a foreign culture and the Greeks or between one foreign culture and another. Formulae of the type “about the same, except for . . . , ” also found in the description of exotic animals and plants,163 place the narratable difference in the context of the overall similarity of certain customs (see, e.g., 1.74.6). Such formulae also point out the overall similarity of two people in a specific area of culture, with the exclusion of a smaller group within a larger ethnos. The similarity must remain pure discourse and devoid of narrative content, while the difference is narrated, often with an unforgettable visual detail (see 4.190). In most cases of this sort, once again, similarity remains unexplained. Nasamones and Massagetae live at the opposite ends of the world, but 162. On Lydian unwarlikeness, see “The Sameness of the Lydians” later in the present chapter. 163. See end of n. 135 in the present chapter.

98

Telling Wonders

(predictably) they are characterized by about the same degree of primitivity; as a consequence, they have similar marriage customs, just as the Babylonians share certain cultural practices with the equally developed Egyptians.164 But Herodotus also establishes connections between societies with discrepant levels of culture.165 The frequent cross-cultural references convey two distinct, but not incompatible, attitudes. Firstly, Herodotus’ historie teaches that long-range coincidences, recurrences, and overlaps are something we can count on in the sphere of culture, just as it verifies the more or less random repetition of small patterns on the physical surface of the earth. This assumption of similarity can even provide some corroboration for the existence of an unseen cultural phenomenon, if one knows about the existence of one like it somewhere else.166 Secondly, however, recurrence and similarity, though likely, are underrated phenomena, which the narrator points out as worthy of great interest. The tradition of ethnographic discourse to which Herodotus is the heir is overdetermined in the other way. Difference is the standard of narratability and what the audience is readiest to expect and perhaps best trained to recognize—difference, most especially, from themselves. Particularly in the sphere of custom, therefore, Herodotus is generally more intent on showing the mutual equivalency of foreign practices and their functional character and normalcy than on letting his audience gape at their strangeness. When, instead of difference or equivalency, actual similarity occurs, it constitutes a “wonder,” that is, a profoundly satisfying discovery that invites reflection.167 The Egyptians follow their ancestral customs rather than acquiring new ones. Among other notable customs, they have also a particular song, “Linus,” which is also sung in Phoenicia, Cyprus, and other countries but changes name from culture to culture, though it  ος

συµ␸ ερεται  happens to be the same [ωυτ ε"ιναι] as the one the 164. On the Massagetae and Nasamones, see 4.172.2; cf. 1.216.1. See Rosellini and Sa¨ıd 1978, 975. On the Babylonians and Egyptians, see 1.182.1– 2, 1.198. 165. He establishes connections between Babylonians and the Lycians of Patera (1.182.3), Babylonians and the Eneti of Illyria (1.196.1), Babylonians and Cyprians (1.199.5), Ethiopian Macrobioi and Egyptians (3.24.2), Satrai and Greeks (7.111, where one item is part of an explicit comparison and another suggests implicit analogy). 166. See 4.33.5, 4.195 (cf. 2.150.2– 4). On the principle that “things that are apparent are the vision of things that are unclear,” see n. 120 and corresponding text in the present chapter. 167. See Lloyd 1975, 147; Corcella 1984, 90– 91; Giraudeau 1984, 121.

Comparison

99

Greeks sing; so, among the many things in Egypt that I am in  ζειν µε], there is also the question of  wonder about [αποθωµα where they found this “Linus.” They seem always to have sung this song. (2.79.1– 2) The Egyptian ethnography begins by presenting Egypt as the land possessing the greatest number of thomata, as well as nomoi and ethea that are all opposite to the rest of humankind (1.35). In the passage just quoted, a minute similarity is ranked among the Egyptian wonders: it connects not only Egyptians and Greeks, or Egyptians and Phoenicians, whose historical contacts can be reconstructed, but Egyptians and many other peoples from whom the Egyptians are isolated and to whom they are, as a rule, opposite. The feature involved in the similarity is moreover a specific—and one could almost say, unessential—ritual, something that belongs entirely to culture. Its peculiarity enhances one’s surprise at finding that it is shared. Pre-Socratic ideas concerning the basic homogeneity of humankind lead up to the principle, later in the fifth century, that the differences among people are a matter of convention, while uniformity of phusis (nature) is what truly counts.168 But Herodotus’ discovery about the ubiquity of “Linus” shows that for him the cultural uniformity one can perceive in the midst of the obvious differentiation of nomos is especially interesting, because it provides evidence for a connection between nomos and phusis and therefore for the equal importance of both in human events. Herodotus does not express himself in terms of an antithesis between the two and is far from proclaiming the importance of the one to the detriment of the other. The narrator or his characters qualify as     “human nature” (ανθρ ωπινος and ανθρωπ ηιος) human problems, sufferings, and limitations; what is humanly possible in contrast to what is not; and all the objective constraints that constitute not so much humanity itself as the condition that is imposed on all people of all ethnea.169 The 168. See, e.g., Antiphon, DK 87 B44, frag. 2. See Thomas 2000, 132. For discussion on ancient ideas of the unity of humankind, see especially Guthrie 1969, 160– 63; Baldry 1965, especially 23– 29.   169. Cf. το ανθρ ωπινον (human nature) as pattern of behavior in Thucydides (e.g.,   1.22.4, 3.82.3). In Herodotus, the phrase η ανθρωπη ι η ␸ υσις once refers to what a person     is capable of doing (3.65.3); the adjectives ανθρ ωπινος and ανθρωπ ηιος tend otherwise to  be referred to human affairs (πρηγµατα) and what happens to people (παθεα). See Powell  1938, s.vv. The notion of limitation is also implicit in Herodotus’ use of the term ␸ υσις in

100

Telling Wonders

complex of collective human responses to these constraints—we would like to say, the natural and at the same time divinely ordained sphere of such responses—is the “Nomos king of all” of 3.38.4, the embodiment of human impulse toward self-regulation and culture that generates all the different nomoi.170 A major task in Herodotus’ study of the internal differentiation of Nomos consists in verifying and pointing out to his audience the cross-cultural correspondences, equivalences, and similarities of the nomoi. Religious customs represent an especially important field of investigation in this respect, because they are less immediately functional and apparently less contingent on a people’s external circumstances than are other customs. They are, moreover, always based on theoretical and unverifiable constructs that vary from ethnos to ethnos. When the narrator states, “what all men know about divine things is equal [ ισον],” this generalizing comparative gloss couples the idea of contingent differences with that of essential uniformity. Herodotus continuously assumes or points out the cross-cultural correspondence of major figures of gods by translating divine names; for him, this constitutes an important proof that human societies tend to conceive of the divine in somewhat similar ways. The consensus of different ethnea over a minor ritual expression, such as the song “Linus,” carries the proof of the unity of cultures one step further. It is amazing precisely because it does not appear to be either mutually learned or a fundamental and predictable expression of human “nature” and the human experience of the world. Implicit Similarity and Analogy The Sameness of the Lydians In the ethnographic sections, glosses of similarity counterbalance the representation of difference. Further, like the far less numerous pointers of reference to one’s character affecting behavior, especially courage (5.118.2, 7.103.4,  8.83.1) or anger (7.16α1); in an ethnological sense, ␸ υσις denotes the temperament of a people (more or less aggressive; see 1.89.2, 2.45.2), which is distinct from, but not in   antithesis with, their νοµοι (in the plural; see 2.45.2). Otherwise in Herodotus, ␸ υσις is something concrete and uncontroversial, such as the family from which one is born (7.134.2) or the physical characteristics of people (3.116.2, 8.38), animals, or lands. Only at  4.39.1 does the implicit idea of ␸ υσις, in the sense of “natural lay of the land,” come into  antithesis with νοµος, in reference to the traditional (and inadequate) way of speaking about continents. See Heinimann 1945, 13– 41; Giraudeau 1984, 131– 32. 170. See chap. 3, “Funeral Customs and Other Nomoi.”

Comparison

101

analogy in the history, these cooperate with the effects of implicit analogy that the text achieves through narrative means. When it comes to foreign peoples and places, Herodotus talks to an audience of disbelievers. Even harder for the Greeks to accept than what may sound like the tall tales of travelers (1.193.4)—and more crucial for Herodotus to display—is the similarity of the other. After the assassination of an usurper to the throne, seven noble Persians come together to reconsider the form of government that would be best for their state. Can Persians discuss political theory and come to a reasoned choice as Greeks would do? Herodotus corroborates, “Speeches were made, unbelievable [απιστοι] to some of the Greeks, but they were nevertheless made” (3.80.1; cf. 6.43.3). The way in which the narrator marks the episode reveals his program to thematize similarity and overcome the prejudice of his listeners. Frequently in the text, an unexpectedly familiar feature is strategically planted in the midst of alien foreign actions or customs. A case in point is the statement “the Egyptians call barbaroi all those who do not speak the same language as themselves” (2.158.5). At a different level, the narrative elicits the audience’s self-identification, for example, by recording a special custom of the Trausians: “The relatives sit around the newborn lamenting the misfortunes of which he, since he has come into this world, will have to fill the measure and enumerating all the   . . . παθεα]. As for the sufferings that are the lot of men [ανθρωπ ηια dead, they bury him in the earth among celebrations of joy and merriment, considering all the evils from which he has been freed so that now  he is entirely happy [ε ν πα σ η ευδαιµον ι  η]” (5.4.2). The Trausian custom as such reverses birth and funeral and is opposite to the Greek practice. But at the same time, it represents the ritual enactment of a perception of human experience widely theorized by the Greeks as a true and superior insight.171 The cultural knowledge encoded in this custom raises an entire foreign people to the rank of those who are, by the audience’s own standards, especially wise.172 Toward the end of the work and within the narrative of Persian 171. The maxim that death is better than life is expressed in the Histories by Solon (1.31.3) and Artabanus (7.46.4). Asheri (1990b, 149) cites the parallels of Theognis 425– 28 and especially Hesiod frag. 377 MW: “et Hesiodus natales hominum plangens gaudet in funere.” 172. Asheri (1990b, 149) and others assume that the Trausian custom is based on a belief in immortality, such as Herodotus attributes to the Getae (4.93), but in Herodotus’ description of the custom, the eudaimonie attributed to the dead person seems to consist only in the deliverance from the evils of life.

102

Telling Wonders

debacles, the enraged Persian commander Artayntes almost kills Xerxes’ brother Masistes, who has accused him of having exercised his leadership  (9.107.1). An in a manner “worse than a woman” [κακι ω γυναικος] ethnographic gloss intervenes to explain that “among the Persians, to be called ‘worse than a woman’ is the gravest of insults” (9.107.2). What is the point of translating a cultural code that is exactly the same as that known by the audience?173 The narrator never uses it in his own voice but attributes it throughout the Histories to various characters, especially Persian, at the same time as he is accumulating evidence of vigorous female actions and of male inadequacies vis-a-vis ` women.174 At this particular stage of the narrative, the gloss underlines the convention to remind the listener that the current humiliating state of affairs for the Persians in the face of the victorious “masculine” Greeks is not the result of that society’s fundamental values. Rather, it derives from a historical process that has interfered with those values and prevented them from becoming actualized. Herodotus’ representation of the Persians proceeds, in fact, on two parallel tracks. On the one hand, they are throughout— from the times of Cyrus and the narrative time of the Persian ethnography to the battle of Plataea and the narration present of that same ethnography—masculine, tough, courageous, and dedicated. On the other hand, their “enslavement” to an autocratic ruler, the expansionism of their kings, the consequent acquisition of material goods, and the corruption of their leaders by luxury have cooperated to produce an inferior performance. This has happened even though the Persians value and cultivate what the Greeks call andreie, “masculine valor,” just as (implicit analogy) the Greeks do. The issue of andreie as a traditional standard of differentiation between Greeks and Eastern barbarians is prominent in Herodotus’ representation of the unwarlike Lydians, the first barbaroi who enter the Histories as an ethnographical subject, and the barbarians who, the text insists in explicit terms, most resemble the Greeks. In the regocentric narrative about Croesus (1.6– 91), one of the few passages that features 173. Greek characters and narrators in literary texts proclaiming that men should be “men and not women” and so on span a broad range, from the Homeric heroes to Socrates and beyond. See, e.g., Il. 2.235, 2.289, 8.163, 22.124– 25; Soph. Trach. 1071– 72; Thuc. 4.27.3; Plato Apol. 35b2– 3; Xen. Lak. Pol. 11.3. 174. 1.155.4, 1.207.5 (with 1.212– 14), 2.102.5, 7.11.1, 7.210.2, 9.20, and especially the whole Artemisia sequence (7.99.1, 8.68α1, 8.88.3). On women in Herodotus as foils for men, see Dewald 1981.

Comparison

103

the collectivity of the Lydians, a historico-ethnographic gloss with quantitative comparison, helps exonerate the Lydians from the defeat in the battle of Sardis: “At that time there was no people in Asia more virile     [ανδρηι οτερον] or stronger [αλκιµ ωτερον] than the Lydians. They fought on horses, carried long spears, and were good riders” (1.79.3). Later on in the history, we learn that after their conquest by Cyrus and subsequent rebellion, Croesus suggested to Cyrus that he impose on the Lydians a cultural change, so that they would not rebel again. The reform included a ban on all weapons and the prescription that the Lydians wear tunics and buskins, play string instruments, and bring up their children to  be shopkeepers (καπηλευειν). In this way, Croesus said, they would soon “become women instead of men” (1.155.4). This is a crude etiology for the effeminacy of the present-day Lydians according to the Greek stereotype; it dramatizes the consequences of their political enslavement to the Persians.175 But positioned between the two passages we have just seen, an ethnographic section provides another model of development and an intelligible context for what the history depicts as an abrupt transformation of the Lydians into “women instead of men.” This insertion, which I shall call the Lydian appendix (1.92– 94), also has another simultaneous—and one might say contrasting— agenda: that of establishing both explicitly and implicitly the cultural similarity of Lydians and Greeks and the status of Lydia as almost a nonforeign country. This explains why the passage, compared to the other ethnographies in the work, contains very little geographic and ethnographic information—no wonders, no lay of the land, no data on climate and production. According to the programmatic introduction, Lydia is the opposite of Egypt, with few thomata and little to tell (1.93.1; cf. 2.35.1). The city of Sardis, Croesus’ prosperous capital (1.29.1) later joined to Susa by the Royal Road (5.52), here receives no description. The golden dust washed down by the Pactolus (see 5.101.2; cf. 5.49.5, 6.125.4) is barely mentioned. The river Halys, Croesus’ fundamental “ethical” boundary in the historical narrative (1.6.2, 72, 75), is entirely absent.176 How does the information Herodotus chooses to include help to explain present-day Lydians while supporting the notion of their affinity with the Greeks?   175. See 1.94.7 (ε δεδουλωντο). Cf. ε ξανδραποδι σασθαι (1.155.1) and ανδραπο δισθεντας πρηθηναι (1.156.1). 176. See Jacoby 1913, 332, 339– 40; Lombardo 1988, 172– 81.

104

Telling Wonders

The Lydian appendix, subdivided into three clearly marked sections, occurs at the point in the logos when the story of Croesus has definitely ended; a dramatic aftermath (1.86– 91) has invited reflection on its complex set of meanings. The first section (1.92.1), on “Croesus’ other offerings in Greece,” is a delayed addition to the narrative of the king’s dedications to Delphi, which was prominent in the history of his policies (1.46– 55). The “other offerings” remind us of Croesus’ ambiguous alterity: on the one hand, his respect for Greek gods and sympathy with Greek culture; on the other hand, the gold of the dedications, which points to the Asian splendor of his rule. A historical gloss at the end of the section enhances the monarchical code: the offerings were partly financed from the estate of a political enemy who supported Croesus’ half brother, Pantaleon, in his claims to the kingship and was subsequently tortured with a carding comb and executed (1.92.2– 4). This entirely new story contains the only act of royal mutilation attributed to Croesus: it rectifies the domesticated portrayal of the Lydian king in the preceding narrative and regularizes his membership in the analogical category of absolute rulers that dominates the rest of the Histories. After the offerings of Croesus, Herodotus only chooses to describe the tomb of Alyattes. He advertises it as the one great εργον that can almost stand comparison with those of Egypt and Babylon (1.93.2). By way of this oversized mark of royal power, the discourse finally arrives at the hitherto much neglected community of the Lydians. They have commissioned the monument, and on the summit of the structure, consisting of a base of huge stones surmounted by a mound of earth, five pillars say which parts have been paid for by merchants, artisans, and prostitutes, respectively (1.93.2– 3). The three professions are emblems of Lydian “femininity” (not effeminacy, as in the passage on Cyrus’ reform). Banausic activities are related to women’s work and prostitution is another form of commerce.177 The ability of women to be agents in the place of men—unlike Greek women, the daughters of the Lydians raise their own dowry (by practicing prostitution) and give themselves in marriage (1.93.4)—points in the same direction. So does the emphasis on building (εργα), a feature we find notably in Babylon, where the commerce of women in some form or another is also traditional (1.196.1– 4, 5, 1.199). Most important, however, is the ethnographer’s choice of the 177. See Xen. Lak. Pol. 1.3. For prostitution as one of the indices of a soft culture, see Redfield 1985, 109– 10, quoted earlier in this chapter.

Comparison

105

monument to represent and symbolize the society as a whole. Standing next to another Mermnad landmark, the Gygaean Lake (1.93.5), the sema (tomb/sign) of Alyattes is a sema of the Lydian people and their subjection to the king. “Femininity” in the sociocultural sense, prostitution, and subjection to monarchical rule are all features that differentiate the Lydians from the Greeks, but the narrator insists on cultural similarity (1.94.1). The section that follows, the last of the ethnography, consists mainly of a long historical gloss that finds the roots of affinity in the past. The Lydians were “the first men we know” to coin gold and silver and the first to become shopkeepers (καπηλοι, 1.94.1), and it is said that they have invented various games the Greeks also now play (1.94.2). Their innovations (ε ξευρ- occurs five times) connote intellectual resourcefulness, or sophie, which the Greeks claim as their national heritage.178 Here they also represent a display of hardiness. During a famine in pre-Heraclidean times, the Lydians “looked for remedies,” and “one man devised one thing and another that.” This occasioned the discovery of popular games as a sort of distraction, so that by playing the whole day and not taking any food for one day out of two, they lasted for eighteen years (1.94.3– 4). The mention of the famine leads in turn to the subordinate narrative of the Lydian migration to Tyrrhenia (94.5– 7), reminiscent of familiar stories of later Greek colonizing expeditions, such as those the text reports elsewhere concerning Therans, Phocaeans, and Teans.179 The Lydian king divides the population into two groups and draws lots (as the Therans do), to decide who should stay and who should go; he then places his own son at the head of the latter group (1.94.5). Unlike Alyattes and Croesus, this king stands outside the monarchical model: he rather resembles the archaic Greek basileus in his role as founder.180 Thus, seen as a whole, the Lydian appendix pursues the idea of similarity and difference with the Greeks by tracing in reverse order, through description and embedded historical narrative, a telescoped evolution of 178. Cf. 1.60.3. See Diller 1961, 67– 68. 179. After seven years of drought, the Therans draw by lot a party from themselves, at the rate of one son out of two, and set off to Libya under the leadership of Battus (see 4.151– 53). The Phocaeans and the Teans abandon their homes and sail for the West rather than face Persian subjection (see 1.164– 68). A few generations later, the Athenians apparently consider doing the same (see 8.62.2). See also the Ionians at 1.170 and 9.106. 180. The traditional figure of the founder, as the embodiment of the trials and tribulations of the city and the antithesis of the tyrant (though also with the potential of becoming a tyrant), has been examined by McGlew (1993, 22– 26, 157– 82).

106

Telling Wonders

the Lydian people—from the remote struggles for survival of a small and courageous nation, to their resourceful inventions and economic development through banausic and “feminine” trades (καπηλοι at 1.94.1 corre sponds to καπηλευειν at 1.155.4), to the growth of their monarchy’s wealth and power. In this representation, the Lydians are fundamentally similar to the Greeks, not Hellenized barbarians. Croesus is Hellenized, but his rule also marks the maximum distance between the collectivity of the Lydians and the Greeks. The difference between the two ethnea is in fact a by-product of the Mermnad monarchy’s development. By the time of Alyattes, and even more so with Croesus, Lydian society has become increasingly passive and subjected: it expresses itself through the king, his monuments, his wealth, his personal prestige and connections, and his more or less well-advised policies. In the first section of the Lydian appendix, Croesus is dominant, and the people are ethnographically absent, just as they were politically irrelevant in the preceding historical narrative. In the chapter on Alyattes’ tomb, they place their activity in the service of their king. Among the early Lydians of the third section—who are the closest analogues of the Greeks and historically explain the affinities that still exist between the two peoples—those who left and changed their name to “Tyrrhenians” perhaps avoided this development. Like the Phocaeans and Teans, at any rate, and unlike the rest of the Ionians, they were spared what eventually befell those who had stayed behind.181 As the narrator abruptly concludes, “the Lydians, then, were enslaved by the Persians” (1.94.7). The alleged transformation into “women” under Cyrus will represent the next stage. The insistence on similarity and sameness in Herodotus’ representation of foreign peoples fits in with the analogical thrust of his historical account. A major concern of the apodexis as a whole is to describe the world so that the histories of foreign peoples may be comprehensible in light of their specific cultures. At the same time, however, each model of explanation, each way of being and becoming, is to some extent also applicable to everyone else, including the Greeks. The signals of similarity between different ethnea both conform to the more general principle of the patterned unity of the world and suggest specific ways in which historical processes can reproduce themselves cross-culturally. I am speaking here not of firm historical laws or inevitable cycles but of recognizable 181. Cf. Lombardo 1988, 202 on Herodotus’ Lydians as a negative paradigm for the Ionians.

Comparison

107

models of likely human behavior and of likely consequences. Herodotus’ Persians, Scythians, Lydians, and so on, are each different from the Greeks and every other people. The history of each is to a great extent not repeatable in other cultural settings. Yet specific features shared by different cultures and the sameness of the anthropinon connect all of these in an untidy system of mutual allegories.182 The Sameness of the Scythians The portrayal of the Scythians in the Histories constructs a people so profoundly different from the Greeks that they constitute the virtual embodiment of the other.183 Yet among the ethnea described by Herodotus, the Scythians are a particularly unstable paradigm of alterity: they are alien but also familiar. Their affinity with the Greeks is partially based on their shared experience of finding themselves on the receiving end of Persian aggression and being able to confound the efforts of that vastly superior power.184 But Herodotus’ implicit suggestion that the Greeks can recognize themselves in the Scythians—crude nomads, living in an area of the world where there is little to admire (see 4.46.2)—has a wider scope. It aims at displaying the sameness of what is most distant and foreign. It is part of his overall pursuit of worldwide cross-cultural links. The antecedents of Herodotus’ Scythians belong to an old and conflicted tradition of ethnographic representations of northern pastoralists that goes back to Homer. The idealized “hard primitive” first appears in a passage of the Iliad where Zeus averts his gaze from the painful battle of Trojans and Achaeans and looks into the distance, over “the land of the Thracian riders of horses and of the Mysians, who fight at close quarters, and the noble Mare Milkers, drinkers of milk, and the Abii, most righteous of men.”185 Theoretically warlike, but also just and naturally disinclined to bloodshed (a feature represented by their milk drinking), remote societies suggest relief from the troubles of the more civilized Homeric world of fighting and ships. In his discussion of representations of the Scythians down to his day, Strabo introduces the Homeric passage 182. On the allegorical modes of modern ethnography, see Clifford 1986. 183. See Hartog 1988, passim, especially 11. 184. Hartog (1988, 35– 38) gives a complete catalogue of the points of analogy between Herodotus’ narrative of Darius’ Scythian expedition and that of Xerxes’ Greek expedition. But see Immerwahr 1954, 262– 63. 185. 13.1– 6. See Lovejoy and Boas 1935, 288, citing Riese; Romm 1992, 53. Levy 1981; Marincola 1997b, 4. For “hard” and “soft” primitives according to the definition of Lovejoy and Boas, see “Differentiating from Within” earlier in the present chapter.

108

Telling Wonders

right after reporting a statement by Ephorus that testifies to contradictory ethnologic views in the fourth century b.c. Ephorus says . . . that the ways of life of the Sauromatae and the  and other Scythians are not all alike, for some are harsh [χαλεπους] even eat humans, while others abstain from eating any other living beings. The other writers, he says, talk about their savagery [περ ι   οτητος] της ωµ because they know that the terrible and wonderful are striking; but one should tell the opposite facts and make them into paradigms [παραδει γµατα ποιεισθαι], and Ephorus himself will therefore describe only those who follow most just modes of behavior [δικαιοτα τοις ηθεσι]; for there are some of the Scythian nomads who feed on mare’s milk and surpass all men in justice [τ η  η πα ντων δια␸ ερειν]  δικαιοσυν and are mentioned by the poets . . . (Strabo 7.3.9 C302 ⫽ Ephorus, FGrHist 70 F 42) This passage illuminates the ideological uses of the antiprimitivist and primitivist positions. While the cruel Scythians are savages in antithesis to whom we civilized people define ourselves, the just Scythians are the superior others, paradeigmata who teach us about our own shortcomings.186 We are here in the presence of a moralistic form of ethnographic discourse. It also briefly appears, for example, in Herodotus’ account of the mutually antithetical Androphagoi and Argippaeans, but Herodotus does not normally idealize his primitives, whether hard or soft.187 Ephorus would no doubt rank Herodotus’ Scythians among the savage 186. On the tradition of “inverse ethnocentrism” and “ethnologic satire” in the Greek representation of various peoples on the edges, see Romm 1992, 46– 48, 49– 60 (Ethiopians), 61– 67 (Hyperboreans), 67– 77 (Arimaspians and Scythians). See also Levy 1981, 57– 59. 187. Thus, for the traditionally idealized Hyperboreans (see Romm 1992, 61– 67), Herodotus is not certain of their existence and is only interested in a custom he can compare to another (4.33.5). His Androphagoi are, however, negatively idealized “hard” primitives (4.106). The Argippaeans, or Bald Men, are positively idealized “soft” primitives: they feed exclusively on the milk from their abundant flocks and on the fruit of certain local trees that also provide them with shelter. Their congenital baldness shared by men and women alike points to a utopian equality between the sexes; it also recalls the holiness of the shaven priests in Egypt and is symbolic of nonviolence: “No one among men does them wrong, for they are said to be sacred, and they do not possess war weapons. In the first place, they are the ones who settle disputes among their neighbors, and secondly, if someone who is a fugitive seeks refuge with them, no one does him wrong” (4.23.5). For the connection of hair with belligerence, see Xen. Lak. Pol. 11.3, 13.8; Plut. Lyc. 22; Tac. Germ. 38.4. See also Loraux 1977, 119. Cf. Hdt. 1.82.7 and 7.208.3. For equality between the sexes as an index of justice, see Herodotus’ Issedones (4.26.2).

Comparison

109

and cruel kind. Though they are not cannibals by custom any more than are the Persians or Medes (1.73.5; cf. 1.119.3), they drink human blood, make human sacrifices, and scalp their enemies. The point of Herodotus’ account, however, is objectively to represent the foreignness of the Scythians and at the same time to overcome it by promoting the audience’s discovery of their affinity with them. Herodotus’ work is made easier by the contemporary Greek ambivalence toward the Scythians as it emerges from fifth-century texts. Here the stereotype appears to be more complex than Ephorus’ simple opposition of good and bad Scythians. For the most part, the Scythians of fifthcentury political discourse are intractable xenophobes living at the extremities of the earth, but this very isolationism also makes them autarchic and impregnable.188 In the Eumenides, a passage reflects on what is desirable for the safety of a state by pairing up Scythia and Sparta in a sort of political kinship. When Athena proclaims that a new court of law will be established in Athens on the Areopagos, the hill where the Amazons attempted to establish a rival city (685– 93), she promises that this court will preserve rightful fear within and provide “a bulwark for the land and a means of salvation for the city such as no one among men possesses, not even among the Scythians or in the places of Pelops” (Aesch. Eum. 700– 703). Identification and separation are here operative at the same time. To the extent that Scythia is viewed as a society founded on a rigorous order, it is comparable to Sparta, on the one hand, and to Athens, on the other.189 Like Sparta and Athens, and in antithesis to the wild Amazons of this passage, the Scythians occupy a political space “neither without rule nor ruled by a master.”190 At the same time, the coupling of the land of Pelops with Scythia in the Eumenides points to their difference from Athens. Discipline and dedication to war make both communities, one 188. See Hartog 1988, 12– 13, on representations of Scythian savagery outside of Herodotus. Examples are Aesch. Prometheus Bound 709– 14; Aristoph. Acharnians 702– 3; the Scythian archer in Thesmophoriazusae 1070– 1175, discussed by Long (1984, 138). On the setting of Prometheus Bound, see especially Hall 1989, 113– 15. The Hippocratic treatise Airs, Waters, Places (17– 22) is idiosyncratic in representing the Scythians not as a “hard” culture but as a people made “soft” and sluggish by climate and effeminate by their riding practices. See Thomas 2000, 68– 71. 189. For Scythian paradigmatic justice, cf. Aesch. Sept. 727– 33.   190. Aesch. Eum. 696. Cf. frag. 198 Nauck, αλλ ι ππα κης βροτηρες ευνοµοι Σκυθαι (Strabo 7.301), where ευνοµοι could mean “possessing a just and well-ordered society.” See Levy 1981, 61.

110

Telling Wonders

Greek and one barbarian, invincible; but they will be no match for Athens, a city defended from within by a juridical body of righteous and godfearing citizens armed with ballots (707– 9). A society that is placed under the aegis of deliberation is here contrasted with the sort of order, common to both Scythians and Spartans, that privileges war.191 Herodotus’ representation of the Scythians is similarly based on an interplay between otherness and partial identification with Spartans or Athenians. This thematic seesaw is programmatically reflected in the first two semiautonomous narratives at the beginning of book 4, one (“Scythians blind their slaves,” 4.2) serving as a long ethnographic gloss within the other (“return of the Scythians from Asia,” 4.1.3, 4.3– 4). The latter is announced first by a summary introduction: when the Scythians came back to their land after ruling Media for twenty-eight years, they had to confront “a trouble not inferior to the Median trouble,” namely, opposition from their slaves, whom the Scythian women had married in their absence (4.1.3). The second narrative is attached at this point to the mention of slaves. With a shift to the ethnographic present, it reports how the Scythians, who are milk drinkers and nomadic, blind “whomever they capture” and put them in charge of milking their mares. Though the connection the discourse establishes between the Scythian method of milking, the practice of blinding the slaves, and nomadism is opaque, the mutilation has presumably functional aims.192 Scythian culture is at once comfortable and deprived, living in abundance (4.47.1, 4.53.2– 3, 4.58, 4.59.1) but with few resources. In the absence of a more painstaking sort of labor, the Scythians indiscriminately use whatever they have on hand.193 While pointing out their drinking of milk (a feature of a “good savage”), the text here also first brings to the fore the brutal pragmatism that is the principal mark of the alterity of the Scythians with respect to the Greeks. 191. Cf. Thuc. 2.97.5– 6, which praises the Scythians for their strength in battle but   disparages their ευβουλ ι α and ξυνεσις. For the lethal warlikeness of the Scythians, see also Aesch. Choeph. 161– 63. 192. See Macan 1895, 1, 2; How and Wells 1928, 1:303; Legrand 1946, 4:48; Benardete 1969, 100– 101; Hartog 1988, 18. Mutilation is normally envisioned as an affirmation of despotic power or as a punishment. See chap. 3, n. 54, in the present book. 193. Bones are used as firewood, the ox’s stomach as a cauldron (4.61). Human skin is fashioned into clothes, and hand-skins with the nails attached serve as lids for quivers; enemy scalps become napkins, and skulls become drinking cups (4.64– 65); as hunters, the Scythians do not always make fine distinctions between animal and human quarry (see 4.134.1; cf. 1.73). See Hartog 1988, 40– 44.

Comparison

111

The continuation of the story of the Scythians’ return juxtaposes next to the functional action we have seen (the blinding of the slaves) a symbolic action that offers a rather different perspective on Scythian culture. The sons whom the Scythian women had borne to the slaves tried to block the returning Scythians and confronted them in battle. The Scythians overcame the young men’s resistance only after hearing the following advice from one of their own. What are we doing fellow Scythians! Fighting with our slaves, we are ourselves killed and become fewer; and killing them, we will in the   ι τε κτεινοµενοι future rule over fewer men [αυτο ε λασσονες      γινοµεθα κα ι ε κει νους κτει νοντες ε λασσονων το λοιπον αρξοµεν]. I propose that we set aside spears and bows and that each take instead the whip for his horse and approach them. For so long as they see us with weapons, they will believe themselves to be equal to us   ι ων]; but if they see us and born from equals [ οµοιο ι τε κα ι ε ξ οµο with whips instead of weapons, they will understand that they are our slaves and, recognizing this, will not stand their ground against us. (4.3.3– 4) This episode entails a sudden shift of Hartog’s metaphorical mirror in Herodotus’ representation of the Scythians.194 The barbarians of the previous narrative are given to strange practices and almost unintelligible brutality in the pursuit of their elementary daily living; here they suddenly reflect in a direct way the Greek audience’s ideology of freedom and mastery over the symbolic forms of their status as free men. Freedom as a Scythian value will be a fundamental element in the historical narrative of Darius’ expedition. The Scythian king Idanthyrsus  claims to Darius that he recognizes as masters (νοµι ζω δεσποτας, 4.127.4) only his ancestor Zeus and Hestia, the queen of the Scythians; this parallels Demaratus’ statement to Xerxes that the Spartans have no other master than their law/custom (7.104.4). The Scythians hold the Ionians in contempt on the grounds that by the standards of free men, they 194. See the study and critique of Hartog’s approach and metaphor (Hartog 1988) in Dewald 1990, 218, 220– 21: according to Hartog, “by looking at how a Greek constructed the Other, we also see much more clearly how a Greek understood that which distinguishes the Same: hence the ‘mirror of Herodotus.’” But Dewald cautions: “the Same unexpectedly becomes the Other”; “to extend the governing metaphor that Hartog uses, Herodotus warns us . . . that his mirrors are not bolted on their walls.”

112

Telling Wonders

are the worst and most unmanly of beings, but “if one speaks of them as slaves, they are the most master-loving and nonrunaway of human stock” (4.142). This evaluation anticipates the Greek Artemisia’s discussion later on about good and bad slaves (8.68γ). Like no other people in the Histories besides the Greeks, the Scythians are represented as a people who define themselves in opposition to literal and political slaves. But the story of the Scythians’ return at 4.3– 4 conveys a more specific parallel. We find again the political situation whereby slaves take charge of the state and marry the citizen women in the absence of the men, in Herodotus’ story about the depopulation of Argos after the battle of Sepeia (6.83; cf. 6.77). It belongs to a pattern of historical traditions concerning Greek states like Sparta that have a system of slavery of the helot type.195 In the passage quoted earlier (see especially the underlined phrases), the rhetoric of the anonymous Scythian reveals the pointedly political character of the anecdote for a Greek audience.196 In contrast to their slaves, the free Scythians perceive themselves as “equals and born   from equals” [ οµοιο ι τε κα ι ε ξ ο µοι ων] (4.3.4). In the Histories, οµοιος  alludes to the political system of the Spartan Homoioi on at least three occasions.197 In two out of the three cases, somewhat as in the Scythian episode, the word helps to express the idea that valor in battle is connected with citizen status.198 All three cases allude to the Spartan ideology 195. The pattern is identified and analyzed by Vidal-Naquet (1981, with numerous examples), though he does not mention Herodotus’ Scythian story as an analogue. 196. It is considered a piece of political theorizing by Macan (1895, 1:3). See Finley 1980, 118– 19. Corcella (1993, 230) cites later parallels. 197. In the strictly technical Spartan sense, the word first occurs in Xenophon (Lak. Pol. 10.7; 13.1, 7; etc.), but as Finley (1968, 146) observes, this fact is not very significant for the meaning of the word earlier on. 198. At 7.234.2, Demaratus says to Xerxes that Sparta is a city of about eight thousand  men and that these are all οµοιοι to those who fought at Thermopylae; the rest of the  Lacedaemonians are brave men, though not οµοιοι. At 3.55.1, an interpretive gloss attached to the narrative of the Spartan expedition against Samos states that “if those of the  Lacedaemonians who were there that day had been equal [ οµοιοι ε γι νοντο] to Archias and Lycopes, Samos would have been captured.” Both passages exploit the ordinary sense of the word to refer to the Spartan code. See Shimron 1989, 61; 1979, 132. At 7.136.2, Xerxes replies to Sperthias and Boulis, who offer to expiate the Spartan murder of the Persian  heralds, that he will not be equal ( οµοιος) to the Lacedaemonians, who have overturned basic human laws. This passage (as also 3.55.1) is an ironical reference to the discrepancy between ideology and reality. Herodotus’ Xerxes (who of course does not come from a society of homoioi; see, e.g., 1.134.1), finds the Spartan notion especially amusing; see 7.103.1, where he brings forth a contradiction in the system by a joke on the Spartan kings being worth double. On the problem of inequality of performance and in the political sphere among the Homoioi, see Finley 1968, 147– 49.

Comparison

113

of equality among themselves and their moral and political superiority vis-a-vis ` everyone else. As an exclusive community of equals, the Scythians are implicitly “like” Spartan citizens. The display of whips suggests a method of psychological conditioning of slaves to which Sparta provides the closest parallel;199 the narrator’s assessment that handling their slaves was no less of an ordeal for the Scythians than was fighting the Medes (4.1.3) hides a comparison with Spartan difficulties with the helots after the Persian Wars. A Spartan model is also confirmed by the anonymous Scythian’s mention of the problem of becoming fewer,200 by the avoidance of any reference that may suggest private ownership of the blind slaves,201 and by the narrator’s corroboration later on of the report of a Scythian invasion into present-day Scythia in preference to traditions of autochthony.202 In a later section of the narrative, we also find the statement that beyond the river Gerrhos, and “as far to the east as the ditch dug by the children of the blind men,” are the so-called royal territories and “the best and most numerous Scythians, who consider the other Scythians as their slaves” (4.20.1). The narrative does not provide much help for reconciling the existence of the blind slaves—described in the ethnographic present at 4.2 and remembered here—with these newly mentioned putative slaves of the royal Scythians.203 The blind men are prisoners of war (and therefore possibly non-Scythians), while the others, who probably include the agricultural Scythians (4.54), are in a state of political subjection. Despite the lack of explicit coherence, it is at least clear  that we are supposed to envision a privileged ruling group of αριστοι in the eastern part of the country, and two different types and degrees of servitude, roughly on the tripartite model of Spartiates, perioikoi, and helots.204 While the Spartans are citizen-hoplites, the Scythians are cityless, horse199. On the systematic degradation of Spartan helots, see Myron of Priene, FGrHist 106 F 2 ⫽ Athenaeus 14.657d; Plut. Lyc. 28. See also Garlan 1988, 153– 55. 200. 4.3.3. On Spartan oliganthropia in the fifth century, see Figuera 1986, 165– 81.  201. Contrast, for example, Persian ο ι κεται (see 1.137.1), and cf. 4.72.1 on the absence of slaves “bought with money.” 202. 4.11, cf. 4.5– 10. On the tradition of the Dorian invasion, see Hall 1997, 56– 64. 203. On the historical, geographical, and logical difficulties of Herodotus’ description, see especially Macan 1895, 1:2– 3 and 14; Macan 1895, 2:1– 30; Kothe 1969, 71– 80. 204. Darius’ opponents appear only to include nomadic and free Scythians. See Hartog 1988, 194– 98. For a distinction between royal/free and other Scythians, see also 4.120, 4.59.1, 4.110.2.

114

Telling Wonders

riding nomads; they avoid pitched battles even against invaders of their own territory.205 But somewhat like the Spartans, the Scythians do not allow the enemy to interfere with their customary way of life, never flee, and will fight when necessary.206 The ethnographer, to be sure, must translate the notion of courage into Scythian cultural terms: they reward as an  ηρ  αριστος  αν (brave man) the warrior who has collected the greatest  number of scalps (4.64.1); they demonstrate what “they call ανδραγαθ ι η [valor]” by showing off to their guests cups made of the skulls of enemies killed in battle.207 But by emphasizing the social control that among the Scythians surrounds the warrior’s achievement, Herodotus again enhances the parallel with Sparta. Xenophon reports that at Sparta “anyone would be ashamed to take a coward into his mess or be matched against him in a wrestling match” (Lak. Pol. 9.4). Among Herodotus’ Scythians, those who have not killed any enemies may not partake of the local version of the symposium sponsored annually by the governor of each province: they sit apart from the others, dishonored and ashamed, while those who have killed a great number of enemies are invited to drink double (4.66). Regulation of drinking on the Spartan model here replaces anarchical excess and the habit of drinking straight wine normally attributed to the Scythians and even mentioned by Herodotus’ Spartan sources.208 Scythians and Spartans are also similar in certain aspects of their respective codes of communication. Herodotus’ Scythian narrative devotes special attention to the language of the Scythians, which is first of all peculiar to them and consistent with other aspects of their reduc205. See Hartog 1988, 50– 53. On Greek—especially Spartan hoplitic—arete (valor), see 7.104; Tyrtaeus frags. 12.13– 20 and 11.31– 34 W. In Plato’s Laches, Socrates bridges the gap between Greek commitment “to stay at one’s post and face the enemy and not run away” (190D– E) and the Scythian method of fighting by flight and pursuit, by recalling the Spartan performance at Plataea (191A– C). 206. See 4.127.1– 3. Flory (1987, 103) remarks that Idanthyrsus’ statement that what he does in this war is only “what I have been accustomed to do” parallels Demaratus’ explanation to Xerxes that arranging their hair when they are about to risk their lives is the Spartans’ custom (7.209.3). I discuss the Scythian and Spartan defensive conception of war in chap. 3, “The Evils of War.” 207. 4.65.2. Contrast Plut. Lyc. 22, which says that the Spartans thought “it was shameful to cut to pieces those who had conceded defeat.” 208. 6.84.1– 3. Scythian drinking was notorious: see Anacreon frag. 11b Page (⫽ PMG 356) in Athen. 10.427b and Theog. 825– 30. For Spartan moderation in drinking, see Xen. Lak. Pol. 5; Plato Laws 637a; Critias frag. 6 West (⫽ Athen. 432d– 433b); DK 88 B33 (⫽ Athen. XI 463E). The sources are collected and discussed by Fisher (1988). Cf. Plut. Lyc. 12.

Comparison

115

tionistic culture. Their simplified code is illustrated by the mouse, frog, bird, and five arrows they send to Darius in response to his request of earth and water. This message turns out to mean that since the Persians are not birds, mice, or frogs, they are trapped in Scythia and will be pierced by these arrows (4.131– 32). The flattening out of the syntax is here truly amazing, showing that Scythian communication is as unique as their entire way of life and system of warfare.209 To the extent, however, that their “language” is spare and concrete, it resembles Spartan discourse. The Scythian displays of whips, scalps, and skulls find their parallel in countless visual images the Spartans use to convey the idea of their power.210 It is perhaps not entirely by chance that the description of King Ariantas’ crater, which measures the Scythian population in an approximate “Scythian” way (i.e., by size rather than by number), attracts a comparison with the crater of the royal Spartan Pausanias (4.81.3). The Spartans in Herodotus dislike long speeches, literalize metaphors, and mistrust abstractions.211 The Scythians designate snow with the word for “feathers” because of the two objects’ resemblance (4.31.2), and they use an ancient akinakes as the sacred image of Ares (4.62.2). In both cultures, the use of signs as a device for economical communication goes hand in hand with an anti-intellectualistic attitude, contempt for a certain type of sophie, and a materially simple way of life. The analogy between Spartans and Scythians in the sphere of discourse comes to the surface of the text through the reported judgment of the Scythian sage Anacharsis: he decrees that the Spartans are the only ones   ολους of the Greeks not “busy pursuing all sort of cleverness” [ασχ . . . ε ς  πασαν σο␸ ι ην] and able “to send and receive speech with good sense”    ι τε κα ι δεξασθαι  [σω␸ρονως δουνα λογον] (4.77). Anacharsis’ words make a distinction between sophie (intelligence/cleverness/wisdom) and sophrosune (good sense/wisdom). The first, which here connotes theorization and fancy rhetoric, is rejected by both Spartans and Scythians but is, 209. See West 1988. Contrast metaphorical prophecies of the type “When a mule sits on the throne of Media . . . . ” (Hdt. 1.55), where decoding the message requires that one replace the sign “mule” with the correct referent, but the referent occupies the same synctactical position as the sign. 210. Examples are red cloaks and long hair. See Powell 1988a on Sparta’s use of the visual in communication. 211. See 3.46 (Spartan response to Samian refugees), 7.226.1– 2 (Dieneces), 7.135.3 (Sperthias and Boulis). See the Spartan economic use of words at 9.11.2 (cf. Thuc. 4.40.2). For Laconian brevity, see also 9.90– 91, and cf. Sthenelaidas at Thuc. 1.86.1.

116

Telling Wonders

implicitly, a specialty of Ionians and Athenians. Sophrosune, by contrast, is valued by both Scythians and Spartans and displayed by them through economical speech.212 The narrator identifies this Anacharsis narrative as an invention of the Peloponnesians (4.77). If we believe that Herodotus’ glosses of source are historical, this points again to the existence before his time of a Greek tradition representing the Scythians as in some way analogous to the Greeks and in particular to the Spartans.213 In relation to speech and other spheres of culture, Herodotus exploits existing traditions for the purpose of establishing his own implicit analogy between the two cultures. Starting from the early episode of the whips, he does so in a context not of positive or negative idealization but of an ostensibly objective and scientific ethnography. The analogy serves therefore neither to praise nor to disparage; rather, it forms a part of the broader message that trends of similarities invariably link ethnic groups with widely discrepant levels of culture and differing customs. The Scythians are not at any rate connected only to Sparta. The multiple traditions concerning the origins of the Scythians relate them to various other ethnea of the world.214 Hartog has thrown light on the implicit analogy Herodotus’ narrative establishes between Scythians and Athenians, especially owing to the nonhoplitic role both nations played in their defensive war against the Persians and to the nonhoplitic way with which the Athenians of Herodotus’ time chose to respond to the Peloponnesian invasions.215 Athenian strategy in the earlier and more recent past has come to define the Athenians culturally, just as the Scythians’ strategy in war, Idanthyrsus says (4.127.1), coincides with their custom in peace. A major interpretive gloss that connects Scythian strategy and nomadism even enhances the implicit analogy between Scythian 212. See Georges 1994, 146. Cf. the Spartan Archidamus at Thuc. 1.79.2, 1.84.3, 1.84.2. 213. On Anacharsis’ and the Scythians’ mistrust of speech, see also Pherecydes, FGrHist 3 F 174; Hesiod frag. 150 MW; Levy 1981, 60. For the Anacharsis legend and apophthegmata, see Kindstrand 1981. 214. See 4.5– 12. On the self-contradictory quality of these multiple accounts, see Hartog 1988, 19– 30. Greek Hellenocentric traditions that make barbarian peoples derive from characters of Greek mythology (see Bickerman 1952; Georges 1994, 1– 9) are again utilized by Herodotus for the purpose of relating barbarians and Greeks. See Corcella 1993, 232; Vandiver 1991, 169– 80. 215. Hartog 1988, 39– 40, 49– 51. Cf. especially 4.122 with 8.41, 4.127.2 and 4.120.1 with Thuc. 1.143.5.

Comparison

117

and Athenians by attributing to the Scythians sophie, the very quality they elsewhere despise.216 The Euxine Sea, where Darius’ expedition was directed, of all the   regions contains the most ignorant peoples [εθνεα αµαθ εστατα]. For within the Pontic region, we cannot mention any people as  excelling in cleverness/wisdom [σο␸ ι ης] or any learned man [λογιον  ανδρα], except for the Scythian people and Anacharsis. In the one matter that is of the greatest importance for man, the Scythian race  made the cleverest/wisest discovery of all that we know [σο␸ ωτατα       πα ντων ε ξευρηται των ηµεις ιδµεν]. For the rest, I do not admire it, but they have devised this one most important thing, that no one who goes against them can escape and it is not possible to catch them if they do not want to be found. For how could men who have no constructed houses or walls but carry their homes with them and are all archers, who live not from agriculture but from livestock, and who have their homes on carts not be invincible and impossible to deal with in battle? (4.46.1– 3). The sage Anacharsis is not here “Laconic” and sophron, as in the   Peloponnesian story considered earlier. He is rather a λογιος ανηρ. Through him, the Scythians partake of the sort of sophie consisting of knowledge and of the ethical wisdom that derives from it, which are elsewhere displayed by the Athenian Solon and possessed in the highest degree by the Egyptians.217 The Scythian people as a whole, however, possess at least sophie of a cunning sort, which is also considered “Athenian.”218 The Scythians are nomadic and primitive, and these two ethnographic characteristics that most identify them as other with respect to the Greeks in general also create the preconditions for their sameness. So far as they are nomadic, with the peculiar strategy that their way of life entails, they resemble the Athenians, autochthonous city dwellers. They resemble the Spartans in their social war ethics and spare way of life, because they are primitive. At the same time, they are clearly neither Spartans nor Athenians nor Greeks, just as they are not Egyptian or even, 216. For the stereotype of the lack of intelligence of northern people, see Arist. Pol. 7.6.1 (1327b23– 25); Hall 1989, 122.  217. See 1.29– 33 (Solon), 2.77.1 (the Egyptians as λογι ωτατοι). 218. I am following here the distinction made by Dewald (1985, 52– 55) between two kinds of knowledge that Herodotus represents.

118

Telling Wonders

as Herodotus is at pains to specify, Budini or Geloni. They are entirely themselves, that is, Scythians. Identification with The Other: Anacharsis and Scyles An ethnographic-historical sequence tells about how first Anacharsis and later Scyles were killed by the Scythians for adopting foreign customs. This set of stories illustrates the role of implicit analogy in forcing the audience to reflect themselves in an alien people.219 Herodotus attributes one trait to all ethnea of the world: subservience to the constraints of their own culture. Formulated in abstract terms in the far-ranging interpretive gloss at 3.38, Herodotus’ theory of universal cultural chauvinism is borne out again and again by specific cases. Thus, a gloss of interpretation discloses the meaning of the stories of Anacharsis and Scyles by  introducing them as evidence (δι εδεξαν) that also [the Scythians] utterly avoid following foreign customs, both those of other peoples and most especially those of the Greeks.
    [ξενικοι σι δ ε νοµαι οισι κα ι ουτοι α ι νως χρ ασθαι ␸ευγουσι, µητε      τεων αλλων, Ελληνικοι σι δ ε κα ι ηκιστα.] (4.76.1) Adverbial κα ι (amounting to a gloss of similarity) perhaps refers first and foremost to the Egyptians. They are an implicit term of comparison throughout the Scythian narrative, including in the immediately preceding passage on Scythian σο␸ ι η.220 Unlike the chauvinism of the Egyptians, however, the Scythians’ dislike of foreign nomoi manifests itself in the summary violence that characterizes their culture as a whole. The two narratives that show this ferocious protectionism differ, however, in one important respect: the story of Anacharsis promotes identification with the victim; that of Scyles, with the Scythians. The Scythian sage Anacharsis embodies Spartan sophrosune in the subsidiary Peloponnesian narrative already considered (4.77). In the story about his death, he is connected with Athenian sophie (4.76). Here, like the Athenian Solon, he travels all over the world for the purpose of sightseeing 219. The set of narratives is discussed by Hartog (1988, 62– 84). 220. See Benardete 1969, 99. For Egyptian chauvinism, see the totalizing statements at 2.79.1 and 2.91.1, as well as specific evidence, e.g., at 2.41.3. On Scythia and Egypt, see n. 91 and corresponding text earlier in the present chapter.

Comparison

119

and acquires (or displays?) much wisdom.221 On his way back to the 222 he   “haunts,” or “customs,” of the Scythians (ε ς ηθεα τα Σκυθεων), lands at Cyzicus, where he makes a vow that if he returns home safe, he will celebrate the Mother of the Gods as the Cyzicenes were doing. He does so in the relative privacy of the Hylaea, but the Scythians spy on him, and their king kills him with his bow (4.76.2– 5). The story ends with a unique explanatory gloss self-referentially marked by historie: “As I heard from Tymnes, a deputy of king Ariapeithes, Anacharsis was the uncle of Idanthyrsus, king of the Scythians, and the son of Gnurus, son of Lycus, son of Spargapeithes.” From this genealogy, which the source Tymnes has provided as if oblivious to its $ implications, the narrator draws his own inference: “If, then [ων], Anacharsis belonged to this family, let him know [ιστω] that he was killed by his brother. For Idanthyrsus was the son of Saulius, and it was Saulius who killed Anacharsis” (4.76.6). The pathetic appeal to the dead Anacharsis in the third-person imperative constitutes the closest thing we find in Herodotus to the Homeric apostrophe to a character in the second-person singular.223 Without making recourse to explicit evaluation, the narrator underlines for his audience the horrible fate of a wise and pious man, caused by the ferocious intransigence of a savage people. We recall the typical Scythian of the ethnography, showing off to his guests the skulls of family members with whom he has feuded (4.65.2).  In the structurally similar story of Scyles that follows (παραπλησια, 4.78.1), a shift in perspective renders Scythian protectionism less alien from a Greek point of view. Unlike Anacharsis, Greek-raised Scyles does  ουδαµ   ηρ  εσκετο  not like the Scythian way of life (διαι τ% η µ εν ως  Σκυθικ% η). His Hellenization, which he pursues during surreptitious visits to the Greek city of Olbia, involves different areas of custom, both secular and sacred (4.78.4). In the sphere of the sacred, it culminates in Scyles’ desire to become initiated to the rites of Dionysus. But rather than promoting sympathy for Scyles’ piety, here the text sends negative signals.  µενος καταυτ  θεωρησας    ην  σο␸ ι ην πολλην  (4.76.2), 221. In γην πολλ ην κα ι αποδεξα  the second participle could derive either from αποδε ι κνυµαι, “display,” or from the far less   frequent αποδ εκοµαι, “receive.” The ambiguity is enhanced by the implicit analogy with Solon, whose sophie derives from his “sight-seeing” or (θεωρ ι η, 1.29.1, 1.30.2), but who is also represented as displaying his sophie to Croesus. The same is true for the narrator,   οδεξις) whose work is a display (απ of historie conducted all over the world. 222. See Hartog 1988, 65; cf. 4.80.1. 223. See, e.g., Il. 1.146. See also de Jong 1987, 13, 60.

120

Telling Wonders

At the metannarrative level, a gloss of anticipation of doom directs the audience to interpret Scyles’ initiation to Dionysiac religion as the precipi tating event, or προ␸ασις, in a career already bent on self-destruction, “since he was bound to incur a bad end.”224 The narrative, for its part, reports a sign of divine disapproval: while Scyles is preparing to undergo the ritual, “the god” strikes Scyles’ house in Olbia with a thunderbolt and sets it on fire, a warning that Scyles ignores (4.79.1– 2). In the tradition of resistance myths, which will eventually produce the alien and threatening Stranger of Euripides’ Bacchae, Dionysus comes to a Greek city from abroad, displays his power to those who resist him, and exacts worship from the entire population.225 In the story of Scyles, a foreigner embraces Greek Dionysus, and the divine opposes resistance. Here the author of the portent is not the god whom the Scythians refuse to recognize but a less culturally determined entity.226 Almost as a divine representative of the “Custom king of all” of 3.38.4, this power objects to an individual’s asocial adoption of a religion not his own by nomos and repulsive to his people as a whole. The Scythians reject Dionysus for reasons that are strikingly similar to those of the god’s opponents in the Greek resistance myths. A cognitive statement added as an explanatory gloss to the narrative reports, “the  to go find [ε ξευρι σκειν] Scythians say it is not reasonable/natural [ο ι κος] a god like this one, who drives people mad.”227 Described in these terms, the Dionysiac cult appears as un-Greek—or, as Herodotus himself else where acknowledges, not “consistent” (οµοτροπον) with the rest of 228 Greek culture —as it is un-Scythian. The same thing could indeed be said about the cult of the Mother of the Gods, which Anacharsis adopts in the preceding story.229 In the Scyles narrative, however, a new empha γενεσθαι.  224. 4.79.1: εδεε ο ι κακως For this type of interpretive intervention, see chap. 1, n. 64, in the present book. 225. See Guthrie 1950, 167– 70. In Herodotus’ version of one of the resistance myths (9.34), Dionysiac madness causes a social and political crisis in Argos. 226. It is Zeus according to most commentators. See Linforth 1928, 219; Hartog 1988, 73. 227. 4.79.3. On cognitive statements, see in the present book, chap. 1, n. 40; chap. 3, “Cognitive Statements and Polemical Negations.” On the negative sense of ε ξευρι σκειν, see n. 35 and corresponding text in the present chapter. 228. 2.49.2. For Dionysus in the Bacchae as a symbol of “the blurring of distinctions between Greek and barbarian,” see Segal 1982, 124. 229. For the parallel roles of these two Greek divinities in the Anacharsis and Scyles narratives, and for their equal ambivalence in the Greek perception, see Hartog 1988, 70, 72, 74– 83; Hall 1989, 153– 54.

Comparison

121

sis appears on the irrational aspects of the divinity in question,230 Scyles lacks a good motive for his initiation and disregards a divine sign, and the Scythian criticism of Dionysus sounds reasonable even from a Greek point of view. All these factors force the listeners to blame Scyles as they would not have blamed Anacharsis. Like Anacharsis, Scyles is finally decapitated by his own brother (4.80.4– 5). Here, however, the narrative delays the last act of the drama by a series of functions focalized through the Scythians as an ethnos and body politic. It represents the humiliating mockery to which they are subjected by a foreigner (4.79.4), the grief they experience in the face of  µεγαλην ε ποιησαντο,  their own king’s display of madness (συµ␸ορην 4.79.5), their deposition of Scyles and appointment of a new king, and the negotiations with a foreign power for the extradition of the criminal (4.80.1– 4). By the time the characteristic ferocity of the Scythians even against family members is fully brought back to the fore with Scyles’ decapitation, the Scythians have emerged almost as a civilized and earnest  community, concerned with preserving (περιστελλειν in the conclusion at 4.80.5) their order and integrity. The situation recalls the alarm the Spartans experienced at the adoption of foreign ways and erratic behavior by one of their own who was almost a king. Herodotus has sufficiently “turned the mirror” so that the Greeks of the audience might recognize themselves in the Scythians. The Pausanias model is not a far-fetched subtext to this story.231 The secular and more strictly “civic” side of Scyles’ adoption of foreign nomoi is represented by Scyles’ change of dress, building of a house in Olbia, and marriage to a Greek woman (4.78.4– 5). These cultural shifts go hand in hand with a separation of Scyles from his people, which is also, since Scyles is the king of the Scythians, a stepping down from his royal position. Thus, every time Scyles leads the army to Olbia, he leaves it outside the city, and after going inside the wall and closing the doors, he walks about the marketplace dressed like an ordinary Greek, “not accompanied by bodyguards or anyone else” (4.78.4). But though Scyles intermittently declines to be a king at home and becomes a private citizen abroad, his behavior fits in the symbolic pattern of monarchy we have described in our discussion of analogy in the history. The verb referring to his irrational desire to become initiated into 230. This is noticed by Hartog (1988, 75 n. 47). 231. See “The Seers of Plataea” earlier in this chapter.

122

Telling Wonders

 Dionysiac religion, ε πεθυµησε, is a term of the monarchical code.232 In response to Scyles’ behavior in the sphere of religion, the divine thunderbolt incinerates precisely the house in Olbia “that I have mentioned shortly before this,” as the narrator recalls to reemphasize the main symbol of Scyles’ civic violation. Since it was “a big and rich house, around which stood sphinxes and griffins made of white stone” (4.79.2), this supposed index of Greekness would signal to Herodotus’ audience a northern variation on the theme of Oriental extravagance.233 As Artabanus will say to Xerxes, it is against the biggest houses that the god strikes with his thunderbolt (7.10ε). The transition from nomadic king to Greek polites is therefore made to resemble a climb to despotic rule in the monarchical pattern of the history, with the same sort of metaphorical paradox that equates Tisamenus’ acquisition of Spartan citizenship with an acquisition of the kingship.234 This time, like Pausanias, who would roam in Byzantium in Persian dress and make plans to marry a Persian woman, and like Deioces, enclosed in his new royal palace, Scyles comes to embody the alienation of the individual from the community, an alienation interpreted as some form of self-exalting lapse into barbarism. Through symbols and vertical analogy, the monarchical pattern can even be stretched to apply to a case like this one, where the literal kingship actually represents the traditional good order of society, while the potentially disruptive element is an individual who excludes himself from this institutional status. Scyles, moreover, becomes a foreigner to the Scythians by becoming Greek, yet the way in which he does so resembles a barbarization in the Greek sense. Placed in the middle of a logos where the representation of a profoundly alien people is carefully designed to show the Greeks intermittent glimpses of their various selves, the Anacharsis/Scyles sequence ends up demonstrating that what from the Greek point of view represents the ignorant rejection of their own civilized customs on the part of the barbarians is really analogous to the Greek contempt for barbarian practices that are repulsive to them. As a narrative of past events concerning a foreign people’s customs  (to desire) in Herodotus, two refer to 232. 4.79.1. Of twenty-five cases ε πιθυµεω sexual lust and the instinct of animals, respectively. One occurs in Artabanus’ gnomic saying that it is wrong to desire many things (7.18.2). Of the remaining twenty-two occurrences,  qualifies Marfifteen describe monarchic or aggressive desires. At 7.6.1, ε πιθυµητης donious’ imperialistic ambitions. 233. See Macan 1895, 1:53; Hartog 1988, 73. On the austerity of Greek private houses in the fifth century, see Morris 1998, 67– 75. 234. See “The Monarchical Model in Sparta” earlier in this chapter.

Comparison

123

and institutions, moreover, the Anacharsis/Scyles unit also reveals the congruence of Herodotus’ analogical interpretation of the world in the sphere of history, on the one hand, and of ethnography, on the other. Synchronically, the story of Scyles conveys the attachment of any society to its own cultural values, whatever these may be, as well as the similarity of some of those values among Scythians and Greeks. Diachronically, through the monarchical model, it shows the danger of social and cultural change in a familiar form: in one way or another, such a change has overtaken the Lydians and the Persians and threatens the Greeks as well. Similarities and overlaps counterbalance the synchronic diversity of humankind, so that we are able to discern similarities and overlaps among historical processes that unfold at different times and in different settings of the world. We can make the experience of others our own because, in some respects at least, the others are same. The Other Is Same: Making Peace with the Amazons That the other is same constitutes part of the underlying message of another ethnographic-historical narrative in the Scythian logos, the story of the birth of the Sauromatian nation from Scythians and Amazons.235 Since the Greek concepts of alterity and self in the fifth century are closely related to a political ideology of power, we should not be surprised to find that Herodotus’ narrative combines the reassessment of such views with a recipe for correct foreign-policy relations. The exemplary behavior of the protagonists, who are strictly collective and non-Greek, suggests the prescriptive character of the Amazon story in Herodotus, in contrast with the predominantly epideictic function of the Anacharsis/Scyles sequence. The Amazons are hardly a politically neutral subject. Unlike all the other peoples mentioned in the Histories, they are an extinct race, and the text reports no logos, reliable or unreliable, that assigns to them a place in Herodotus’ map of the present-day oikoumene.236 They are characters of Greek traditions about the remote 235. 4.110– 17. The most important discussions of the episode are by Dewald (1981) and Hartog (1988, 217– 24). See also Flory 1987, 108– 13. According to Cole (1967, 143– 45), the story derives from a fifth-century source that bears connection with the source of the social history in Polyb. 6. Such an origin is well suited to the different but nevertheless theoretical use Herodotus makes of the anecdote. 236. See especially Strabo 11.5.4. Cf. Diod. 17.77.1– 3. On the extinction of the Amazons, see Lysias 2.6.

124

Telling Wonders

past. As far as we are able to determine, they play only one role, that of being repeatedly defeated, subdued, or captured by various heroes, like other bestial or lawless opponents of the mythical ancestors of the present-day Greeks.237 In the political discourse of the fifth century, the story of the Amazons becomes a prominent mythical model for the affirmation of Greek superiority vis-a-vis ` the barbarians. Eventually Athens appropriates it as a charter for the justification of its imperialistic policy after the Persian Wars.238 The visual art of the period depicts the Amazons side by side with Centaurs, Giants, Trojans, and, at least on one occasion, the Persians and Greek enemies of recent history, to represent the forces of disorder confounded by the representatives of civilization—Lapiths, Greek heroes, Olympic divinities, or modern Greek hoplites, as the case may be.239 The ideological message encoded in the combination of these unrelated struggles on a single monument is based on the well-established polarities and analogies between Greek and Barbarian, male and female, human and animal.240 After the Persian Wars, the myth most prominently includes the Amazon invasion of Attica for the purpose of conquest, an invention that seems to have supplanted the preexisting tradition of heroic expeditions to their part of the world, to serve as the antecedent and analogue of 237. For the Amazons versus Bellerophon, see Il. 6.152– 206; Pind. Ol. 13.63– 92. For the Amazons versus Heracles, see Pind. Nem. 3.36– 39; Pindar fr. 172 SM; Hellanicus, FGrHist 4 F 106; Eur. HF 408– 18; Eur. Ion 1144– 45; Apollod. 2.5.9. For the Amazons versus Theseus and Heracles, see Eur. Heracl. 215– 17. For the Amazons versus Theseus, see Pherec., FGrHist 3 F 15, F 151 and F 152; Pindar fr. 176 SM; Eur. Hipp. 10, 305– 9; Plut. Thes. 26– 28. See Gantz 1993. 238. See Tyrrell 1984; Loraux 1986, 147– 48; Tyrrell and Brown 1991, 159– 215, especially 198– 200; DuBois 1982; Castriota 1992, 43– 58; Holscher ¨ 1998, 167. 239. Attic Amazonomachies appeared paired with a Centauromachy in the paintings of the Theseion at Athens; in the Stoa Poikile, juxtaposed to depictions of captured Troy, the battle of Marathon, and the battle of Oinoe against the Spartans (see Paus. 1.15, 1.17; Aristoph. Lys. 672– 80); on the western metopes of the Parthenon (a Centauromachy, a Gigantomachy, and an Iliupersis are on the other three sides); and on the outer surface of the shield of Pheidias’ Athena Parthenos, with a Gigantomachy on the inner surface (see Plut. Per. 31.4). See especially Castriota 1992 (33– 63, 76– 89, 134– 83) and the iconographic study by von Bothmer (1957). Amazonomachies also appeared on the Athenian treasury at Delphi (juxtaposed to other exploits of Heracles and Theseus; see DuBois 1982, 57– 71), on the temple of Apollo Daphnephoros at Eretria, on the temple of Hephaestus in Athens. The temple of Apollo at Bassae in Arcadia, built probably in the last quarter of the fifth century by the architect of the Parthenon, Ictinus, testifies to the popularity of the myth outside of Athens. See DuBois 1982, 64– 66. 240. See Thales (or Socrates) in D.L. 1.33, quoted by DuBois (1982, 4– 6).

Comparison

125

the recent barbarian aggression.241 The passage from the Eumenides that compares Scythians and different groups of Greeks also mentions the Amazons; it shows how they could be invoked when speaking of the dangers that threaten the polis from abroad and of the hubris of her opponents.242 The Amazons are first of all presented as antithetical to the pious and law-abiding Athenians, whose new court will sit precisely on the Areopagos, which the Amazons once occupied. They are also contrasted to the mutually analogous Spartans and Scythians, both of whom rely on war, rather than on political institutions, for their defense, but who nevertheless, unlike the Amazons, possess civic order. Herodotus’ Sauromatian logos does not contradict the notorious “historical” events featured in the myth, but it radically revises its ethical slant. Greek authors of Herodotus’ time unanimously represent the Amazons as the fulfillment of female nature out of control: because they were aberrant and wild, driven by lust of domination, and posed a threat to the civilized world, one needed to fight and conquer them.243 Herodotus, by contrast, portrays the Amazons as a people who possessed certain peculiarities but otherwise were not all that alien from other ethnea or abnormal with respect to the moral sense of the rest of humankind, Greeks 241. See Castriota 1992, 46– 47. In Aesch. Suppl. 234– 37, 277– 90, the Amazons represent a generalized type of aggressive barbarian women hostile to marriage. Aristoph. Lys. 672– 80, where the chorus compares the rebellious Greek women to both the Persian ally Artemisia and the Amazons as depicted by Micon, confirms the public’s interpretation of the Amazon invasion on their city’s monuments as the analogue of the Persian invasion, as well as conveying the idea that the Amazons embody the threat of the female. For the gender-related aspect of the myth, see especially Tyrrell 1984, 22, 113– 28. In fourthcentury oratory, the Athenian defeat of the Amazons in Attica is a conventional topos of the list of glorious Athenian achievements that culminates in their historical defeat of the Persians. See especially Lysias Epitaph. 2.4.– 26; Isoc. Paneg. 4.68– 70; Demos. Epitaph. 60.4– 8. The topos goes certainly back at least to the time when Herodotus was composing his work, as is demonstrated by its occurrence in his version of the Athenian speech at Plataea (9.27.4). On the tradition of the epitaphios in Athens, see Loraux 1986; Tyrrell and Brown 1991, 189– 215. 242. Aesch. Eum. 681– 706. See especially 685– 93 (cited earlier in this chapter), where the Amazons are the archetypal enemy of the polis. In Aeschylus’ Oresteia, the Amazons are also models for the man-killer Clytemnestra (Agam. 1231– 32; Eum. 625– 28). See Zeitlin 1984, 160– 70, especially 163. 243. Hellanicus mentions that the Amazons removed their right breasts, and he calls them “a golden-shielded, silver-axed, female, male-loving, male-infant-killing host” (FGrHist 323a F16 and F17); cf. Hippocr. On Joints 53 [see Thomas 2000, 61– 62]). Other details about Amazonian society do not appear in fifth-century sources and may be later elaborations.

126

Telling Wonders

included. This representation is consistent with Herodotus’ approach to foreign peoples throughout the ethnographies, but as far as our evidence for the Amazons goes, it is unique in literature.244 In the logos reported by Herodotus, a group of Amazons have survived the defeat at the Thermodon River (one of the traditional events in the myth) and are carried off by the Greeks as prisoners. They prove true to their Scythian name “Mankillers” (4.110.1) by dispatching their captors on the ships. Since they find themselves without pilots and are inexperienced in navigation, they drift about until they land on the shores of the Maeotis Lake.245 Here they make their way toward the inhabited area, take possession of some horses, and start raiding the territory (4.110.1– 2). This accidental but aggressive arrival of an Amazon contingent in Scythia seems designed to replay in reduced and modified form the story of the Amazons’ invasion of Attica, which is not mentioned. Replacing the Greeks, the local inhabitants naturally undertake to defend their land. The theme of fighting goes hand in hand with the question of difference. In wonder at the language, dress, and ethnic identity of the Amazons, the  Scythians cannot “put the matter together” [συµβαλεσθαι το πρηγµα] (4.111.1). They cannot, that is, do what the narrator of the Histories frequently does in the course of his ethnographic and geographic research: find grounds for comparing a new phenomenon with something already experienced, to conjecture about its nature. The Scythians perceive the Amazons as entirely different from themselves, except for thinking that the Amazons are young men. Eventually, when they realize from the bodies of the Amazons dead in the battle that they are really women, the Scythians discover a difference that suggests complementarity rather than conflict. They stop the fighting immediately and send a group of young Scythian men, wishing to have children from the Amazons. The differences between the two ethnic groups that now confront one another are considerably reduced. Both are detachments from their respective societies, equal in number (4.111.2) and occupying the same marginal space in the wild.246 They resemble one another in appearance, 244. In art, it is perhaps paralleled by the beautiful Amazons from the sanctuary of Artemis at Ephesos. These, according to Holscher ¨ (1998, 173), “represent an act of selfassertion against overpowering Athens.” 245. These Amazons’ inexperience of ships and navigation points to their idealized primitive status. See Romm 1992, 74– 75. Herodotus’ other Amazon figure, Artemisia, is the opposite (see 7.99.3). See chap. 4, “Vertical Analogy.” 246. On the ephebic character of Scythian youths and Amazons, see Hartog 1988, 217, 219– 20 (cf. 54– 55); Tyrrell 1984, 64– 87.

Comparison

127

since the Amazons look like young men and these particular Scythians are   πρωτην   young men (ανδρας τ ην η λικι ην εχοντας, 4.110.1; τους  νεωτατους, 4.111.2). The boys have, moreover, received instruction “to do what [the Amazons] do”: withdraw when they pursue and draw near when they stop, thereby avoiding all hostile confrontation (4.111.2). Just like the Amazons, the young Scythians have no material possessions ex εζωον τ ην  cept their horses and weapons: both lead “the same life” [ζοην  ην]  of hunting and plundering (4.112). The assimilation of the two αυτ groups parallels their physical rapprochement.247 Their camps come closer and closer each day, and at one point a meeting ensues between one of the Scythians and one of the Amazons. Other individual encounters follow (4.112, 4.113.1– 3). Eventually all the young Scythians pair up with the Amazons, and the two groups, “having joined their camps, permanently live together, each man having as wife the woman to whom he had originally joined himself” (4.113.3– 4.114.1). Fusion and the acceptance of complementarity entails further assimilation and compromise on both sides. The men do not become women and the women men, as some scholars have maintained.248 Reversal is to some extent inherent to the conception of a masculine woman, but within the boundaries of this idea, Herodotus radically modifies the system of polarities in the myth. The Amazons and the young Scythians achieve a society without the inequalities of conventional marriage, since both groups play the male role in identical fashion, while the necessary (though on principle undesirable) female functions are distributed between them across gender lines. The Amazons first of all consent to femininity in the sphere of sexual relations. The words by which the narrator designates intercourse unmistakably connote the submission of the passive partner, in conformity to the asymmetrical Greek conception of the sexual union.249 The Amazon of the original one-on-one encounter proved willing to become a little less Amazonic and culturally male when she “allowed the young Scythian to  have his way with her [περιειδε χρησασθαι]” (4.113.1). In the final stage of the fusion of the two ethnea into one, “the remaining young men 247. The action at 4.111– 15 is marked by alternate verbs of separation and approaching. 248. See DuBois 1982, 36; Tyrrell 1984, 42; Brown and Tyrrell 1985; Hartog 1988, 216– 24. All the scholars here cited emphasize the pattern of reversal, though Hartog realizes the difficulty of applying it to this episode. 249. See Halperin 1990, 29– 36.

128

Telling Wonders

 tamed [ε κτιλ ωσαντο] the remaining Amazons” (4.113.3). In this particular area of behavior, there is really no alternative to the female’s assumption of a role that Herodotus, like his Greek contemporaries, interprets as subordinate.250 Strabo’s Amazons will react to this necessity by resorting only to momentary and furtive unions for the sake of reproduction; those of Diodorus compensate by keeping their mates in an inferior—even crippled—state in all other respects.251 Herodotus’ Amazons find a less radical solution. In the tradition of heroic myths, marriage to an Amazon means primarily conquering her in war, subjecting her in a social as well as in a sexual sense, and taking away her Amazon identity by integrating her into a patriarchal order. Thus, when Heracles kills the Amazon queen for the sake of her belt, this symbolically prefigures Theseus’ abduction to Greece of the Amazon Antioche or Hippolyte, whom, according to the earlier versions of the myth, he has raped.252 In Herodotus’ story, the Amazons’ concession to the physiology of sexual intercourse leads to a stable union. Their subjection, however, is both voluntary and limited to the sexual sphere. It is compensated by the assumption of a share of the social female role on the part of the men, who agree to leave their parental homes, bring the equivalent of a dowry, and be monogamous (4.114.1– 4.115.1). Both groups equally assume the masculine social role, just as had been the case before the fusion: “the women of the Sauromatae follow the old way of life [διαι τ% η]; they regularly go to the hunt, both in the company of the men and separately from them, and also go to war and wear the same dress as the men” (4.116.2). In the Scyles narrative, the national dress is the symptom of a more profound and insidious cultural change. At the beginning of this story, it represents an external and deceptive sign of alterity (4.111.1). The eventual assimilation of ideologically unproblematic features, such as dress and language (4.113.2, 117), is here made 250. See Rosellini and Sa¨ıd 1978, 999– 1000. Brown and Tyrrell (1985), who want to interpret the story as representing the triumph of the female over the male, consider the use  of the word ε κτιλ ωσαντο as a contradiction. 251. Strabo 11.5.1; Diod. 2.45.1– 3, 3.53.1– 2. 252. Eur. Hipp. 305– 9. Cf. the rape depicted on the pediment of the temple of Apollo Daphnephoros in Eretria (ca. 510 b.c.), of which the scene on the amphora of Myson is probably an imitation. See Tyrrell and Brown 1991, 166– 67. Theseus’ capture of the Amazon is mentioned by Plutarch (Thes. 26, on the authority of Pherecydes and others; cf. Pherecydes, FGrHist 3 F 151 and F 152) and by Pindar (fr. 175). The Amazon’s zoster carried off by Heracles in his ninth labor (Apollod. Bibl. 2.5.9) is a war belt but has ambiguous sexual connotations.

Comparison

129

possible by the fact that in a more fundamental sense, the ethnos that appeared to be alien was actually both similar and complementary. Thus, the fusion of the two peoples, based on their resemblance, preserves their mutual equality of status and their integrity. Among the Scythians, no warrior will partake of the distribution of spoils if he does not bring a head to the king (4.64.1), and he is excluded from the annual symposium if he has killed no enemies (4.66). An analogous nomos forbids a girl of the Sauromatae to marry before killing an enemy (4.117). This implicit parallelism means that for the Sauromatian adult women, not only is marriage conceived as a privilege, but their social prestige and full integration into the society is linked to their contribution as warriors, not wives.253 The report adds that some of them die unmarried in old age because they have not been able to fulfill the nomos; this testifies to the determination with which the women of the Sauromatae guard their Amazon identity as much as possible in a society founded on marriage. In Greek society, marriage, for a woman, not only is the counterpart of war but also excludes war. The Amazons in Greek tradition make war and reject marriage. The Amazons of Herodotus’ Sauromatian logos have accepted marriage as a carefully circumscribed change in their customs. Finally, among the Sauromatian female descendants of the Amazons, marriage remains secondary with respect to the Amazon activity of war. If the cultural differences between Amazons and Scythian unmarried men prove ultimately to be imaginary or unimportant, the gulf separating the Amazons from Scythian women (and therefore also from the Scythian men who are married to them) is real and insurmountable. Because they are better ethnologic observers than are the Scythian youths, the Amazons realize this clearly. When their new husbands propose that they all stop leading “this sort of life” in the wild and go and live with the multitude of Scythians, where the young men have families and property, the Amazons describe their difference in a series of negations: “We could never live with your women, because we do not have the same customs   α νοµαια]. [τα αυτ We fight with our bow and spears and did not learn women’s work. Your women do none of the things we have mentioned but rather do women’s work sitting inside their carts and do not go out to 253. The Hippocratic Airs, Waters, Places (17), describes a similar initiation for Sauromatian women as Herodotus does, but with the crucial difference that marriage puts an end to their war activity.

130

Telling Wonders

 the hunt or anywhere else. We could not get along [συν␸ ερεσθαι] with them” (4.114.3). Aside from the fact that they live in wagons instead of houses, the Scythian society of families is here described in Greek terms—women indoors and men outdoors (cf. Xen. Oec. 7.23). The Amazons differ from the Scythians in this one important respect. The articulate and civilized character of the Amazons’ deliberations with the Scythian young men, the lucidity of the Amazons’ reasoning, and their appeal to conventional  and shared notions of justice (see δικαιοτατοι at 4.114.4) are designed to revise the traditional notion that the Amazons are wild women and the generalized negative other, the opposite of normality in every way. Nevertheless, the one peculiarity that identifies them as an ethnos cannot be imported into a patriarchal culture. Peaceful cohabitation between Amazons and the Scythian young men is possible, but not within the society of Scythian families. “Let us live on our own,” say the Amazons, and their husbands are persuaded.254 Herodotus’ Sauromatian logos conveys two lessons. First, on objective consideration, such as both the ethnographer himself and the characters of this story are able to exercise, the other may turn out to be same. By exposing as illogical the Amazon myth that masculine women are antithetical (rather than analogous) to men, the story undermines the notion of alterity for which the myth stands. Second, it prescribes that a real incompatibility is to be dealt with peacefully through separation. The exemplary behavior of the characters in fact denies the ideology, part and parcel of the traditional Amazon myth, that war is necessary because the other is the enemy; if we do not conquer it, it will conquer us. This inescapable alternative, conquer or be conquered, is mentioned by Xerxes to justify his expedition against Greece and is formulated again by Cyrus at the end of the work.255 But to Herodotus, war is an evil second only to “intertribal” struggle (στα σις . . . εµ␸υλος).256 In the Sauromatian logos, the collectivities involved, though all warlike to an extreme degree, seem remarkably determined to avoid both war  254. 4.114.4– 115.1. Notice the noncoercive nature of the deliberations: ε ι βουλεσθε    (4.114.4); ε πει θοντο (4.114.4, 115.3); ε πειτε αξιο υτε, ␸ ερετε (4.115.3). 255. 7.11.3, 9.122.3. See Raaflaub 1987, 228, for varied formulations of the same idea in Herodotus, Thucydides, and other texts. The third option, which Otanes favors in the  . . . αρχειν   vertically analogous context of a society’s internal organization, is ουτε ουτε  αρχεσθαι (3.83.2). 256. 8.3.1; see chap. 3, “The Evils of War.”

Comparison

131

and intertribal struggle. After the initial battle, the Scythian adults and the Scythian young men do not fight with the Amazons, nor do the Amazons fight with them. Hoping to coopt the Amazons as child bearers, the Scythian adults find themselves deprived of the sons they already have, but they let them go without resistance. The young men even obtain from the Scythians their share of family property before leaving for good (4.115.1). More directly emphasized, for obvious corrective reasons, is the behavior of the Amazons themselves. The Amazons realize that their initial sea crossing into Scythia has entailed a number of more or less voluntary and aggressive violations that now put them at risk. We are afraid and frightened to have to live in this place, because, firstly, we have deprived you of your fathers and, secondly, we have greatly devastated your country. So they plan to undo the invasion and leave. Since you think fit to keep us as your wives, do the following  together with us. Come, let us move away [ε ξαναστεωµεν] from  this land and, crossing [δια βαντες] the river Tanais, live over there. (4.115.2– 3) In the Histories, the fearless men who cross rivers (with the verb διαβαι νειν) do so to subdue and conquer, appropriate what belongs to others, add to their rightful share, and, in concrete metonymic terms, acquire more land.257 Here the Amazons and their new husbands, who once again “were persuaded” (4.116.1), abandon a country that is not or is no longer their own and cross the river (δια βαντες), removing themselves—at a distance of a three-day journey away, to be exact—from conquest and war.258 According to the symbolic code of the Histories, the Amazons’ crossing of the Tanais is a violation of boundaries in reverse and a spectacular display of sophrosune. The subtext of the Amazon myth determines the 257. See my introduction, n. 26 and corresponding text. For words of the διαβαι νω family marking “unwise imperial ventures,” see Lateiner 1989, 131– 32; Payen 1997, 140. 258. The “three days” may belong to the contemporary political code. Diodorus (12.4) reports a clause of the Peace of Callias that stipulated that Persian armies should not come nearer to the coast than three days’ march. The crossing of the river in reverse is also noted by Flory (1987, 112– 13).

132

Telling Wonders

polemic intent of the story and enhances its participation in the general message of Herodotus’ ethnographic historie. According to the traditional image, the Amazons were invaders who lusted for conquest, while the autochthonous Athenians were and are by nature just and would never deprive another of his land.259 Representing the product of Herodotus’ historical and ethnographic research, the Sauromatian logos shows that the Amazons were neither wild, nor violent, nor cowardly, nor gutless, nor eager to enslave, nor ignorant of justice, nor the enemies of the race of men. Here, as elsewhere in the work, female is not the antithesis of male, barbarian is not the antithesis of Greek, and the alternative of conquering or being conquered appears invalid. Herodotus’ pursuit of the similar within his representation of difference confounds mythical constructs of alterity. His scientific ethnography teaches that difference pervades the world, to be sure, but not according to the schematic intellectual map devised by the Greeks. Conclusion I have begun by exploring the extent to which the narrator’s explicit comparative interventions confirm the cohesiveness of the Histories and throw light on their meaning. In the historical narrative, I have argued, the rule of diachronic similarity predominates on its own by implicit analogy. The juxtaposition of narrated data in itself conveys that similar events and features recur (horizontal analogy) and that certain actions or objects figuratively represent more abstract qualities and general processes (vertical analogy). Thus, the narrator rarely points out in his own voice, “Such and such is analogous to such and such.” Two cases of this sort provide to the recipient of the narrative additional help on shocking or particularly obscure connections. In doing so, they confirm the correctness of the listener’s interpretation of the history through both horizontal and vertical analogy. To the extent that these glosses make patent the narrator’s own process of interpretation, they also reveal the encompassing range of the analogical field that revolves around kingship as both historical reality and historical symbol. If historical events emerge as being like each other diachronically, 259. See the Athenian ambassadors at 7.161.3. The Amazons’ lust for conquest is already implicit in Aesch. Eum. (685– 90). See also Isoc. Paneg. 4.68– 70; Lysias 2.4. In Demosthenes (6.4– 8) the connection between autochthony and justice is immediately followed by the mention of the Amazons’ invasion. See Tyrrell 1984, 114– 16.

Comparison

133

ethnography and geography describe synchronic difference around the world. Without difference, there is nothing to narrate. Glosses explicitly declaring the difference of a phenomenon from another or its uniqueness periodically underline this inherent presupposition of the narrative. Glosses of difference are not always advertisements of narratability, however. Rather, in certain cases, they aim at breaking down ethnographic categories into smaller, if often more elusive, groupings. This maneuver seems designed to contradict and scramble excessively schematic notions, much in the same way as when Herodotus devalues the conventional subdivision of the earth into large sections. Conversely, explicit glosses of difference that engage as whole categories the Greeks, on the one hand, and the barbarians, on the other, are almost entirely absent. This constitutes again a move away from convention. Herodotus is reluctant to theorize the Greeks as special or even as the norm in an absolute sense. Most important is the frequency with which the whole representation of difference is counterbalanced by glosses that explicitly state that an ethnographic or geographic phenomenon is like another one somewhere else, either foreign or Greek. These statements scan the objective account of specific facts and add plausibility to its intended direction. When Herodotus describes how various ethnographic subjects differ from the Greeks and emphasizes their separate identities—the different ways in which they differ from the Greeks—this also conveys the different ways in which they resemble the Greeks or different groups of Greeks. The glosses of similarity compensate for the propensity of ethnography to result in a discourse on alterity, especially the alterity of the barbaroi as a whole to the Greeks as a whole. They are reminders of an ideology of sameness that manifests itself also in the unmarked representation of shared features in the context of the objective description of difference. The identification with the other, like the partial devaluation of geographical boundaries, participates in Herodotus’ overarching idea of a world that is differentiated and homogeneous at the same time. This in turn is designed to spoil for the Greeks the pleasure of contrast effects, to uncover for them surprising paradigms, and also to deny them the separate role of spectators of barbarian woes, such as Aeschylus’ Persians had allowed them to take. Each ethnos possesses its special identity and history, but pervasive and unexpected likenesses among all ethnea guarantee on principle eventual resemblances in their historical experience, as happened with the Scythians and Greeks. The synchronic patterning of the world confirms and explains the predictability of the patterns of history.

Chapter 3

Interpretation and Evaluation

A narrative text encourages its recipient to understand the events it recounts in a certain way through strategies that are largely indirect, such as the selection of the facts it narrates and the words by which it chooses to represent them. But when we hear the narrator express the main point about the facts of a narrative, speculate on its less obvious aspects, express a judgment, or draw the moral of the story, we have an interpretive gloss. This is a very broad metanarrative category, potentially coinciding with the range of referential metanarrative itself. Among explicit comparisons, a bold case of interpretation appears in the gloss that points out that Cleisthenes of Athens reproduced in his democratic reforms certain actions of his grandfather, the tyrant of Sicyon. The gloss “just like the Greeks” attached to the factual statement of Egyptian monogamy bids the listeners to think of the Egyptians in terms of affinities with themselves, rather than differences. Both these elements of the discourse provide, as do interpretive glosses in general, directions for “reading” (in the sense of decoding) the narrative. The narrator also enhances the listener’s understanding of what is being narrated by expressing or implying his approval or disapproval. Evaluation of worth is always based on certain interpretive assumptions about the meaning of the action that is praised or blamed; conversely, to explain meaning even in neutral terms often promotes a judgment in terms of “Good” or “Bad.” The distinction between evaluation and interpretation, in other words, is bound to lack rigor, but its usefulness for the purposes of the present book has to do, once again, with the differences we notice in the approaches of history and ethnography, respectively. History first and foremost investigates the meaning of events—why they occur and what general laws one can derive from the accumulation of factual data. We therefore provisionally subsume evaluation to interpretation. Herodotus’ ethnographical descriptions, in contrast, often leave unanswered (and unasked) why people customarily do what they do (e.g., 134

Interpretation and Evaluation

135

sacrifice) in a certain way. Even in cases that may appear controversial to us, Herodotus tends to report the meaning of actions as a matter of fact.1 Although cumulatively ethnography leads Herodotus to an understanding of the world, the main problem in individual ethnographic statements and sections is rather the extent to which his discourse either valorizes difference or encourages the assumption of its inferiority. As I examine the ethnographies, therefore, I will begin by subsuming interpretation to evaluation and focus on the ways and means by which Herodotus promotes approval or disapproval of his ethnographic subjects. Evaluation in the Ethnographies Explicit Evaluation I have argued in the preceding chapter that Herodotus’ insistence on the strands of cultural similarity connecting different ethnea around the world contributes to undermining the notion of a clear-cut separation between Greeks and non-Greeks and questions the validity of those categories. But the canonical antithesis pervades the logos and implicates the histor. The Spartan ambassadors in Athens declare that “with barbarians there is no faith or truth” (8.142.5). Pausanias tells the Aeginetan Lampon that to mutilate the corpse of Mardonius would be an action “befitting barbarians rather than Greeks” (9.79.1). Both these judgments occur in ironical contexts and have no authority from the point of view of the text.2 When Xerxes lashes and brands the Hellespont, however, the evaluation of his words as “barbarian and impious” [βαρβαρα τε κα ι

 σθαλα] (7.35.2) comes directly from the narrator and testifies to his ατα acceptance of the negative ethical connotations of the word barbaros. The exclusive application of the word in this sense to the most symbolic monarchical transgression in the entire logos helps to explain the partial dissociation between the barbaroi of the ethnographies and the concept of barbaron in the history. Because of the regocentric character of the historical narrative of non-Greek actions from Croesus to Xerxes, 1. E.g., from our point of view, Herodotus “interprets” the ritual described at 4.62 as a sacrifice to Ares. Mora (1985, 124– 25), e.g., disagrees on both counts. But from the point of the text, this is a sacrifice to Ares. 2. On irony in the sequence of speeches at 8.140– 9.11, see Fornara 1971a, 84– 86; Raaflaub 1987, 240. Spartan deceitfulness is a contemporary stereotype, at least in Athens. See Hdt. 9.54.1, 9.11, 6.108; Powell 1988a.

136

Telling Wonders

entire cultures tend to become identified with their less typical specimens and bear the burden of providing the negative paradigm as either ruler or ruled. Monarchy carries such powerful negative connotations that it cannot help affecting the evaluation of the peoples who accept and perpetuate this form of government.3 This evaluative constraint also has a certain carryover in points of representation of a nonpolitical nature. Thus, in the ethnography, the Lydians are authors of inventions later adopted by the Greeks (1.94.1– 2), but in the history, the Greek Thales scientifically predicts the eclipse that appears as a terrifying portent to the Lydians and the Medes (1.74.2). In the ethnography, the Egyptians do things earlier and more intelligently than the Greeks, yet in the history, a Greek doctor proves superior to the Egyptian physicians (3.129– 30). The main feature of the discrepancy is, however, that the historical narrative of the actions of kings is likely to be the locus of orientalistic representation.4 The ethnographies—which describe a wider range of cultures and take a broader outlook on culture—go a long way to repair the damage. Indeed, one might argue that from the point of view of the overarching message of the Histories, this is one of their primary functions. Nowhere in the logos, at any rate, do we find the narrator evaluating a known historical people with a statement remotely resembling the following statement that Thucydides appends to one of his narratives, despite Thucydides’ more limited opportunities for characterizing foreigners. For the Thracian race, like the bloodiest of the barbarians, is even more so when it has nothing to fear. (Thuc. 7.29.4, trans. Crawley in Strassler 1996). Herodotus applies negative evaluations to a few remote peoples, who play a very small role in the historical narrative and whose customs he cannot regard as morally equivalent to those of more advanced ethnea.5 An extreme and unique case is represented by the utterly unregulated Androphagoi, who illustrate Herodotus’ identification of law/custom/ culture with morality: they “have the most savage ways of all men 3. See the implicitly evaluative interpretive gloss to the report that the Egyptians, after gaining their freedom, set up twelve kings, “for at no time they were able to live without a king” (2.147.2). In the historical narrative, cf. 3.143.2, on the Ionians of Samos, and 1.62.1, on some Athenians in the time of Pisistratus. 4. On orientalism, see Sa¨ıd 1978, 5– 9 and passim. 5. E.g., he describes peoples who couple “like cattle” (1.203.2, 3.101.1, 4.180.5).

Interpretation and Evaluation

137

  [αγρι ωτατα ηθεα], and they do not practice justice as a custom nor do   ουδεν  δι κην νοµι ζοντες, ουτε  νοµω

they have any law/custom [ουτε ι 6  χρε ωµενοι].” Two evaluative statements that explicitly praise a people’s morality are designed to overcome the “Bad” impression that the report of their customs may have created. The funeral practices of the Issedones involve cannibalism and the fashioning of the dead man’s skull into a cup. Here, Herodotus’ effort to promote in his audience a relativistic attitude has already expressed itself in the comparative gloss that made these practices manifestations of piety on the same level as a Greek ritual. In the subsequent gloss that “otherwise these also are said to be just and to enjoy equal rights, both men and women” (4.26.2), also (και ) assimilates the Issedones to the Bald Men, who are at the opposite end of the spectrum to the Androphagoi and paradigmatically just.7 More impor tantly, otherwise (αλλως) insists on the principle that justice is a virtue that transcends cultural differences; but it also reveals the difficulty of the relativistic position. Similarly, the totalizing statement that the Getae are  

the “most courageous and just” [ανδρηι οτατοι κα ι δικαιοτατοι] of the Thracians (4.93) counterbalances, on the one hand, the information that they were conquered by the Persians and, on the other hand, the description of their cruel ritual in honor of their god Salmoxis, whose absurdity the text can hardly conceal.8 A different application of this compensatory strategy consists in punctuating the nonjudgmental account of practices that are bound to strike a listener as savage with very specific praise of certain items of material culture. In his description of the brutal Scythian sacrifice, for example, the narrator conveys the idea that despite differences and deficiencies, things nevertheless get done in a satisfactory way. Since the land of Scythia is  “terribly poor” in wood, the Scythians “have invented” [ε ξευρηται] a special procedure for cooking the meat by lighting the fire with the bones. They cook the meat in local cauldrons, “if they have them, very similar to Lesbian craters, except that they are much bigger.” If they do not have a cauldron, then they throw all the meat into the stomachs of the victims: “The meat cooks beautifully [αιθεται καλλιστα], and the stomachs 6. 4.106. As Redfield explains (1985, 98– 99), animals also have ethea, but only human beings have nomoi. 7. See 4.23.2– 5. Equality between the sexes also obtains among the Bald Men. See chap. 2, n. 187 and corresponding text. 8. 4.94; see Hartog 1988, 84– 109; Lateiner 1990, 235– 45.

138

Telling Wonders

 contain the meat easily [ευπετ εως] after it has been boned. And so the ox cooks itself, and the same goes for the other victims.”9 Sophie constitutes a conspicuous field of ethnographic evaluation.10 It is a diverse quality that may include resourcefulness, practical or theoretical intelligence, learning or wisdom; when it is attributed to societies, it lacks the ethical dark side it occasionally connotes in individuals. The dogma of Greek superior intelligence is the subtext that makes all the explicit or implicit attributions of sophie to foreign nations indices of similarity to the Greeks. At the only time when the narrator formulates the dogma directly, however, he encases it in a demonstration of Athenian foolishness (1.60.3). Conversely, the totalizing statement that evaluates the group of ethnea around the Pontos as “the most ignorant people”

 [εθνεα αµαθ εστατα], occurs in the context of Herodotus’ praise of the   Scythians for making the “cleverest discovery” [σο␸ ωτατα ε ξευρηται]— 11 a mode of life that enables them to elude subjection. Several ethnographic evaluations of worth are marked by self-referential signs I have grouped in the broad category of “opinion.”12 Some of them, however, express a more subjective and less reasoned inclination  than a γνωµη. They place the individuality of the narrator in the foreground by emphasizing his entitlement to a subjective reaction to something foreign, and they appear to constitute one of his responses to the tension between a relativistic ideology and an evaluative impulse. Thus, concerning the Scythians, the narrator points out the paramount strategic value of nomadism, but aside from this, he adds, “nothing else pleases  

αγαµαι]  me” [τα µεντοι αλλα ουκ (4.46.2). While acknowledging the argument of those societies that allow intercourse in temples, he still proclaims his personal dislike of the practice: “I just do not like it”  ].13

αρεστα

[εµοιγε ουκ Couched in objective terms, in contrast, is the negative evaluation of the Babylonians’ “most shameful custom” (1.199.1). This includes not 9. 4.61.1– 2. The narrator also praises Scythian and Thracian hemp (4.74), the Scythians’ lustral steam bath (4.75.1– 2), their aromatic body pack for women (4.75.3), and the availability to them of essential resources (4.59.1). See Romm 1998, 111. 10. See, e.g., Herodotus’ representation of the Assyrians, discussed in my introduction. 11. 4.46.1– 2. The importance Herodotus attributes throughout the logos to a nation’s ability to maintain its freedom motivates also evaluations concerning the population size and military strength of Thracians (5.3.1) and Ionians (1.143.2). See Evans 1976. 12. See chap. 1, “Self-Referential Glosses.” 13. 2.64.2. Cf. the subjective formulations of 2.3.2 and 2.46.2, discussed in chap. 2, “The Texture of Nomos.”

Interpretation and Evaluation

139

merely intercourse in a sacred place but the obligation for all local women to subject themselves to ritual prostitution in the temple of Aphrodite once in their lifetime. The narrative provides a neutral but eloquent account of the humiliation of the wealthiest among the women, each compelled to follow the first stranger who comes along and throws whatever amount of money he wishes at her feet.14 The narrator underlines the uniqueness of the occasion with a direct challenge to the listener: after the woman has returned home, “you are not going to get her no matter how much money you give.” We also receive a vivid impression of the ordeal of the less attractive women, each of whom must sit in the temple even for three or four years until some man chooses her and allows her to fulfill the ritual (1.199.1– 5). This exceptional case of explicit negative moral evaluation in an ethnographic context begins to suggest a distinction between customs that are oppressive and customs that are welcome and helpful to the people who follow them. The exact counterpart of “the most shameful custom of the Babylonians” is one that Herodotus introduces as “their most intelligent    η µετερην]  custom, in my opinion” [σο␸ ωτατος . . . κατα γνωµην τ ην (1.196.1) and which he later calls “their most beautiful custom” [ο  καλλιστος νοµος] (1.196.5), thereby signaling moral approval as well. This practice also involves a mercantile exchange of women, with the attendant problem of an inequality between the more and less beautiful. This time, however, the institution consists in a yearly auction of brides for the achievement of stable marriages, carefully designed to compensate for preexisting disadvantages and to make everyone happy. In Aristophanes’ Ecclesiazosae, a roughly comparable utopian arrangement is called a  δηµοτικη γνωµη (631). This “democratic” quality lies at the basis of the excellence of Herodotus’ Babylonian nomos, unfortunately superseded since the advent of hard times by the definitely “undemocratic” and exploitative practice of prostitution (1.196.5). A similar criterion of evaluation also emerges from the Babylonian custom judged “second in sophie” (1.197), where the marketplace principle is applied to the field of health care. Since the Babylonians have no professional class of doctors, they bring all the sick people to the agora, where passersby proffer advice and prescriptions, each according to his personal or indirect experience of this or that ailment. In the case of the 14. A form of ritual prostitution also existed in Greece, but not for citizen women. See Halperin 1990, 104– 7, for sources.

140

Telling Wonders

market of brides, various prohibitions were in place to prevent one from eluding the system (196.3– 4); so also in this case, no one is allowed to walk by a sick person without asking him why he is suffering. Both nomoi provide for the exchange of resources according to need, and both preclude individualistic violations. The Babylonian custom of gathering the sick in the center of the city for the purpose of caring for them implicitly contrasts with the Persian practice of banishing from the city altogether whoever suffers from leprosy or the “white disease.”15 To the explicit praise of the Babylonian custom, however, there corresponds no explicit negative evaluation of the Persian. What we find instead is a justification of the (implicitly “Bad”) Persian practice in terms of native knowledge: they banish the sick because “they say” that these particular diseases afflict those who have sinned against the Sun. The fact that explicit praise is more frequent than explicit blame points to a search for whatever “Good” features a certain culture has to offer from out of its own store of nomoi, institutions, or resources.16 In the description of the highly hierarchical Persian society, where there is no trace of the demotikon that Herodotus is able to find among the Babylonians, Herodotus chooses two nomoi of an entirely different sort on which he dispenses personal praise through the per “I approve” (1.137.1). Both these customs aim at formative verb α ι ν εω, protecting the patriarchal head of the family and state from succumbing  to excessive emotions—respectively, grief (αση) at the possible death of  against servants and an infant son (1.136.1) and impulsive anger (θυµος) subjects (1.137.1). As in the case of explicit comparisons, evaluations point to the ideological direction in which the text travels by implicit means. In the marketplace or other community space, people can observe what they like or need among different items, as in the wisest and second wisest Babylonian practices: this setting also serves as the metaphor for the activity of the ethnographer, who places “in the middle” a variety of different peoples and their nomoi for the benefit of his audience. Look15. 1.138. The antithetical character of the two customs is observed by Asheri (1988, 380). 16. See the praise of the Persian postal system (8.98.1). See also statements praising foreign natural resources (1.193.3, 4.198, 4.194, 3.106.1, 3.112, 3.113.1), the “correctness” of certain barbarian names (4.59.2, 6.98.3), and a people’s physical health (4.187.3, 2.77.3). No negative evaluations concerning health occur, as they do in the Hippocratic Airs, Waters, Places.

Interpretation and Evaluation

141

ing on entails evaluation and possibly a negative judgment, since people like their own customs best (3.38.1). Herodotus recognizes this inevitable tendency through his own evaluations but also keeps it under control.17 He models for his listeners an attitude of charitable observation; when he does not lead them to the realization of unexpected likeness, he promotes the discovery of understandable difference and creative solutions.18 This attitude dissociates the notion of barbarian from that of the barbaric and replaces a generalized contempt for alien customs with a more self-conscious definition of what must necessarily be the furthest limits of one’s tolerance. Strategies of Evaluation Revising Greek Traditions Herodotus’ descriptions of foreign cultures frequently imply a context of Greek ignorance and prejudice and thereby signal the ethnographer’s corrective aims. In the field of interpretation and evaluation, one of Herodotus’ strategies consists in refuting or countering Greek stories that perpetuate damaging stereotypes of the barbaroi. In book 2, passages that criticize Greek traditions concerning Egypt or the Egyptians are a part of the polemic against different branches of Greek knowledge (chronology, geography, the gods), which Herodotus pursues while more broadly demonstrating the limited power of historie for all sides involved.19 Programmatic in this respect is the initial set of two logoi (2.2.1– 5). To the experiment of Psammetichus as a positive, though imperfect, barbarian instance of historie in the narrative, the metanarrative juxtaposes a negative Greek one: according to the Greek version of the story, for the purposes of his experiment, Psammetichus cut off the tongues of his servant women. This is one of the “many frivolous things [µαταια πολλα ] that the Greeks say.”20 Similarly, at the end of his radical 17. Contrast the evaluative mode of the Hippocratic author of Airs, Waters, Places or of Strabo/Ephorus (see chap. 2, “Does Climate Determine Culture” and “The Sameness of the Scythians”). 18. For the concept of “charity,” see Asad 1986, especially 147 in reference to modern ethnographers’ tendency to give a favorable account of native phenomena. 19. See Hunter 1982, 50– 92, especially 61, 64– 65, 67, 69– 70. 20. On Psammetichus in this passage as a model histor and analogue of the histor Herodotus, see Christ 1994, 184– 86, especially 185. On the limitations of both Psammetichus’ interpretation of his experiment and the Greek version of the story, see, e.g., Benardete 1969, 33; Dewald 1998, 615.

142

Telling Wonders

revision of Greek traditions concerning Heracles in light of information acquired in Egypt (2.45.1– 3), the narrator dubs the Greek story that the Egyptians attempted to sacrifice Heracles another typical foolishness  

ι ρους) of Egyp(ευ ηθης . . . µυθος). Here the Greeks’ inexperience (απε  tian culture and nature (␸ υσις in the sense of “national character”) contrasts with Herodotus’ experience of these things (see δια πειραν at 2.77.1); it leads them to believe in an occurrence that even contradicts  nature (␸ υσις in the sense of “what is humanly possible”). As at other times in Herodotus, custom and nature, or disregard and ignorance thereof, are here closely related.21 While the Greek version of Psammetichus’ experiment reveals a generalized Greek stereotype of barbarian cruelty, the tale of Heracles’ sacrifice is based on a contemporary prejudice that is specifically tailored to Egypt. Confronted with Egyptian aloofness, the Greeks constructed a fantasy of Egyptian hostility toward guests, as is represented outside of Herodotus by the legend of how the Egyptian king Busiris habitually sacrificed travelers arriving to his land.22 The Busiris model of xenophobia also lies behind the Egyptian version of a Greek story that Herodotus reports to provide an implicit reversal of the Busiris myth. In this passage (2.112– 20), both the historical knowledge and the national character of the Egyptians are vindicated through no less a topic than the archetypal Greek saga of the fall of Troy. Herodotus has learned from the priests of the sanctuary of Proteus in Memphis that Alexander and Helen landed in Egypt on their way from Sparta to Troy. The slaves of Alexander sought refuge in Egypt, as suppliants in a nearby temple of Heracles, and told the priests about their master’s abduction of Helen. The warden of the mouth of the Nile, a fellow named Thonis, informed King Proteus. After confronting Alexander, Proteus banished the wrongdoer from his land but kept Helen, her treasure, and the suppliant slaves 21. Cf. the rhetorical questions “How does it accord with nature that Heracles, being only one and a man, should kill many tens of thousands?” (2.45.3) and “How could a dove speak with a human voice?” (2.57.2) in Herodotus’ refutation of the Greek story of the foundation of the sanctuary of Dodona, another tradition born from the Greeks’ inexperience of ethnic difference. Herodotus also indicts Greek “incorrect speaking” in his refutation of the story that the courtesan Rhodopis built the pyramid of Mycerinus (2.134.1– 2.135.6). 22. On Busiris, see especially Froidefond 1971, 177– 79; Lloyd 1976, 212. The earliest testimonies of the story date to the sixth century: see [Hesiod] frag. 378 MW; Pherecydes, FGrHist 3 F 17; Panyassis frag. 26K (⫽ Athen. 4.172D); and the late sixth-century b.c. hydria from Caere in Vienna (3576).

Interpretation and Evaluation

143

under his protection until the time when Menelaus would come and claim them back (2.113– 15). Alexander sailed back to Troy, where the Greek army was already gathered. The Greeks requested the return of Helen, but their embassies bore no results: the Trojans kept saying that Helen was in Egypt, but the Greeks did not believe them. The war went on until, finally, the city was captured; only at that point did the Greeks realize that Helen was not at Troy and send Menelaus to look for her in Egypt.23 Herodotus again calls what “the Greeks say” (which here means the  canonical Homeric version) a µα ταιος λογος (2.118.1), and he corroborates this Egyptian story in three different ways. First, he reports his independent conjecture, based on autopsy, that the shrine in Memphis uniquely named after “Foreign Aphrodite” or “Aphrodite the Guest” must be a shrine of Helen (2.112.2). Second, he examines four epic passages for the purpose of demonstrating that Homer knew about the sojourn of Helen in Egypt, though he followed the incorrect version because it was “appropri ε ποποιι ην ευπρεπ

 (2.116.1). Finally, he ate to epic poetry” [ε ς τ ην ης] refutes the Greek version, showing its logical flaws on the basis of the likely (2.120.1– 4). The Egyptians prove again to be superior sources, even on a fundamental issue of ancient Greek history.24 The truthful Egyptian version has the advantage that it more clearly reveals a fundamental pattern of history, as I shall show when I discuss the most generalized level of Herodotus’ interpretation.25 But from an ethnographic standpoint, which concerns me here, it counters “foolish” Greek prejudices with a correct representation of Egyptian phusis and nomoi. The theme of xenie (hospitality) links the myth of the Egyptians attempting to sacrifice Heracles (with its Busiris subtext) and the story of Helen in Egypt, as Plutarch noticed.26 In the story of Helen, the stem ξενoccurs a total of twelve times, and the stem δικ- (justice) occurs eight times. A great deal of character text discusses what is or is not a just and  holy (οσιος) treatment of guests. Proteus reproaches Alexander in terms 23. 2.118.2– 4. On this story, see Fehling 1989, 59– 65; Hunter 1982, 52– 61. Austin 1994, 118– 36. 24. At 2.118.1 and 2.119.3, the text refers to Herodotus’ Egyptian sources with the same vocabulary of inquiry and knowledge that Herodotus elsewhere applies to himself:

 “to know through inquiry” [ ι στορι ησι . . . ε πι στασθαι], “knowing precisely” [ατρεκ εως

ε ιχον ει πειν]. For the narrator’s evaluation ε πιστα µενοι], “they were not able to tell” [ουκ  of the Egyptians as λογι ωτατοι (most learned), see 2.77.1; cf. 2.4.1. For the real Herodotus as a logios, see Nagy 1990, 221– 24; Evans 1991, 94– 98. 25. See 2.120.5 and “Divine Retribution” later in the present chapter. 26. Plut. De Malign Herodot. 12 ⫽ Mor. 857A– B. See also Froidefond 1971, 178– 79.

144

Telling Wonders

that characterize this Egyptian king as a reversal of the guest-killer Busiris: If I did not consider it of the greatest importance not to kill any of the guests who are driven by the winds to my land, I would seek from you retribution on behalf of that Greek, O most wicked of men, you who after receiving hospitality have perpetrated a most impious deed: you have come to the wife of your own guest . . . and this was not enough for you, for you come here after plundering the house of your guest. But now, since I consider it of the utmost importance not to kill guests, I will at least prevent you from taking away this woman and the treasure, and I will hold them in my keeping for the Greek stranger/guest [ξει νω  ] until he wants to come himself and take them back (2.115.4– 6). The presence of a temple with a special nomos protecting suppliants enhances the representation of Egyptian piety. The role of the warden Thonis ensures that Proteus’ righteousness will be perceived as also typical of ordinary Egyptians. The impeccable Egyptian hospitality contrasts with Alexander’s “injustice” and “impious deed” (2.113.3; 2.114.2; 2.115.3, 4; 2.120.5). But the story gains a special polemical edge (again not lost on Plutarch) when the  Greek guest in this story, Menelaus, “behaves unjustly” [ε γενετο ...  

αδικος] and “devises an impious action” [ε πιτεχναται πρηγµα ουκ  οσιον] (119.2). After receiving Helen and her treasure back with hospitable gifts from Proteus (ξενι ων . . . µεγαλων, 2.119.1), Menelaus sacrifices two local children to obtain favorable winds for his sailing. Adapted to the xenie theme of the narrative, this reenactment of the sacrifice of the Greeks at Aulis produces a host-sacrifice that reverses the alleged guestsacrifices of the Egyptian Busiris.27 Like the anti-imperialistic Amazons in the Sauromatian logos, the hospitable Egyptians of this story set the record straight with regard to the representation and evaluation of the other. Native Voices Through native logoi, the text revises the assumptions of the Greeks about barbarians. A strategy for “challenging evaluative preferences for Greek characteristics over barbarian,” in comparison, consists in focaliz27. Fehling (1989, 62) notices the analogy between Menelaus’ sacrifice of the Egyptian children and Agamemnon’s sacrifice of Iphigenia at Aulis.

Interpretation and Evaluation

145

ing Greek and foreign customs through the barbarians.28 Appearing in the logos in the guises of characters, ethnographic subjects, or epichoric sources, “they” participate in the din of voices that resound on many different issues throughout the work. When these speakers talk at different narrative levels about the way in which they or others lead their lives, their utterances serve the purpose of ethnographic evaluation. In Culture and Truth: The Remaking of Social Analysis, anthropologist Renato Rosaldo explains, “If you ask an older Ilongot man of northern Luzon, Philippines, why he cuts off human heads, his answer is brief . . . : He says that rage, born of grief, impels him to kill his fellow human beings.” During his stay among the Ilongots in the late 1960s, Rosaldo was troubled to discover that head-hunting was still surreptitiously practiced and that “every man in the settlement had taken a head.” A few months later, when the ethnographer was drafted for Vietnam, his Ilongot friends surprisingly urged him not to go and offered to conceal him in their homes: “They told me that soldiers are men who sell their bodies. Pointedly they interrogated me, ‘How can a man do as soldiers do and command his brothers to move into the line of fire?’”29 Set in the context of Rosaldo’s self-conscious analysis of available forms of ethnographic discourse (not a type of analysis Herodotus does), these exchanges provide a not too distant parallel for the strategy of representation by which Herodotus bridges the gap or evens the score between foreign and Greek customs by opening the possibility for what Rosaldo calls “reciprocal critical perceptions,”30 in which a native explains himself and/or criticizes aspects of the ethnographer’s culture. Ethnocentric remarks from the other at the very least communicate the idea that what one takes for granted as normal may also be perceived, with some justification, as undesirable and abnormal.31 When Herodotus’ foreigners evaluate Greek customs, they are almost always critical.32 Though they often display ignorance, exaggeration, or a 28. See Pelling 1997. The two functions often overlap. 29. Rosaldo 1989, 1 and 63. See Rosaldo’s entire chapter “After Objectivism” (46– 67). 30. Rosaldo 1989, 64. 31. In Herodotus, see, e.g., the Ethiopian king’s contempt for bread (3.22.3– 4) and the Egyptians’ ethnocentric criticism of the Greeks’ dependence on rainfall, the last both corroborated and countered by the narrator (2.13.2– 2.14.1). 32. Two exceptions (7.208.1– 3 and 8.26.3) occur in the highly celebratory narrative of Thermopylae, where we also find the similarly exceptional case of an entirely misguided criticism of Greek culture by a foreigner (7.103.3).

146

Telling Wonders

quaint partiality of their own (as Rosaldo’s Ilongots do when they define soldiers as “men who sell their bodies”), they also make valid points.33 Among observers of Greek nomoi who combine misunderstanding with brilliance is Mardonius. His main point to Xerxes that the Greek method of fighting is ineffectual and uneconomical constitutes “foolish words”  [µαταιοι λογοι], “slanderous” against the Greeks, as Artabanus will re34 tort. But when he implies that it is the peculiar nomos of the Greeks to fight one another incessantly, this marks one of those great polemic moments that leave no doubt as to what the Histories are about, whom they address, and what solution they prescribe: “And yet, since they speak the same language, Mardonius says, they should settle disputes with heralds and messengers and in any other way except by fighting” (7.9β1– 2). The reference to a common language, which Mardonius intends as a practical consideration, defines for the audience a conventional aspect of their Panhellenic cultural identity (8.144.2). The phrase “settle disputes with heralds and messengers and in any other way except by fighting” belongs to the political code of Herodotus’ time. Thus, in Thucydides, Pericles maintains that the Spartans “prefer to settle disagreements by war rather than by words”; Thucydides cites the suspension of parleys through heralds on both sides as the first decisive sign that their disputes have finally determined the outbreak of the Peloponnesian War.35 Cognitive Statements and Polemical Negations Since negative statements serve to contradict the audience’s expectations, they often enhance the representation of difference by underlining a culture’s lack of features typical of “normal” civilized living.36 At other times, however, negations reflect a people’s ideology and its polemics against foreign customs. The ethnographic gloss explaining that the Per33. See, e.g., 4.79.3 (Scythians’ polemic against Dionysus; see chap. 2, “Identification with the Other”; 4.142 (Scythian criticism of the Ionians as “slaves”), 2.160.2– 4 (Egyptian criticism of the rules of Olympic Games). 34. 7.10η1. For µαταιος applied to ethnic prejudice, see discussion under “Revising Greek Traditions” earlier in this chapter. On the economical nature of Greek warfare, see Hanson 1989, 1– 18. 35. Thuc. 1.140.2 (Pericles here echoes the exhortation of the ephor Sthenelaidas at 1.86.3), 1.146. 36. See Hartog 1988, 257. Among statements of true “lack,” the limiting case occurs at 4.106. On the portrayal of deficiencies and primitivity through “empty negatives” in modern ethnography, see the critique of E. Le Roy Ladurie’s Montaillou (1975) in Rosaldo 1986, 85.

Interpretation and Evaluation

147

sians “do not use marketplaces, and do not even have a marketplace to begin with” for example, is attached to Cyrus’ contemptuous definition of the Greek agora as “a designated place in the middle of the city where people gather and deceive one another under oath” (1.153.2). This cognitive use of negative statements underlines that foreign peoples are different in ways that are desirable to them. They do not have, do, or believe something that is normal for the Greeks, not because they are more limited in some sense, but because they do not consider that item useful, holy, just, or true.37 Negative statements, whether in the voice of characters, ethnographic subjects, sources, or the narrator, may also serve to preempt biased judgments in the listener (“It is not true, as you think, that . . .”). In book 4, Herodotus relates a remarkable system of commercial exchange that takes place between the Carthaginians (who are also the source for this information) and an unidentified Libyan people living beyond the Pillars of Heracles (4.196). After arriving to the shores of these men, the Carthaginians display their wares in orderly array along the beach, then return to their ships and make smoke signals. When the natives see the smoke, they go down to the beach, deposit the amount of gold they offer for the merchandise, and withdraw. The Carthaginians in turn disembark and look at the gold. If they judge it sufficient to pay for the merchandise, they take it and sail away; if not, they return to their ship and wait. The Libyan natives return and keep adding gold until the merchants are satis

 δ ε ουδετ

 fied, and “there is no foul play on either side [αδικε ειν ερους]: neither do the Carthaginians touch the gold before they judge that it has reached the price of the merchandise, nor do the others take the merchandise before the Carthaginians have taken the gold” (4.196.3). Despite peculiar circumstances and the stereotype about dishonest Phoenician merchants,38 the system works well. The reassuring negative statement here serves as a positive evaluation that defies common assumptions. There happens to be a thematic link among what Mardonius says 37. The statements that a certain people will not borrow foreign nomoi belong to this category (2.79.1, 4.76.1). For negative prescriptions, see, e.g., 2.37.3, 4– 5; 2.45.2; 2.46.1; 4.63. For negative belief implying criticism, see, e.g., 2.50.3. On the Greek side, cf. Sperthias and Boulis’ statement that it is not their custom to prostrate themselves in front of a human being (7.136.1). 38. Phoenician dishonesty, notorious in Homer, is, e.g., built into the tradition corroborated by Herodotus at 2.54, 56.

148

Telling Wonders

about the Greeks who speak the same language but prefer fighting to negotiations, Cyrus’ perception about the Greeks’ deceiving one another when they trade in the agora, and the Carthaginian account of their seashore transactions that overcome a language barrier and entail no wrongdoing. This thematic link is the issue of fair dealing within and between different societies. It demonstrates the web of meaning that runs through the logos and the consistency of Herodotus’ ideology of mutual exchange and cooperation. The “most beautiful” Babylonian custom of the market of brides and their custom “second in wisdom,” the exchange of medical advice in the marketplace (1.196– 97), present the same structure. Their prostitution ritual is “most shameful,” but the listener should not misunderstand its import and meaning: after a woman has fulfilled the requirement, “you are not going to get her no matter how much money you give” (1.199.4). In the episode of the marketplace on the beach, Herodotus’ epichoric sources report what they do. The boundary between informants and ethnographic subject is here ill defined, but the phrase “the Carthaginians say” counts as a gloss of source identifying the narrative as someone else’s logos. In other cases, ethnographic subjects state their own cultural beliefs. In most cases, these cognitive statements, like native criticisms of Greek customs, contribute to portraying a culture’s strong sense of itself. By presenting an alternative but legitimate way of thinking, they tend to convey positive evaluation.39 The most striking illustration of Herodotus’ use of cognitive statements is represented by the Persian ethnography. Here we also find several other phenomena I have mentioned: negative statements of rejection, native criticism of Greek customs, explicit (positive) evaluations, mention of customs that elicit a “Good” impression as well as implicit similarity between the barbarian culture and the Greek. The passage is coherent and illuminating.40 Since the Persians are the most prominent foreigners of the Histories, it deserves to be examined in detail. 39. See, e.g., 2.35.3, 4.65.2. See Redfield 1985, 99. Besides in cognitive statements with   λεγουσι/λ εγοντες, ␸ασι or νοµι ζουσι /νοµι ζοντες, and so on, customs are focalized by their owners in sentences of the type “For them it is most beautiful [καλλιστον] to gather at drinking parties in groups according to age and friendship, men, women, and children together” (1.172.1). See n. 44 in the present chapter. Unmarked implicitly “native” explanations also occur; see, e.g., 4.104: “They have their women in common so that they may all be brothers . . . ” 40. Pace Gould 1989, 98– 99. For a historical discussion of Herodotus’ ethnographic representation of the Persians, see Briant 1988.

Interpretation and Evaluation

149

Persian Ideology The Persian ethnography is a list of nomoi, pure and simple, just as the prospective introduction announces.41 But such predicates as “they do X,” “they have learned to do Y,” and “they have X” alternate with “they assert/believe Y.” The narrator’s first-person glosses guarantee accuracy and testify to the ethnographer’s involvement with his subject. As for the Persians, I know that they have the following customs, that   ποιευµενους]

ε ν νοµω  they do not consider it in their custom [ουκ to build statues, temples, and altars but even attribute foolishness  [µωρι ην ε πι␸ ερουσι] to those who do, because, it seems to me, they 

. . . ε νοµισαν] never believed [ουκ the gods to be of the same nature as men, as the Greeks do. (1.131.1) Here the Persians attack Greek beliefs on a weak point.42 The negative statements that follow either connote Persian rejection of nonsensical accessories in sacrificial ritual (1.132.1: no altars, fire, libation, flute, fillets, barley grain) or enhance the polemic by pointing out a feature that the Persians, unlike the Greeks, would consider unthinkable to do without. After the sacrificer has arranged the meat, a Magus stands by and chants a theogony; for they do not have the custom to make sacrifices without a Magus. (1.132.3) Interdictive negations abound, displaying intransigent correctness. The one who sacrifices is not allowed to pray for good things for himself privately but rather wishes for the prosperity of all Persians and of the king, for he himself is included among the Persians. (1.132.2) They do not make water into rivers, spit in them, or wash their hands in them, nor will they allow another to do so, for they revere rivers most of all things. (1.138.2) 41. There is no subordinate narrative of how institutions came into being, such as we find in the Lydian ethnography (1.92– 94), no mention of particular rulers or historical events, and no description of monuments. 42. See Burkert 1988, 20– 21. Cf. Xenophanes, DK 21 A14.

150

Telling Wonders

The Persians not only are different from the Greeks but also find the Greeks inadequate, with a negative of true lack. They say that the Greeks stop eating when they are still hungry, because after the main meal, nothing else to speak of is served to them. (1.133.2) Persian difference is “Good,” based on a sensible and rigorous value system. To their sons, from the age of five to twenty, they teach only three things: to ride, to shoot arrows, and to tell the truth. Before he is five, a boy does not come in the presence of his father but lives with the women. (1.136.2)  Here the narrator intervenes to approve (α ι ν εω) in the explicit evaluations we have seen. The clipped rhythm of the list of Persian customs contributes to creating a brisk impression of strength. We may compare the Lydian ethnography, which differs from this passage in both the type of information it contains and the form of discourse it employs.43 In relation to the marketoriented Lydian culture, Persian culture is masculine, characterized by rough physical activities, the value of andragathie (manly excellence, 136.1), anti-intellectualism, and the rearing of sons (on Lydian daughters, see 1.93.3– 1.94.1). In their practice of prostitution, the invention of games, and the migration to Tyrrhenia, the Lydians appear to cope rather than act by choice. By contrast, the cognitive approach of the Persian ethnography makes Persian action stem not from historical contingencies but from collective ideology, likes and dislikes. The Persians do not simply prepare a particularly large dinner on their birthday; “they think it right” to do so (1.133.1). Sixteen predicates indicate what the Persians consider, believe, assert, honor, or deem right, while the Lydians acquire a voice only to speak of their long-lost resourceful past.44 Correspond43. See chap. 2, “The Sameness of the Lydians.” 44. See also “this is [for them] a display of manliness” (136.1) and explanations of customs where the focalizer is not marked (1.132.2, 1.136.2). Herodotus’ attempt to capture Persian national ideology is emphasized by Wolff (1934, 158– 62); see also Cobet 1971, 117– 18. For the Lydians, three verbs of saying occur (at 1.94.2) in the narrative of their invention of games and migration to Tyrrhenia, which is all in indirect speech.

Interpretation and Evaluation

151

ingly, we find here a total of twenty-four negations, either in interdictions or in polemical representative statements. The Persians even appear ready to revise reality if it does not conform to their value system. They say that no one yet has ever killed his father or mother, and each time something of the sort happens, they say that when the 

 νγκην] it matter is investigated, of absolute necessity [πασαν ανα would be found that the children were adopted or bastards. For they say, it is not natural that a true parent may be killed by his own child. (1.137.2) Aggressively dogmatic, the Persians consider themselves the best of all cultures. To their internal hierarchicalism corresponds a view of the inhabited world based on ethnic rank. Most of all, after themselves, they honor those who live closest to them, secondly those who live second closest, and then the others in proportion to their distance. Their literally geometric notion of ethnocentrism is reproduced by the ring structure of the ethnographer’s discourse, which places “the Persians  at the center and the decreasingly inferior “oththemselves” (ε ωυτους) ers” at the periphery on either side. They hold in least esteem those who live the farthest from them, believing that they themselves are by far the best of men in everything, that other peoples live more or less far in proportion to their

 and that those who inhabit the lands farthest excellence [αρετ η], from themselves are the basest.45 By association with this Persian mental map, a gloss of similarity recalls a feature of the old Median Empire.  

οντων], During the rule of the Medes [ε π ι . . . Μεδων αρχ the various peoples even ruled [ ηρχε] one another—the Medes over all and over those who lived closest to themselves, these in turn over their neighbors, and these over those near them. The Persians give 45. 1.134.2. Cf. the gradation of barbarism from a Greek point of view in Xen. Anab. 5.4.34.

152

Telling Wonders

 αυτ  δ ε

ον respect according to the same criterion [κατα τον   λογον . . . τιµωσι]. [Among the Medes] each people farther and   δ η το father removed held rule and supervision [προεβαινε γαρ   εθνος αρχον τε κα ι ε πιτροπευον]. This imprecise analogy between the Persian system of “graduated respect” and the Median system of “graduated rule”46 connects ethnocentrism, the primary ethnographic feature of the Persians, to the notion

η),  which turns out to be their primary feature in the of empire (αρχ history. An imperialistic ideology also emerges from a different order of signs. The stark simplicity and prescriptiveness of some aspects of Persian culture combine with acquisitive tendencies. The Persians “believe that  δ η γεαται  ε ιναι] (1.136.1).  multitude is strength” [το πολλον ι σχυρ ον This taste for to pollon manifests itself in their insistence on having many trimmings to their meals (1.133.2), many legitimate wives, and an even greater number of concubines (1.135). Every year, the one who displays the most children receives a prize from the king (1.136.1). In the historical narrative, royal Persians delight in possessing large armies and many subject nations.47 Though multitude signifies masculine strength and though acquisitiveness is part of the Persian desire for domination, acquisitiveness also represents the most indulgent and feminine side of Persian culture. Despite their contempt for foreigners, “the Persians like to appropriate foreign customs more than does any other people” (1.135). To their austere nature religion, they have added the cult of Aphrodite, learned from Assyrians and Arabs (1.131.3), and they import from other cultures luxu

ries and “pleasures of all sorts” [ευπαθε ι ας . . . παντοδαπας]. They have adopted the Median national costume (“because they considered it more beautiful than their own”), Egyptian corselets, and the Greek practice of 46. This is the terminology of How and Wells (1928, 1:116), whose interpretation I have followed in my translation. The analogy misleadingly implies that just as the nation farthest from the Persians is lowest in their hierarchy of esteem, so the nation farthest from the Medes used to be lowest in the hierarchy of power. 47. Xerxes enjoys the sight of his army at 7.44 and counts it at 7.59– 60. An army is a typical Persian gift (see 9.109.3). Konstan (1987) examines many instances of the fondness of Persian kings for quantity and numbers in connection with imperialism, but he strangely does not cite any part of the ethnography. Here also the judicial rule of “calculating whether the offenses are more in number than the services” (1.137.1) is consistent with the habit of counting and measuring goods.

Interpretation and Evaluation

153

making love with boys (1.135). The Persians are accumulators of consumption goods. In the historical narrative, Atossa desires accomplished Greek maids, and Mardonius praises the orchards of Europe in the context of imperialistic schemes.48 We should underline the extent to which the Persian ethnography avoids mentioning orientalistic features, even some that regularly occur in the historical narrative. The description of Persian customs keeps the role of monarchical despotism to a minimum. In the Lydian ethnography, the Lydians en masse, in their triple socioeconomic subdivision, raise a monument to their king with the fruit of their labor (1.93). Here the Persians worship the gods (1.131– 32), celebrate birthdays (1.133.1– 2), engage in deliberations and social interaction (1.133.3– 1.134); they pursue pleasure and war and raise large families (1.135– 1.137.1). When they sacrifice, they include a prayer on behalf of the king (132.2). Honor to the king represents a portion of their busy lives, and monarchy itself is a part of a social structure, marked by a hierarchicalism that is stereotypically “oriental” but at least suggests a distribution of rights according to status (1.134.1). What the history represents as prerogatives of the Persian king (the prostration and the birthday banquet, elsewhere called tukta) are here cultural forms with broader applications.49 Conversely, the king is subject to rules of fair dealing with his inferiors, according to criteria that also apply to ordinary Persians.50 The equivalence between the relationship of master and servants, on the one hand, and that of king and subjects, on the other (1.137), implies the same metaphor of political enslavement that we find in the history. But the history tends to emphasize the overbearing and punitive actions of the king, his violation of established custom in various spheres,51 his subjects’ service to him, and the powerlessness of the Persians in general, regardless of their 48. For Atossa, see 3.134.5; cf. 5.12– 13 (Darius likes the prospect of skilled Paeonian servant-women). For Mardonius, see 7.5.3; cf. 7.8α2 (Xerxes), 9.122.2 (Artembares wishes for a land better than the Persian). 49. Cf. Egyptian salutations (see 2.80.2). For proskunesis to the Persian king, see 3.86.2, 7.136.1. On the tukta, see the ethnographic gloss at 9.110.2. 50. See Briant 1988, 85. 51. The Persian nomos not to inflict grave penalties at the first offense or before balancing offenses and benefactions is followed only once (7.194.2; see 3.127 for a reverse case) and broken several times by the king in the history. See Lateiner 1989, 153. See 3.35, 36; 4.84; 7.38– 39; 8.118 (though this story is refuted on different grounds). For royal violations of the Persians’ traditional reverence for rivers (1.138.2, 1.131.1), see especially 1.189, 7.35. Xerxes’ mutilation of dead Leonidas is explicitly glossed as a violation of  Persian custom (παρενοµησε, 7.238.2).

154

Telling Wonders

rank.52 This description, by contrast, represents the assertiveness of the ethnos as a whole and places it firmly in control of its own nomoi. As it downplays the monarchical code, the Persian ethnography also leaves out all references to nomoi that involve abusing the body.53 Impaling and castration are not mentioned. Cutting off the ears and nose of a wrongdoer, flaying, and other punitive practices are adumbrated by the antiseptic

ηκεστον  euphemism αν πα θος, “incurable harm,” in the section that describes the Persians’ judiciousness in disciplining their slaves or subjects.54 In the history, suffering mutilation even appears as a sign of valor, from the actions of Zopyrus and Boges and from the Persian soldier’s fascination with the horrible wounds of the Aeginetan Pytheas.55 In the ethnography, Persian andragathie simply means “fighting well” and producing many sons (1.136.1). Human sacrifice, attributed to the Persians on several occasions in the history, is out of the question.56 The narrative even omits the moment of the victim’s slaughter in the context of animal sacrifice. We find instead a description of the sacrificer’s cutting of the meat in tiny pieces, which are fastidiously arranged on a soft bed of grass.57 Among the vulgar stereotypes that are surprisingly transformed is the barbarian addiction to wine: Herodotus’ Persians put it to good use in deliberations.58 To the 52. See especially the episode of the anonymous Persian at Attaginus’ banquet (9.16), which has also the function of dissociating the community of the Persians from the actions of the king, as Corcella (1984, 181) remarks. 53. In his Persian ethnography, largely based on Herodotus, Strabo finds it necessary to add the following item: “They are governed by hereditary kings; he who disobeys has his head and arms cut off, and his body is thrown off” (15.3.17). The funeral custom of the magi, which includes a ritual mutilation of the corpse (Hdt. 1.140), will be discussed later in the present chapter. 54. 1.137.1. In the Histories, mutilations are first and foremost terms of the monarchical code, not merely Persian or barbarian. See Hartog 1988, 332– 34. Passages mentioning royal Persian mutilations include 3.69.5 (cutting off of ear); 5.25 (slaying and flaying); 7.35.3, 8.90.3 (decapitation); 7.39.3 (cutting in half); 7.194.1– 2 (crucifixion, stayed); 7.238 (decapitation of corpse and impaling); 9.112 (mutilations by queen); 3.16, 3.27.3, 3.29, 3.30, 3.31.1, 3.32, 3.35 (mutilations by Cambyses). Mutilations are also attributed to other Persians, not necessarily acting on behalf of the kings (3.79.1, 3.118.2, 3.125.3, 6.30.1, 6.32). 

ηκεστος).  55. For Zopyrus, see 3.154.2 (called a λωβη αν For Boges, see 7.107.2. For Pytheas, see 7.181.2– 3. Cf. the andreie of the medizer Hegesistratus (9.37.2). 56. See 7.114, with an ethnographic gloss inferring that burying people alive is a Persian custom. See also 7.180, 1.86.2. 57. See 1.131.3. The omission is exceptional. See Gould 1994, 99. 58. See 1.133.3– 4. In other texts, the stereotype of barbarian drunkenness applies especially to Thracians and Scythians (see Hall 1989, 133– 34; chap. 2, n. 208 in the present book), but it fits in with the notion of barbarian intemperance. Cf. Cambyses at 3.34.1– 3. The Persian custom in the ethnography recurs among Tacitus’ idealized Germans (Germ. 22). See How and Wells 1928, 1:114.

Interpretation and Evaluation

155

amazement of Plutarch, the notorious Eastern use of eunuchs is here replaced by pederasty in the Greek style.59 The near absence of expected indices of barbarity (torture, despotism, lack of restraint), the attribution to the Persians of a strong collective voice in defense of their nomoi, and the intrinsic righteousness of some of the nomoi contribute to create a “Good” impression of the culture as a whole. The negative side of the evaluation is then only conveyed through the symbols and signs of imperialism, acquisitiveness, and material abundance. Among these, the banquet represents a key event in Persian culture, and Herodotus’ description of Persian meals plays an important role in representing the Persians historically.60 The Persians are, as the present tense of the ethnography describes them from the time of their conquest of Lydia. Previously they used to be, according to Croesus, hubristai (arrogant/violent) by nature, but poor (1.89.2). In the words of Sandanis (1.71.2– 3), they used to dress in leather and had neither wine, nor figs, nor any other good thing (all negatives of true lack). The narrator’s ethnographic-historical gloss corroborates this speaker’s assessment: “before the conquest of Lydia, the Persians had nothing good and no luxury  ουτε  ουδ  αγαθ

 αβρ  ον

εν]”  at all [ουτε ον (1.71.4). In the ethnography, which records their definitive cultural forms, the Persians are elegantly dressed (1.135). On their birthday, wealthy men serve a cow or a horse or a camel or an ass, cooked whole in the stove; a variety of side dishes; and a great deal of wine (1.133.3) If in the positive sides of their culture, the Persians appear as good as or better than the Greeks, Herodotus’ representation of Persian wealth and acquisitiveness implicates the Greeks explicitly in two different contexts. 59. See Hdt. 1.135. Plutarch writes (De Malign. Herod. 13 ⫽ Mor. 857C): “How can

the Persians owe the learning of this intemperance [ακολασ ι ας] to the Greeks, when practically everyone recognizes that this people has practiced castration of young boys before even seeing the Greek sea?” See Hall 1989, 157, for this stereotype. In Herodotus, the existence of a market for eunuchs in Persian-dominated Asia Minor is acknowledged in the ethnographic gloss at 8.105.2, which does not, however, connect it with sexual practices. Another orientalistic feature that Herodotus fails to attribute to the Persians is the (apparently historical) practice of consanguineous marriage (see Dissoi Logoi, DK 90 2.15), which in the Histories only appears as one of the monarchic perversions of Cambyses (see 3.31). See Mora 1985, 165– 66. 60. In the historical narrative, banquets mark crucial moments at the beginning of Persian history (see 1.125– 126); at the end of their imperialistic dream, with the defeat of Plataea (see 9.82); and in the disastrous aftermath at Xerxes’ court (see 9.110.2). For other instances, see 1.207.6– 7 with 1.211, 3.79.3, 7.119, 7.135.1. In Thucydides (1.130.1), the orientalized Pausanias “keeps a Persian table.”

156

Telling Wonders

The Greek contribution of pederasty to the range of Persian pleasures appears in a list of importations from abroad that recalls the celebrated influx of foreign comforts to Athens after that city’s rise to imperial status.61 Though the Persian banquet is still an index of the discrepancy between Persian and Greek cultures, “the Persians say that the Greeks stop eating when they are still hungry, because after the main meal nothing else to speak of is served to them, and if it were served, they would not stop eating.”62 Just as the Persians, who now eat what they want, used to eat not what they wanted but what they had (1.71.3), so the Greeks of Herodotus’ present either had first, now have, or still want the material pleasures the Persians have acquired. Positioned precariously between contrast and similarity, the Greeks partially share in the strength and the vulnerability Herodotus’s ethnography attributes to the Persians from the time of their conquest of Asia to that of their defeat at Plataea. Dispassionate Narrative and the Limits of Relativism When the metanarrative does not intervene to express doubt or mention a source, ethnographic descriptions rely on a presumption of autopsy.63 The position of Herodotus with respect to the ethnographic material he presents is analogous therefore to that of the author of a modern ethnographic report. The facts are external to the text, and the narrator is not responsible for them. The problem is, then, how things are said more than what is said. How does one produce a culturally unbiased—that is, nonsensational and nonjudgmental—description of a foreign culture whose customs are inherently bound to produce a “Bad” impression? One solution is to avoid mentioning such features, and I have argued that the Persian ethnography is somewhat selective precisely in this sense. But this is not a practice an ethnographer can resort to throughout a work. The question of how things are said is raised by the argument of Hartog that the objective, dispassionate, nonevaluative style of discourse that Herodotus generally favors when describing foreign culture is in fact one of the many tricks of his trade, a part of his “rhetoric of otherness.”64 We should then consider other style options. In the context of a discus61. 1.135. Cf. Thuc. 2.38; [Xen.] Ath. Pol. 2.7. 62. 1.133.2. Cf. 9.82 with Thuc. 1.130.1– 2. 63. In ethnographic descriptions, hearsay plays only a very limited role. See, e.g., 1.183.1– 3 (description of Babylon); 4.176, 4.178, 4.183.4, 4.184.3, and 4.187.3 (ethnography of Libya). 64. Hartog 1988, 256.

Interpretation and Evaluation

157

sion about the possibility of our understanding cultural forms that are alien to ourselves, Clifford Geertz quotes L.V. Helms’ description of human sacrifice in Bali, which we may take as a modern term of comparison with Herodotean discourse.65 It is a striking and elegant passage, but since my specific purpose is to make a point about biased and unbiased narrative, I regrettably quote only a few excerpts, with not much regard for the continuity of the action described or the integrity of the account. 1. While I was in Bali one of these shocking sacrifices took place. The Rajah of the neighboring State died on the 20th of December 1847; his body was burned with great pomp, three of his concubines sacrificing themselves in the flames. It was a great day for the Balinese. It was some years since they had had the chance of witnessing one of these awful spectacles, a spectacle that meant for them a holiday with an odour of sanctity about it. . . . 2. They looked little enough like savages, but rather like a kindly festive crowd bent upon some pleasant excursion. The whole surroundings bore an impress of plenty, peace, and happiness, and, in a measure, of civilization. It was hard to believe that within a few miles of such a scene, three women, guiltless of any crime, were, for their affection’s sake, and in the name of religion, to suffer the most horrible of deaths, while thousands of their countrymen looked on. . . . 3. The victims of this cruel superstition showed no sign of fear at the terrible doom now so near. Dressed in white, their long black hair partly concealing them, with a mirror in one hand and a comb in the other, they appeared intent only upon adorning themselves as though for some gay festival. The courage which sustained them in a position so awful was indeed extraordinary, but it was born of the hope of happiness in a future world. From being bondswomen here, they believed they were to become the favourite wives and queens of their late master in another world. . . . Round the deluded women stood relatives and friends. Even these did not view the ghastly preparations with dismay, or try to save their unhappy daughters and sisters from the terrible death awaiting them. Their duty was not to 65. Helms 1882, 59– 66, quoted by Geertz (1983, 37– 39).

158

Telling Wonders

save but to act as executioners; for they were entrusted with the last horrible preparations, and finally sent the victims to their doom. . . . 4. The women were carried in procession three times round the place, and then lifted on to the fatal bridge. There, in the pavilion which has been already mentioned, they waited until the flames had consumed the image and its contents. Still they showed no fear. . . . Meanwhile their attendant friends prepared for the horrible climax. . . . The supreme moment had arrived. With firm and measured steps the victims trod the fatal plank; three times they brought their hands together over their heads, on each of which a small dove was placed, and then, with body erect, they leaped into the flaming sea below, while the doves flew up, symbolizing the escaping spirits. . . . This terrible spectacle did not appear to produce any emotion upon the vast crowd, and the scene closed with barbaric music and the firing of guns. This autobiographical narrative in the past tense describes a particular occurrence of a customary, if infrequent, event. The features that make Helms’ narrative the very antithesis of the altogether neutral fashion that Hartog rightly attributes to Herodotus, however, could just as well characterize third-person descriptions in the iterative ethnographic present. Even when we see no grammatical first person, the narrator, Helms, is pervasively there to direct the reader’s perception of it and to communicate his ideological stance.66 Evaluative modifiers occur at every turn, from “shocking sacrifices” in excerpt 1 to “barbaric music” in excerpt 4. What pretends to be the report of the native attitude is actually the narrator’s distanced interpretation of it. “A holiday with an odour of sanctity about it” (1) and “their duty was . . . to act as executioners” (3), for example, in no way reflect what “the Balinese say.” Actual native 66. In case there should be any doubt left, Helms’ narrative ends with an advertisement of narratability followed by a gloss of interpretation and evaluation: “It was a sight never to be forgotten by those who witnessed it, and brought to one’s heart a strange feeling of thankfulness that one belonged to a civilization which, with all its faults, is merciful, and tends more and more to emancipate women from deception and cruelty. To the British rule it is due that this foul plague of suttee is extirpated in India, and doubtless the Dutch have, ere now, done as much for Bali. Works like these are the credentials by which the Western civilization makes good its right to conquer and humanize barbarous races and to replace ancient civilizations.”

Interpretation and Evaluation

159

beliefs concerning the meaning of the event are judged to be absurd in the same breath as they are reported (3: “cruel superstition,” “deluded women”). Negative sentences signal amazement, especially with regard to the discrepancy between civilized appearances and the abnormal savagery of the ritual (2, 3, 4). Among the stylistic means that are available to Herodotus, the choice is between an evaluative and a neutral description.67 Herodotus prefers the latter form, which reproduces the style generally employed by the Hippocratics in their account of symptoms, but which eluded one of these authors, who undertook to discuss foreign peoples. It is more important to verify the extent to which a neutral ethnographic description fails to be truly neutral and why it does so in particular cases than automatically to rank objectification among the various devices for enhancing otherness. In Herodotus’ description of what the Scythians do with the skin of the enemies they kill in war (making napkins, cloaks, and the like), we find the following gloss.  ην αρα,  

  δερµα δ ε ανθρ ωπου κα ι παχυ κα ι λαµπρον σχεδον   δερµατων πα ντων λαµπροτατον λευκοτητι. [and, as a matter of fact, human skin turned out to be both thick and bright, almost the brightest in whiteness of all skins.] (4.64.3)  Two words here deserve attention. First, the particle αρα (as a matter of fact) expresses the sudden interest of an unexpected but revealing datum.68 Second, the imperfect ην (was, turned out to be) interrupts the string of ethnographic presents. To what time does ην refer? Evidently it  refers to the time of “fieldwork.” The construction ην αρα is a gloss of historie that places the narrator on the scene, inspecting the local crafts, as he brings his audience’s attention to the result of his observation. Somewhat analogous in tone to Herodotus’ implicit display of firsthand expertise with regard to garments made of hemp (4.74), this gloss is nevertheless bound to produce a different rhetorical effect, perhaps enhancing, rather than toning down, the sensational character of the description. It advertises the deliberately objective stance of an ethnographer who is 67. The type of discourse that puts this problematic on display, as in what Geertz (1988, 97) calls “author-saturated texts,” is of course not an option. 68. See especially Denniston 1934, 32– 43; Corcella 1993, 286. Herodotus’ nonevaluative description is equivalent to “minimally narrated narrative.” See chap. 1, “What Is Metanarrative?”

160

Telling Wonders

determined to eschew cultural bias and to apprehend a foreign universe on its own terms; but in the face of this data, he is bound to appear too determined. There are countless passages of this sort in Herodotus. In them, the absence of an ethical viewpoint in a detailed but compressed style of narration conveys an almost cheerful detachment of the narrator vis-a-vis ` his 69 subject. Because “no mode of composition is a neutral medium,” Herodotus’ professional disengagement, either voluntarily or involuntarily, sometimes produces both horror and humor.70 In other cases, however, the discourse attaches the factual description of a practice to its proper ideological motivations by means of cognitive statements, native reported speech, or the direct portrayal of native attitudes. This type of description is also likely to communicate a specific criterion for judgment. Exemplary in this respect is the account of a Thracian widow slaying (5.5), the ritual in Herodotus that most resembles the Balinese suttee described by Helms. Herodotus’ discourse builds up an animated momentum by sectioning the narrative by prospective sentences that increase expectation but withhold praise or blame. 1. And those above the Crestonaeans do this: Each one has many wives; when a man dies, there is a great competition among his wives, with intense pleading of friends and relatives on the following issue: which one of them was most loved by her husband. The report of the results of the competition continues to emphasize the enthusiastic adherence of all involved. 2. And the one who is chosen and awarded this honor, after receiving the praises of both men and women, is slain over the grave by her closest male relative and, having been slain, is buried with her husband. Finally, a cognitive statement reports the state of mind of the survivors. 69. See, e.g., the account of the Scythians’ human sacrifice to Ares (4.62.1– 4), with the last chilling sentence, “and there’s the arm, lying where it has fallen, and, somewhere else, the body.” 70. Rosaldo 1989, 46– 52, especially 49. For a genre of ethnographic writing that exploits humorous representation, see, e.g., Barley 1992, especially 74– 75.

Interpretation and Evaluation

161

3. The other wives consider it a great calamity; for this is for them the greatest shame.   µεγαλην ποιευνται᝽   [α ι δ ε αλλαι συµ␸ορ ην ονειδος γαρ σ␸ι  µεγιστον  τουτο γι νεται.] The sequence of joyful actions in passages 1 and 2, suddenly landing on the chilling predicate “is slain” [σ␸αζεται], strikes a sensational note (cf. 1.45.3). But all three sections convey the natives’ lack of ambivalence, without the incredulity Helms demonstrates at the apparent acceptance of the victims and their relatives. Although lack of pathos and even a comic element are here once again the price of detachment, Herodotus’ narrative implements two serious principles of ethnographic writing. It focalizes the custom through their owners, and though, unlike Helms’ description, it does not explain the native perception in light of existential beliefs, neither does it undermine or contest it. The case just examined communicates that a nomos that the ethnographer’s culture considers inhumane or unholy can be integrated into a relativistic view if it is internalized by all those whom it affects. Herodotus’ report about a custom of the Massagetae conveys the same message. No other limit of human life is set for them. But when one becomes very old, all his relatives assemble and sacrifice him and other animals together with him, and after boiling the meat, they feast upon  it [κατευωχεονται]. This is considered by them to be the most  σ␸ι νενοµισται],    τα ολβι ωτατα  blessed thing [ταυτα µ εν but if one dies of disease they do not feed upon him but rather bury him in the  ποιευµενοι]  ground, considering it a misfortune [συµ␸ορ ην that he did not reach the age of being sacrificed. (1.216.2– 3) The initial negative statement has the corrective function of dispelling the audience’s possible confusion between the Massagetan custom and more radical practices of other peoples.71 The combination of the code of   sacrifice (θυουσι, τυθηναι) with the secular term κατευωχεονται is jolting; but the statements about what this community regards as “most 71. It parallels the correction at 1.216.1. Asheri (1988, 386) cites the (explicitly evaluative) accounts by Strabo (11.11.3) and Aelian (VH 4.1) about cultures that fix the limit of human life at seventy or sixty years of age or kill the elderly or the sick in different ways (starving them, throwing them to the dogs, etc.).

162

Telling Wonders

  overcome the ideoblessed” [ολβι ωτατα] or a “misfortune” [συµ␸ορην] logical instability of the description by referring to a peculiar but deliberate and shared worldview of this society.72 The criterion of evaluation implicit in these passages is consistent with Herodotus’ explicit condemnation of ritual prostitution at Babylon (1.199.1) and with the implied disapproval of customs in which one portion of the society victimizes another. The Indian tribe of the Padaeans was particularly savage, according to ancient tradition, though predictably no statement to that effect appears in Herodotus.73 He merely calls them “eaters of raw meat” (3.99.1) and applies to their particular brand of cannibalism the same form of discourse employed for the Thracian and Massagetan funeral customs—a neutral description from the native point of view. When one of the townspeople gets sick, be it a man or a woman, if it is a man, the men who are his closest kin kill him, saying that as he wastes away with the disease, the flesh becomes spoiled for them. The other denies being sick; but hearing no reason, they kill him and   feast upon him [αποκτε ι ναντες κατευωχεονται]. And if a woman gets sick, the women who are closest to her do with her just as much as the men do with the men. For when one comes to old age, they  γαρ   δ ε ε ς γηρας απικ  sacrifice him and feast upon him [τον οµενον   θυσαντες κατευωχεονται]. But not many people reach this theoretical point, for before this, they kill anyone who falls sick. (3.99.1– 2) The final statement, with the narrator’s most intrusive evaluation, is the counterpart of the negative gloss introducing the Massagetan custom (1.216.1: “No other limit of human life is set for them”). In the immediately preceding sentence, the verbal correspondence establishes a parallel  . . . γερων   with the Massagetan ritual (cf. 1.216.2: ε πεαν γενηται     καρτα, . . . θυουσι µιν, . . . κατευωχεονται); this renders the differences between the two practices all the more conspicuous. The Massagetae sacrifice and eat the very old but do not eat those who die of disease, 72. They recall how unconventional Solon’s use of these terms appeared to Croesus (see especially 1.32.2– 9): see Flory 1987, 97. The phrase “to consider a misfortune” recurs at 5.5, quoted earlier, and at 4.79.5, in cognitive descriptions of native attitudes. 73. How and Wells (1928, 1:99) and Asheri (1990a, 326) suggest the derivation of the ethnic name from the Sanskrit padja, “bad,” and cite Tibullus 4.1.144– 45, testifying to the negative reputation of this tribe.

Interpretation and Evaluation

163

while the Padaeans kill and eat a person as soon as he or she gets sick, which means that few Padaeans reach old age. The killing of the elderly, both among the Massagetae and the Padaeans, is called a “sacrifice”   [θυειν], but the Padaeans’ killing of the sick is a “killing” [αποκτε ι νειν]; the secular verb κατευωχεισθαι occurs conjoined with both terms. Whereas for the Massagetae the dietary aspect of the custom is subordinate to its religious character, the reverse happens among the Padaeans. All the Massagetae agree that the sacrifice and the eating are a “blessed” thing, although as in the case of the Thracian widow slaying, we are not told on what grounds.74 The utilitarian/dietary ideology of the Padaean practice, by contrast, produces in their society a split between those who are killers and those whose turn it is to be victims, between those who benefit from the sacrifice and those who perceive they do not. The native voice is divided, as the typical dialogue embedded in the description testifies: the healthy say that the sick must be killed in a hurry, and the sick protest in vain that they are not sick. Herodotus’ cultural relativism is not ethical relativism in an extreme sense. Just as the moralistic historian objects to monarchical abuse, so the ethnographer signals disapproval of customs that oppress a society’s free members.75 This ethical principle is, however, supracultural and capable of accommodating a wide range of diverse practices and perceptions. If difference is not always “Good,” Herodotus’ logos is more concerned in establishing that it is not automatically “Bad.”76 This, the legitimacy of cultural difference as such, is his fundamental lesson as an ethnographer. One of the means through which he conveys it is his commitment to a nonevaluative mode of ethnographic description. Equal Knowledge: Cognitive Relativism Herodotus’ acceptance of the validity of foreign nomoi and native attitudes is subject to certain judgmental reservations in particular cases. In 74. Contrast the explanation given by the Dissoi Logoi (DK 90 2.14). See Mora 1985, 164. 75. See the evaluation implicit in the juxtaposition of Scythian royal funerals (4.71– 72) and ordinary Scythian funerals (4.73). 76. On Herodotus’ relativism, see Burkert 1988, 27– 28 and later discussion in the present chapter. I will apply the term relativism somewhat loosely to fifth-century thought (see Guthrie 1971, 164– 75; Kerferd 1981, 83– 110). See however the objections put forward by Bett (1989). On the modern controversy between relativism and antirelativism, and on the difference among different relativistic positions, see Geertz 1984.

164

Telling Wonders

the sphere of religious beliefs and beyond ethics, however, his cognitive relativism is more radical, theorized a priori, and often conveyed through the deliberate suspension of his normal activities of inquiring and narrating.77 As we have seen in the cases of the Thracian widow slaying and Massagetan cannibalism, Herodotus rarely describes the religious beliefs underlying customs. When he does, he is usually pursuing a polemic against the Greeks (e.g., at 1.131.1). His discussion of Salmoxis-Gebeleizis compensates for the ambivalence of his own description of the Getic ritual by exposing the more flagrant chauvinism of the Black Sea Greeks.78 After rejecting their version, which reduces Getic religion to an inferior byproduct of their own civilized culture, Herodotus quickly brings the section to a close: “Whether Salmoxis was a man or whether he is some local  Getic divinity, farewell to him [χαιρ ετω]” (4.96.2). Given the particular context, this rare form of programmatic conclusion recalls the formulaic hymnal farewell to the gods. The dismissal of the topic as such encodes a sort of agnostic acceptance of the religion.79 We have already seen how the first programmatic introduction of the Egyptian ethnography (2.3.2) connects Herodotus’ reticence to explain religious beliefs to his relativistic position.80 Here, his profession of belief (νοµι ζων) that “all men know equivalently about these things” [παντας   ισον περ ι αυτ  ων  ε πι στασθαι] goes beyond the evidence that ανθρ ωπους components of the different religions of the world can often be translated from culture to culture.81 Rather, it means that, even regardless of similarities or overlaps, the degree of accuracy possessed by the whole body of theological beliefs of one society is equal to the degree of accuracy possessed by the whole body of theological beliefs of another. When Herodotus attributes an equivalent value to the religious knowledge of different cultures, from the monotheistic Getae to the complicated Egyptians, this of course includes the idea that divine matters “cannot be   η and γνωµη  objects of enquiry ( ι στορι η) by οψις, ακο and, hence, can77. On cognitive relativism and related attitudes in modern thought, see Hanson 1979. 78. On Herodotus’ ambivalent evaluation of the Getae, see “Explicit Evaluation” earlier in this chapter.  79. Cf. Hom. H. 1.20, etc. The only other conclusion with χαιρετω occurs at 2.117 (see Lateiner 1989, 63), but the dismissal that is closest in function to 4.96.2 is at 1.140.3, quoted later in this chapter. On programmatic conclusions, see chap. 1, “What Is Metanarrative?” 80. See chap. 2, “The Texture of Nomos.” 81. This evidence exists, e.g., in the case of major divinities, worshiped by different peoples with different names. See Hartog 1988, 107 and n. 162, quoting Veyne 1971, 141.

Interpretation and Evaluation

165

not be objects of certain knowledge.”82 This is again too reductionist a statement, however, if by it one means that for Herodotus the real state of affairs is entirely out of reach. It is not Herodotus but Protagoras who denies the possibility of human knowledge concerning the gods. Consider  the first sentence of Protagoras’ Περ ι θεων. About the gods, I am not able to know that they exist or do not  θεων  ουκ  ε χω ε ι δ εναι,  exist or what shape they have [περ ι µεν  ε ι σ ι ν ουθ  ουκ  ως  ως  ε ι σ ι ν ουθο   ποιοι τινες ι δ εαν],  ουθ because many are the factors that prevent one from knowing, such as their invisibility and the shortness of human life. (DK 80 B4) The indirect questions in this passage bear resemblance to those Herodotus formulates in another interpretive gloss (2.53): here he asserts,  on his own authority (ε γω λεγω), that before the codification of knowledge recently effected by Homer and Hesiod, the Greeks “did not know whence each of the gods came into existence, whether they were forever, !  !  θεων,  and what kind of shape they had” [ οθεν δ ε ε γενοντο εκαστος των 83 "             ε ιτε α ιε ι ησαν πα ντες, οκοιο ι τε τινες τα ε ιδεα, ουκ ηπιστεατο]. This statement is designed to cut Greek theological representations down to size rather than to proclaim their validity as real knowledge.84 At 2.3.2, however, Herodotus generalizes in positive terms: all men know ison about divine things. They all “really know” something, and they all know an (indeterminably) equal amount.85 As an inquirer of the past, Herodotus is able to identify actions that appear attributable to something other  the than human planning or random contingencies—το θει ον, ο θεος, divine in purely theological terms—with no need or possibility to be more specific.86 At the same time, however, he also verifies the existence of a line of communication between human beings of all nations and god (as an ontological principle transcending culture) that passes through cultural channels, such as prayer, sacrifices, and oracles. A space that is 82. Lloyd 1976, 17. 83. 2.53.1. Burkert (1988, 26) cites the fragment by Protagoras as a parallel to this passage; Lloyd (1976, 17) cites it as part of the intellectual background of 2.3.2. Both times,  Herodotus’ ε πι στασθαι, in lieu of Protagoras’ ε ι δεναι, places less emphasis on the sensible origin of the knowledge. See Untersteiner 1967, 1:67 n. 37. 84. Cf. 7.129.4. See Burkert 1988, 20– 22. 85. In positive terms, they know not “equally little” (Stein 1883, I.2.6) but “equally much” (see, e.g., Grene 1987, 666). For a different rendering, see Mora 1985, 136– 39. 86. See Linforth 1928.

166

Telling Wonders

sacred by nomos is sacred in absolute terms; the god whom “they” worship in the shape of a bull is objectively divine.87 The objectivist component of Herodotus’ theological relativism is a paradox that preserves traditional Greek polytheism, reinterprets its openendedness, and places foreign religions on a par with it. It also determines the narrator’s reluctance to take on τα θεια as a part of his ethnographic subject. To inquire and explain is to question and therefore, potentially, to negate. The limiting case of perverted theological inquiry is Cambyses’ brutal practical test of the divinity of the Apis bull (3.27– 29). In matters of religion, however, even verbal inquiry, which goes hand in hand with a sort of arbitration, is at least dangerous. When the narrator displays his own historie in the secular sphere, contradictory logoi that are pitted against each other result, implicitly or explicitly, undermined; indeed, in such cases, to expose their subjectivity and the instability of truth is for Herodotus an important goal. In a somewhat parallel fashion, to discuss the hieroi logoi (sacred history) of different peoples will raise questions concerning their correctness, absurdity, or impropriety, as well as the intellectual and cognitive skills of the people who hold them as true. Though at very different levels, both the vulgar account of the Black Sea Greeks concerning Salmoxis and the philosophical polemic of a Xenophanes of Colophon are deconstructive criticisms of culturally determined religious beliefs. Per of Protagoras had put accounts of different theologies haps the Περ ι θεων 88 to a similar use. As the promoter of the validity of difference, Herodotus sometimes needs to shake his audience’s self-assurance in their own religious traditions or justify those of others; precisely these motivations, more often than not, provide the “overwhelming compulsion” (see  ε ξαναγκαζοµενος in 2.3.2) to discuss sacred knowledge.89 But to subject foreign hieroi logoi to that kind of scrutiny, to expose them to ethnocentric debate, sophistic skepticism, and perhaps ridicule—this, as a rule, he is   . . . προθυµος] “not eager” [ουκ to do (2.3.2). 87. For sanctuaries, see, e.g., 9.65.2, with a cautionary gloss of opinion. A gloss of comparison (at 3.64.3) implicitly presents Cambyses’ death as evidence of the divinity of Apis. 88. Philostratus (DK 80 A2) makes Protagoras’ statement derive from a conversation with the Persian Magi; see Untersteiner 1967, 1:56. For a denial of the possibility of knowledge about the gods, cf. Xenophanes, DK 21 A34. 89. See 1.131.1, 2.43– 44 (concluded with an apology to gods and heroes at 2.45.3), 2.143– 46. For the different case of 2.156.4– 5, on the myth of Leto, see chap. 4, “Herodotus and the Conventional Code.” For the myth at 2.42.3– 4, see chap. 2, “The Texture of Nomos.”

Interpretation and Evaluation

167

Funeral Customs and Other Nomoi: Cultural Relativism The only mutilation tentatively attributed to the Persians in the ethnography is the practice of exposing the dead to be torn by a bird or a dog. After this treatment, which recalls the Homeric indignity against the bodies of one’s enemy, the Persians cover the body in wax and bury it.90 Clearly marked off by a retrospective/prospective system (1.140.1), this custom is deliberately not integrated into the rest of the description, and it is not allowed to contribute to an understanding of Persian ideology or beliefs. The ritual simply exists, either among the Magi or among both Persians and Magi, and it is different, just as the Magi are unaccountably different from the rest of mankind also in other respects (1.140.2). One sign of the custom’s unfathomable arbitrariness is that the narrator, some what as he does for Salmoxis, bids it farewell: “let it be [ε χ ετω], just as it has been established as a custom to begin with” (1.140.3). This ritual appears as the first of the funeral customs in the Histories; it constitutes the limiting and representative case of them all—opaque, arbitrary, of unknown origin, and a symbol of difference, yet not entirely unique. Only one other ethnos in the Histories abandons the dead to corruption (3.100), but many embalm and bury. Similarly, only one ethnos celebrates joyfully when someone dies (5.4), and only one (5.8) holds athletic contests, though that is the old heroic manner. Several eat or cremate the dead or make their skulls into objects; several sing dirges, sacrifice, or mark tombs.91 Individual features recur again and again in different combinations, revealing both people’s mysterious opportunity for difference and the limit of that difference. And since each ethnos is bound to do something—which is always the same ritually prescribed 90. Mutilation of corpses by “dogs and all the birds” in Homer (Il. 4– 5; cf. Il. 2.393, 4.237) is not value-free. See Segal 1971, 9– 17. Cf. Soph. Antig. 204– 6, 258– 59. Boedeker (forthcoming) contrasts Hdt. 1.140 with Artabanus’ “Homeric” threat to Mardonius at Hdt. 7.10θ3. In this case, the discrepancy between Herodotus’ Persian ethnography and history (see discussion earlier in the present chapter) follows an inverted pattern. 91. Embalming (with different techniques) is performed by the Babylonians (1.198), the Egyptians, the Ethiopians (3.24), and the Magi. Burial is practiced by the Magi, the Persians, and the Massagetae (1.216.2), the Libyan nomads, the Nasamones (in a sitting position, 4.190), and the Trausians (5.4). Eating the dead occurs among the Massagetae (1.216.2), the Callatians (3.38.4), the Padaeans (3.99), and the Issedones (4.26.1). Cremation occurs among the Greeks and the Lydians. Skulls are made into objects by the Issedones (skulls of loved ones, 4.26.2) and the Scythians (skulls of enemies, 4.65.2). Dirges are sung by the Babylonians, the Egyptians (1.198), and the Thracians (5.8). Tombs are used by the Greeks and the Nasamones (4.176).

168

Telling Wonders

act—with their dead, funeral customs are inevitable signs of culture. They belong to every society’s body of most compulsive norms—the ι ρα   in the proper τε κα ι νοµαια—and mediate between what is sacred ( ι ρον) sense, connected with the cult of the gods, and what we would more loosely call “sacred.”92 A society’s ritual disposal of dead bodies has no bearing on relations with others, and it does not involve issues of justice or injustice toward its own members in the same way as do customs that have to do with the treatment of the living. In Herodotus, therefore, funerals are an important cultural symbol. They illustrate the sense of alienation that foreign cultures inspire simply as they go about their business of being different from ourselves; consequently, they allow the ethnographer to theorize on cultural relativism in the broadest possible terms. This happens in a far-ranging interpretive gloss that, standing at the intersection of ethnography and history, generalizes on the basis of the narrative about Cambyses’ behavior toward custom (3.38). Cambyses is the extreme representative of Herodotus’ negative paradigm for historical and ethical action, the monarchical model. He is also a scientist of sorts, a researcher of nomoi, and in this capacity the foil of the histor of the Histories.93 Their shared field of observation is principally Egypt, a fundamental source of theoretical and anthropological learning for Herodotus, and a land Cambyses oppresses as well as observes. Through his brutal testing in the sphere of custom—from religion, to funeral procedures, to marriage laws (3.16, 31, 35.5)—Cambyses attempts to find out whether practices, cultural beliefs, statutes, public opinion, and common morality will hold out under skeptical critique, objective scrutiny, and external force. Herodotus’ own respectful inquiry on the value of nomos here takes advantage of the historical case of Cambyses himself. To Herodotus, Cambyses represents especially useful evidence for two reasons. First, his violations of the nomoi of others are strictly connected with his destruction and deconstruction of less culture-specific ethical rules and of the laws of his own society. In symbolic terms, the fulfillment of his early promise of putting everything upside down in Egypt corresponds to the unprecedented upside-down burial of his own Persian sub92. Heinimann (1945, 79) gives nomos as applied to these practices the meaning of fas. In the taxonomy of ethnographic descriptions, funeral customs are sometimes adjacent to religion, other times not. 93. See Munson 1991, 59– 60; Christ 1994, 186– 87.

Interpretation and Evaluation

169

jects.94 Second, Cambyses’ madness is not a hyperbole for criminal behavior but a clinical (i.e., physical) illness.95 This means that Cambyses’ dysfunction with respect to all sorts of customs—foreign and native, religious and secular—coincides with an impairment of the mind that clearly belongs in the sphere of phusis (nature). By Herodotus’ time, phusis, nomos, and the divine, once blurred in archaic Greek thought, have grown increasingly distinct.96 As far as Herodotus is concerned, however, Cambyses’ triple abnormality demonstrates e contrario the existence of a close connection between divine will, nature, and culture as a universal phenomenon.97 Herodotus’ conclusion at 3.38.1 begins by interpreting Cambyses’ deri sion of religion and serious customs ( ι ρα τε κα ι νοµαια [things sacred and pertaining to custom]) as evidence that confirms his madness  is here an understatement that in(ε µα νη). But “derision” (καταγελαν) vites vertical analogy between the crimes of Cambyses (accompanied by laughter: 3.29.1– 2, 3.35.3, 3.37.2) and more commonplace manifestations of contempt. In fact, the case of Cambyses will be paradigmatic for  narrator and audience: any one who makes fun (γελωτα . . . τι θεσθαι) of such things is mad (3.38.2). The positive counterpart of this insane individual is normal people in general, who simply scrutinize all sorts of customs (διασκεψαµενοι) and are attached to their own. Protagoras may ultimately be the source for the interpretive hypothesis at 3.38.1 that if “all men” were able to choose from a display of nomoi, they would choose their own, believing them to be the most beautiful/  καλλι στους). In a very similar passage of the sophistic honorable (τους Dissoi Logoi, the idea has been put to the service of a “different strokes for different folks” brand of relativism that denies the absolute validity of  and the shameful (το the ethical concepts of the honorable (το καλον) 94. 3.3.3, 3.35.5. See also the mutilation of the body of Amasis, by which Cambyses violates both Persian and Egyptian customs (3.16). 95. Explicit references to Cambyses’ medical madness occur at 3.30.1, 3.33, 3.34.1, 3.37.1, and 3.38.1. Only Cleomenes shares this distinction. See chap. 2, n. 71 and corresponding text. 96. See especially Heinimann 1945, 110– 62; Guthrie 1971, 55– 147. Cf. chap. 2, “The Texture of Nomos.” 97. Individual nomoi have of course human origin in Herodotus. See Evans 1965, 145– 46. However, the idea of Nomos as an overriding impulse to culture recalls Heraclitus DK 22 B114 (“all human laws are nourished by a single law, which is divine; for it has as much power as it wishes and is sufficient for all and is still left over”). See Heinimann 1945, 65– 66.

170

Telling Wonders

 98 Herodotus, however, is already traveling in a different direcα ι σχρον). tion: the lack of objective validity in people’s perception that their own nomoi are καλοι only goes to show that all nomoi are equally καλοι . To deride them is therefore madness. The subjectivity of all men, illustrated by the hypothetical scenario of  the display of nomoi, is next confirmed by the evidence (τηκµηριον) of the actual experiment of another royal histor (3.38.2). Darius once proposed to the Greeks that they eat their dead parents and to the Callatian Indians, who customarily eat their parents, that they burn them instead; both groups refused to practice the other’s nomos (3.38.3– 4). Other than showing again that each likes his own customs best, the exchange also dramatizes the repulsion that the harmless other is likely to inspire. The subjectivity of each party’s reaction is enhanced by the fact that the competing nomoi are here no longer the unidentified assortment of the hypothetical scenario envisioned earlier. They are specifically funeral customs, the most suitable for conveying the principle that diverging cultural norms are not α ι σχροι and, since they are all equally compelling, are inherently καλοι . Here the implied equivalence of all men’s nomoi from the point of view of their moral goodness corresponds to the equivalence of all men’s religious beliefs from the point of view of their truth-value at 2.3.3. On the basis of this equivalence, Herodotus’ next generalized statement no longer follows the distributive pattern of 3.38.1—“all men would each  !  ε ωυτων  νοµους] [εκαστοι] believe that their own customs [τους are the most beautiful.” The unified formulation borrowed from Pindar, rather, posits a single human community and, over it, a single rule.99 Thus are these things determined by nomos, and it seems to me that Pindar was right when he said in poetry that nomos is king of all.  !  νυν ταυτα  νενοµισται,  µοι δοκεει  Πι νδαρος [ουτω µεν κα ι ορθως    ␸ ησας  ποιη σαι νοµον παντων βασιλεα ειναι.] (3.38.4) In this definitive maxim, nomos comes to mean custom/law/culture as something abstracted from this or that nomos or set of nomoi. It encompasses all humankind (Androphagoi excepted: 4.106). It represents the 98. DK 90 2.18 (cf. 2.26). On the connection between the Dissoi Logoi and Protagoras, see Lasserre 1976, 73– 74; Robinson 1979, 51– 59. 99. For the meaning of Pindar frag. 169 SM in its original context, see especially Gigante 1956, 72– 102; Schroeder 1917; Stier 1928; Ostwald 1965; Humphreys 1987. Another intriguing fragment of Pindar (215 Bergk) expresses the idea of cultural differentiation.

Interpretation and Evaluation

171

universal fact of having nomoi, whatever these may be, and of behaving according to them, in culturally determined ways. At 3.38, Herodotus takes the opportunity offered by Cambyses to supplement in a crucial way the message that the ethnographies communicate to the audience. The inquiry into the customs of foreign peoples and the presentation of such inquiry to an audience navigate between the Scylla and Charybdis of ethnocentric absolutism and ethical relativism, respectively. Our ethnological observation may reinforce the sense of our  in comparison with worse communities who do monopoly over το καλον things differently than “we” do. As we have seen, the Histories include more or less automatic negative evaluations (“they couple like animals”) as well as ultimately presupposing an absolutist—and deliberate—view of what is kalon (beautiful/moral) or aischron (shameful/immoral). But though certain particular customs may indeed be aischroi (see 1.199.1), the ethnographic discourse of Herodotus attempts to steer the audience clear of unilateral chauvinism toward barbarian customs as such. A relativism according to which different nomoi are on principle equivalent in function and worth is fundamental to this lesson. For someone like Cambyses, however, the realization that different peoples have different nomoi with roughly the same validity leads to denying the validity of them all. This is the Charybdis, the second danger that Herodotus addresses in this gloss. The monarchical position of Cambyses, who obeys his own personal law and “does what he wants” (3.31.4), is in fact similar to the ethical monarchism and extreme relativism of a theoretician of the ilk of Plato’s Callicles.100 To Cambyses/Callicles, nothing is “sacred,” while Herodotus’ brand of relativism teaches that every  [Custom/ thing is. Callicles attributes the saying νοµος παντων βασιλευς Law king of all] to the law of nature that lets the stronger prevail, which is, according to him, the only law that counts (Plato Gorg. 483B). We should not doubt that the phrase served to validate sophistic thought long before the time of the Platonic dialogue. Herodotus, at any rate, uses the same phrase to support a position that is the opposite of that of Callicles and to replace Cambyses as the king- despotes (and 100. Plato Gorg. 482E– 484D. The parallelism between Herodotus’ Cambyses and Plato’s Callicles is enhanced by the continuous intrusions of the monarchical code in Callicles’ speech. Callicles’ models are the Persian kings Xerxes and Darius (483E); he speaks of strong men coming of age as “lions” (483E) and as kings’ sons (492B); he claims (491E) that they should give free rein to their desires (ε πιθυµι αι, a monarchic word in Herodotus: see chap. 2, n. 232) and that they should not impose on themselves conventional nomos as a despotes (492B).

172

Telling Wonders

researcher and destroyer of custom) with the kingship of custom and conventional law. The exchanges Herodotus describes throughout his work show to what extent both Greeks and non-Greeks—“all men,” in other words— are assiduous observers and critics of each other’s nomoi. For each ethnos of histores, the sense of allegiance to their own nomoi cannot prescind from the realization of the allegiance of others to theirs. Couched in these terms, Herodotus’ ongoing polemic against cultural chauvinism, cultural imperialism, and racism—all of which allegedly preserve a person’s attachment and obedience to his own nomoi while allowing for his contempt toward others and theirs—takes the remarkable form of an ideology that squeezes this double standard out of existence. As in the case of Cambyses, imperialistic contempt for others is madness that overrides all laws. Ethnocentrism is universal because, to paraphrase Herodotus with the words of Geertz, we “cannot escape preferring our own preferences.”101 But for the ethnographer of the Histories, morality, sanity, and piety at home as well as abroad are contingent on the acceptance of the universal and absolute rule of culture. This entails respect for the subjectivity of others on a par with ours and the belief, from an objective point of view, in the essential moral equivalence between burning and eating the dead. Thus, the notion of the abnormality of foreigners is replaced with the idea that abnormal is the one who derides their nomoi. Interpretation in the History This section focuses on those metanarrative passages where the narrator directly intervenes to explain in his own voice what an event of the past “means.” The meaning of a specific historical action or event is in most cases connected to its motives, causes, and results; an interpretation of these factors may in turn indicate the action’s worth.102 I begin by examining two fundamental glosses, both the narrative of Xerxes’ invasion, where the interpretive and evaluative functions are strictly combined. One is the famous judgment about the consequences of Athenian policy of naval resistance (7.139); the other is the explanation of Leonidas’ decision to remain at Thermopylae (7.220). These two interpretations are almost symmetrical, one about Athens and the other about Sparta, each 101. Geertz 1986, 261; cf. 257. 102. So Plutarch (De Malign. Herod.) understands Herodotus in terms of praise or blame.

Interpretation and Evaluation

173

vis-a-vis ` Persia and in contrast to other Greeks. Both statements are clearly marked by self-referential signs of opinion as interventions in the voice of the narrator. Specific Glosses of Interpretation: Sparta and Athens The first passage anticipates at long range the Salamis narrative and introduces at short range the account of Athenian deliberations before Xerxes’ invasion (7.140– 43). It is in turn programmatically introduced as a gnome, or opinion, that will cause resentment. At this point, I am compelled by necessity publicly to display an  odious interpretation [γνωµην ε πι ␸θονον] for most people, but   I shall not still, since it appears to me to be something true (αληθ ες), refrain. (7.139.1) The interpretive gloss itself rushes on in a rhetorical fugue of contraryto-fact past conditions sketching the scenario of what would have happened if the Athenians had not opposed Xerxes on the sea. It ends by explaining the meaning of the Athenians’ initiative in terms of both mo tives (“since they chose [ε λοµενοι] that Greece remain free”) and results (“one saying that the Athenians were the saviors of Greece would not 103   miss the mark of the truth [ταληθ εος]”). In his study of the concept of aletheia (truth) in Greek literary discourse, Detienne has traced the early use of the word to connote the uncontested truth that emanates from the gods and finds expression in prophecy and poetry. In the context of the polis, aletheia enters in competition with doxa (opinion), which informs the secular and more provisional discourse of public debates among peers.104 Herodotus rarely invokes aletheia as the foundation of his logos or as a realistic goal of his inquiry.105 Though accompanied by the narrator’s more normal vocabulary of opinion and evidence, this is the Histories’ most unambiguous proclamation of “truth” in the sphere of human knowledge. It emphasizes not only the maximum certainty of the gnome in terms of evidence (␸αι νεται) but also its general validity, its nonlocal and nonrelative status. We can perhaps transfer to historical reports the connotation, 103. 7.139.1– 5. See Demand 1987 for the rhetorical aspects of this passage. 104. See Detienne 1973, 25– 47, 81– 110. 105. On the predominance of opinion over truth in Herodotus, see Darbo-Peschanski 1987, 164– 89, especially 165– 67.

174

Telling Wonders

which Nagy applies to the realm of poetic traditions, of aletheia as the feature that characterizes the canonical version and excludes all other local variants.106 The phrase “one saying that the Athenians were the   saviors of Greece would not miss the mark of the αλη␽ ες” (7.139.5) canonizes what happens to coincide, on the whole, with the Athenian version of the war.107 It validates it as the Panhellenic version, since one (τις) can refer to any member of the audience, whether Theban, Corinthian, or Argive, or any Greek, including the floating histor. Herodotus here publicly performs a gnome that alone must be and must remain accepted by all Hellenes. It is, as he says, compulsory. The true gnome is compulsory but unpleasant. It is ε πι ␸θονος, likely to make those who express it the objects of phthonos (envy) for most men, because so is Athens, the tyrant city whose self-glorifying claims it confirms.108 A contemporary argument, surely a commonplace response to those Athenian claims in the 430s, was to point out that the Athenians perhaps behaved well in the war against the Persians but now were behaving badly toward the Greeks.109 Herodotus here reverses the terms and places the Panhellenic gnome about the past in the foreground while alluding to the accusations concerning the present: if Athens is ε πι ␸θονος to the Greeks now, it has been, at the time of Xerxes’ invasion, their savior. This clear-cut judgment frees Herodotus in the subsequent narrative of Salamis and beyond to continue indicating to his audience the signs of trouble to come within the available record of past events.110 Herodotus’ sense of 106. Nagy 1990, 59– 68. 107. See Loraux 1986, 58. See the Athenian speakers in Thuc. 1.73.4– 74.4, especially the contrary-to-fact condition at 1.74.4; cf. Lysias 2.45. The Peloponnesian version would have maintained the primacy of the battle of Plataea (see Immerwahr 1966, 240 n. 8) and minimized the merit of the Athenians by emphasizing the strategical errors of the Persians (see the Corinthians in Thuc. 1.69.5).  108. See Thuc. 2.8.5, and 2.64.5 (το ε πι ␸θονον). A typical recipient of ␸θονος is the tyrant: see Hdt. 3.52.4– 5, 7.236.1. See McGlew 1993, 31– 33. For the notion of the tyrant city, see chap. 2 n. 49 and corresponding text. 109. See Sthenelaidas in Thuc. 1.86.1. Whatever one thinks about the historicity of  Thucydides’ speeches (see Hornblower 1987, 45– 72, for a discussion), τα δεοντα (1.22.1) guarantees the historicity of the political code—types of arguments, word combinations, and so on. 110. On the qualified nature of Herodotus’ praise of Athens at 7.139, see Payen 1997, 189– 93. I have examined this aspect of Herodotus’ account of the battle of Salamis in Munson 1988, and see further “Disputes, Arbitration, and the Subjectivity of Opinions” later in the present chapter and chap. 4, “Vertical Analogy.” The evidence throughout the Histories does not support Evans’ view (1979, 117) that 7.139 demonstrates Herodotus’ acceptance of the moral justification of Athenian imperialism.

Interpretation and Evaluation

175

continuity between past and present often produces a representation that is nonidealized or, as Plutarch qualifies it, “malicious,” because it interprets the past in light of the present.111 This form of revisionism is nevertheless conjoined with the determination to keep the record straight and not to revise history in a way that detracts from past achievements. The result is a discourse that tends to swing back and forth between explicit praise and more covert blame. The hypothetical history of the war minus the Athenians, which supports the interpretation at 7.139 that “the Athenians were the saviors of Greece,” involves an ambivalent portrayal of the efforts of everyone else—the strategic futility of the wall across the Isthmus, the role of the nonmedizing Greeks as mere followers, and the practical uselessness of the isolated valor of the Spartans. With the realistic hypothesis that even the Spartans, when all hope would be lost, might have come to terms with the Persians, Herodotus slightly corrects the impression just conveyed by the narrative about the intransigent courage of Sperthias and Boulis.112 This instability of evaluation is typical of Herodotus’ interpretive technique in his account of Greek city-states in the Persian Wars. Thus, with the narrative of Thermopylae, we are back to the full recognition of Spartan achievements. Precisely in the narrative of Thermoplyae, we find the Spartan counterpart of the praise of Athens just considered. It similarly highlights a moment of choice in the face of the invader and is again squeezed in between a Greek council and an oracle. The Greeks at Thermopylae learn that the Persians are surrounding them; from sacrifices, they receive omens that death is about to overtake them together with the dawn. They meet to decide what to do, and their opinions are divided. Finally, part of the army leaves the pass, scattering “each to his own city” (7.219.2, index of divisiveness). At this point, the narrator steps in to interpret the action in terms of both motive and results. He corroborates a received logos (certainly Spartan) and expands on it. His praise of one party, somewhat as in the Athenian gloss, goes hand in hand with a certain ambivalence toward the others (7.220.1– 2). 111. See Fornara 1981, 155. 112. 7.134– 36. The hypothesis is a recognition of Spartan pragmatism (see Loraux 1977, 113), but it contradicts the Spartan “image” (cf. Thuc. 4.40.1, 4.36.3). Plutarch (De Malign. Herod. 29 ⫽ Mor. 864A– B) comments that at 7.139.3, Herodotus “obviously praises the Athenians not to praise the Athenians but to speak ill of everyone else.”

176

Telling Wonders

They say that it was Leonidas himself who dismissed them, concerned that they should not die; as for himself and the Spartans who were with him, [he thought] it would not have been seemly for them to leave the post they had come to guard to begin with.  I also am very much of this opinion [γνωµην], that Leonidas, after realizing that the allies lacked eagerness and did not want to share the danger, ordered them to depart, whereas for him it was not honorable to leave. By remaining there, he in fact left behind great  glory, and the good fortune of Sparta was not obliterated [κλεος    ε ξηλει ␸ετο]. µεγα ε λει πετο κα ι η Σπα ρτης ευδαιµον ι η ουκ (7.220.1– 2) Herodotus then explains why he attributes to Leonidas the role of saving Sparta, by reporting an oracle that earlier on had predicted either the destruction of the city by the Persians or the death of one of its kings (7.220.3– 4). The interpretation then resumes in similar terms as before. And [I am of the opinion that] it was because he considered this and  wished to establish glory [κλεος] for the Spartiates alone that Leonidas dismissed the allies, rather than them leaving in disorderly fashion because their opinion differed. (7.220.4) Just as the Athenians “became the saviors of Greece,” so Leonidas “left behind great glory, and the good fortune of Sparta was not obliterated”: this result encapsulates the ultimate significance of Leonidas’  choice in the perception of the narrator. In the poetic tradition, κλεος (glory) is almost a technical term for the glory of heroes, especially in death. With a connection that is again traditional in epic, the two occurrences of kleos in this gloss frame the mention of penthos in the oracle’s  prophecy that the city will mourn (πενθησει) the death of its king descended from Heracles.113 Herodotus, in other words, directs the audience to interpret Thermopylae in terms of the epic code of heroic achievement and commemoration, thereby reinforcing elements of that code that are scattered throughout the narrative.114 Previously Herodotus had   113. 7.220.4, lines 3– 4. On κλεος compensating πενθος in several passages of the Iliad, the Odyssey, and the Theogony, see Nagy 1979, 94—117. 114. In this narrative, what starts out as a hoplitic battle becomes a heroic battle (7.223– 25); see Loraux 1977, 116; Dillery 1996, 235– 42, 245– 49. The notice of the fight over Leonidas’ body (7.225.1) evokes the Homeric fight over the body of Patroclus (Il.

Interpretation and Evaluation

177

stepped in with a true gnome (opinion/interpretation) to mark the essence of Athenian merit vis-a-vis ` the rest of Greece. Here, again on the authority of his gnome, he marks the moment at which Sparta, through Leonidas, fulfills its potential as a Greek city-state in the Panhellenic tradition of heroic valor.  The uniqueness of this gnome is confirmed by the use of κλεος and derivatives in the rest of the Histories. The stem κλε- appears in the narrator’s code of celebration within the program of the first sentence, where he  promises not to let human achievement become ακλε α (inglorious). Other  and αποδε  terms we find there (the verbs ι στορεω ι κνυµαι, the nouns εργον and θωµα)  recur repeatedly in the work, both in metanarrative and in the narrative. Words of the κλε- family, by contrast, appear only three other times besides in the Leonidas gloss, exclusively in connection with Spartans, as if the notion of heroic glory represented a standard of measure   “ingloriously,” evaluates appropriate only to them. The adverb ακλε ως, the abortive expedition of Cleomenes to Attica (5.77.1). At Plataea, the Persians complain, with some justification, that the Spartans are not living  up to their renown for valor (κατα κλεος, 9.48.3). After the battle, the Aeginetan Lampon utters what the narrator evaluates as “the most impi and ous speech,” when he praises Pausanias for having achieved κλεος encourages him to finish the work by impaling the corpse of Mardonius (9.78.1– 3). These are all tainted uses, leaving the battle of Thermopylae to shine forth not only as the superior achievement with respect to the “fairest victory” of Plataea (9.64.1) but also as the only perfect fulfillment the Histories have to offer of their author’s promise in the proem to celebrate heroic glory. Leonidas leaves behind a great kleos and causes the eudaimonie  ε ξηλει ␸ετο). (happiness/good fortune) of Sparta not to be blotted out (ουκ His achievement is obviously complementary to Herodotus’ task not to let the events of men become faded with time (ε ξι τηλα) and not to let the great and wonderful deeds of Greeks and barbarians become aklea. Just as Leonidas acquires kleos and preserves the eudaimonie of the city, so Herodotus preserves the kleos of Leonidas.115 The long gloss of identification that introduces Leonidas at the beginning of the Thermopylae narrative 17.274– 87). See How and Wells 1928, 2.230. The heroic ancestry of Leonidas is noted three times, first with full genealogy (7.204, 7.208.1, 7.220.4). See also the Homeric way of indicating time at 7.215 and 7.223.1 (cf. Il. 11.86 and Od. 12.439, on which see Lloyd 1966, 186). 115. See Nagy 1990, 221– 27.

178

Telling Wonders

already marks him as a prime target for Herodotus’ commemoration: he is a descendent of Heracles and at the same time a private citizen who has  become king (7.204– 205.1). He is ο θωµαζοµενος µα λιστα, the highest object of wonder (7.204.1), and consequently belongs to the category of θωµαστα that the narrator has singled out for attention in the first sentence. As Leonidas, therefore, realizes his full potential at Thermopylae as citizen-hero and king– not king, so Herodotus fulfills the potential of his Homeric role as celebrator of deeds through the narration of Leonidas. The praise of Athens I have considered earlier stresses the collective body of Athenian citizens, while the latter magnifies Sparta through its first citizen. The two passages are, however, parallel in other respects, not least because also in the Athenian gloss, the narrator represents himself as personally achieving something exceptional and hard to do, as we have seen: there he performs a gnome that is alethes, absolutely true, both in the epistemic and in the Panhellenic sense. Generalized Glosses of Interpretation Though important and far-ranging, the two glosses I have just examined are among those that confine themselves to discussing events within a specific historical context. In a smaller number of cases, by contrast, the interpretation makes a shift from the past tense of historical narrative to the timeless present and interprets the particular by predicating what is valid in general for all, regardless of time and space. These sentences are our most precious indicators of what Herodotus’ speech act ultimately “means.” Aristotle calls a generalization of this sort a gnome and defines it as “an assertion—not, however, about particulars, such as what kind of person Iphicrates is, but of a general sort, and not about everything (for example, not that the straight is the opposite of the crooked), but about things that involve actions and are to be chosen or avoided with regard to actions.”116 A statement “such as what kind of person Iphicrates is,” to use Aristotle’s example, represents in the terminology of this book a specific gloss of interpretation (or evaluation) like the two I have just discussed. Since all sorts of interpretive statements, specific and general, 116. Arist. Rhet. 2.21.2 (1394a), in the translation by Kennedy (1991). For an analysis of Aristotle’s treatment of the subject, see Lardinois 1995, 7– 13. Generalizations in Homer are also discussed by Richardson (1990, 144– 45).

Interpretation and Evaluation

179

are identified by Herodotus as his gnomai, or opinions, this is the way I will here use the word gnome; I will call Aristotle’s gnomai “gnomic sayings,” “gnomic glosses,” “maxims,” or simply “generalizations.” These are not necessarily equivalent to our proverbs, ready-made, anonymous, and handed down by tradition. Like “the wise words of the Western Apache,” they can be “the property of a particular speaker and created on the spot.”117 Aristotle calls them gnomai because they represent, or purport to represent, the speaker’s opinion. The generalizations that I will consider directly communicate the opinion of the histor Herodotus. Whether or not they reproduce traditional modes of thinking, they are always based on his interpretation of specific events or on a synthesis of his historical experience. Aristotle does not consider all possible generalizations as gnomic. By excluding such a statement as “the straight is the opposite of the crooked,” he attempts to draw the line between moralistic and scientific discourse. Herodotus’ generalizations are “scientific” in the sense that they are always worded as statements of fact, never as instructions on what would be best or what one should do.118 In terms of speech-act theory, they are “representatives,” that is, sentences that make the words fit the world, not the other way around.119 But if we compare, for example, the statement “Diseases among men derive from changes of seasons” (2.77.3) with the statement “Good fortune never stays in the same place” (1.5.4), we see that only the first is purely a description. The second pointedly makes the historical experience of its primary referent, the cities that have risen and fallen in the past, relevant to “us all.” Its gnomic character has to do with the implicit prescription to the listeners not to disregard the “law” or “rule” that the generalization formulates but rather to regulate their behavior according to it. Most generalizations in Herodotus are of this type, statements of fact with the indirect force of advice or warnings. As they provide a reading of the evidence, they also, as Aristotle says, convey a message “about things that involve actions and are to be chosen and avoided with regard to actions.” 117. Lardinois 1995, 5, quoting Basso 1976, 98. For the “coined” character of archaic Greek gnomai (in the Aristotelian sense), in which traditional recurrent themes are constantly reworded and reshaped, see Lardinois 1995, 22– 26. 118. The same is true for some of Aristotle’s own examples of gnomic sayings, as Lardinois observes (1995, 11). See Rhet. 2.21.2 (1394a– 1395b). 119. At least, this is their primary “illocutionary point.” I am relying on the taxonomy of Searle (1976). See, in the present book, my introduction, n. 41 and corresponding text.

180

Telling Wonders

Through these interventions, more conspicuously than through narrative alone, the scientific researcher appropriates in relation to his audience the stance of the sage of archaic tradition vis-a-vis ` the citizens of the polis. In particular, the narrator comes to resemble somewhat the wise advisers who populate his narratives. But wise or not, most speakers in the Histories generalize relentlessly, on all sorts of topics.120 Whether their gnomic sayings corroborate or contradict those of the narrator, they inevitably— as far as the general message of the work is concerned—tend to complicate the issues: alongside with events narrated, they are all parts of the factual evidence the Histories presents. At the same time, the utterances of speakers are never automatically reflections of Herodotus’ interpretation of that evidence’s meaning. The overt deliberative rhetoric in which Herodotus’ speakers mostly engage affects the form of their maxims. Occasionally characters generalize in a string of gnomic sayings, use maxims in the mode of “should,” posit general truths as the basic assumption or as supporting evidence for their argument, phrase them in the grammatical second person (7.50.1), or deliver them in hortatory form to their listener (µαθε  . . . , 1.207.2, 7.49.3). The narrator, for whom the deliberative aim is ως hidden behind the judicial and scientific/representative modes, never does any of these things. He tends to be a brief, spare, and cautious generalizer of what certain particulars reveal to be the case. The abundance of maxims in the reported speeches of the Histories has caused some confusion between what the narrator says and what his characters say. Moreover, generalizations tend to be strictly connected to their respective narrative contexts. These factors have led some critics to devalue Herodotus’ generalizations as merely the self-contradictory and strictly occasional by-product of the storyteller’s impulse to narrate.121 The impulse to narrate, however, is for Herodotus the impulse to accumulate evidence. A narrative may already function as the paradigmatic account of a particular action that implicitly conveys certain truths applicable to the recipient of the narrative.122 But if the proliferation of narratives and utterances creates a picture that is crowded, nuanced, diverse, 120. The seventy-five maxims of the list compiled by Lang (1984, 58– 66 and notes) occur mostly in speeches. See also Shapiro 2000. 121. See, e.g., Gould 1989, 81. 122. See Lardinois 1995, 110– 16, on the functional equivalence between gnomic sayings and paradigmatic narrative (ainoi) in the utterances of Homeric characters.

Interpretation and Evaluation

181

hard to know on the factual level, and unstable from the point of view of how and why things happen, the narrator’s generalizations highlight certain moments when the meaning of particulars appears especially clear to him. Herodotus does not pretend with his audience that the evidence always goes his way or only one way; this is an important aspect of his persona as inquirer and of the interrogative nature of the text. But from the multiplicity of experience, a moral can be derived, and occasionally the narrator is so bold as to express it. The moral is then in turn both a tool for capturing the diachronic and synchronic patterns of human experience and a guide to human action. Few in number and remarkably consistent, Herodotus’ maxims identify some of the major ethical concerns of the historie. Because they are fundamental reference points for understanding his message, we have already encountered some of them in preceding discussions. I will now survey them as a group, subdividing them primarily according to what they generalize about, and I will use them as a guide for examining some specific glosses of interpretation that are thematically related. We can isolate a category of generalizations of a rather philosophical and metaphysical sort, concerning three major issues: the instability of human fortune; the nature of divine action in the visible world, especially divine retribution; and the phenomenon of divine communication. Within another, political category, we first of all find a subgroup that wants to be considered from the point of view of the topic to which the statements are attached, because here the primary referent and the representative/ evaluative character of the maxims overshadow other aspects. This subgroup consists of two generalizations on government that are inspired by the historical case of Athens. A second subgroup of political generalization concerns war. A third perhaps holds the ideological key to the paucity of generalizations in Herodotus: it has to do with the subjectivity and relativity of opinions. The Instability of Happiness The first maxim in the work is a preliminary generalization that does not stem from the narrative but rather motivates the narrator’s program with regard to the contents and structure of his logos. It appears here underlined, quoted in its context. The self-referential metanarrative appears in bold face.

182

Telling Wonders

. . . after placing in relief the one I myself know to have been the first to initiate the wrongdoing against the Greeks, I will proceed with my logos, going through the cities of men, small and great alike. For those that were once great have for the most part become small, and those that were great in my time were previously small. Therefore, since I know [ε πιστα µενος] that human good fortune  ανθρωπη   never stays in the same place [τ ην ι ην . . . ευδαιµον ι ην   ω  ουδαµ α ε ν τωυτ &  µενουσαν], I shall mention them both alike. (1.5.3– 4) The first part of the program establishes that the first narrative of the logos will be about Croesus, whom the highly interpretive gloss identifies as “the first . . . to initiate the wrongdoing against the Greeks.” He is, in other words, that original aitios (guilty one) for the East-West conflict about whom the historie had started out to inquire (α ι τι ην, first sentence; α ι τ ι ους, 1.1). The second programmatic movement (“I will proceed . . .”) identifies the logos as a whole, describing it as a sort of odyssey (cf. Od. 1.3) in which the wanderer-narrator can choose where he wants to go: he will narratively visit all sorts of astea because most of them have an interesting history of becoming large or becoming small.123 So there are two competing topics in the logos. One centers around aggressions against the Greeks; the other is represented by the histories of “cities,” how they have grown and how they have declined. Implicitly, Croesus is part of both: he is the first aggressor of the Greeks, and the narrative about him as it turns out, relates his becoming great and becoming small. The gnomic saying itself is formulated by the narrator on his own authority (ε πισταµενος), as is typical of the generalizing element in the metanarrative of the Histories.124 The self-referential signs marking Herodotus’ gnomai contribute to displaying the researcher who does not take traditional wisdom for granted but derives a teaching from a careful evaluation of the evidence he has obtained. Here the maxim “Human good fortune never stays in the same place,” though already a result of Herodotus’ historie (research) is preliminary to the apodexis (presentation). Its purpose is to broaden the scope of the historical synthesis about 123. On the metaphor of the logos as a journey, see chap. 1, “Self-Referential Glosses.” 124. This is infrequent in other authors (see Lardinois 1995, 157– 61, 61– 63, 195– 225, 231– 72) and in Herodotus’ reported speeches; exceptions occur, for example, at 1.32.1 (see n. 127 in the present chapter), 3.40.2, and 7.18.2.

Interpretation and Evaluation

183

what has happened to cities, projecting it beyond the boundaries of the historical past. Large cities have become small, and small cities have become great; history has been so consistent in this respect that Herodotus can surmise that this sort of thing will continue to happen, not only to cities, but to nations and states as well as individuals.125 The transition to the present tense signals that the logos of Herodotus, although it will narrate things about the past, is relevant to listeners at the moment of the narration and that it communicates a universal historical norm about their future. Croesus will call this norm of instability the “cycle of human things” (1.207.2), but the narrator’s formulation bears no suggestion of cyclic regularity, no determination of time, and no declaration of historical necessity. Under which circumstances, then, and for which causes does eudaimonie migrate from one place to another? Divine Retribution In the course of the logos, the narrator eventually offers two generalized interpretations that answer this question (2.120.5, 4.205). Both passages connect the decline of cities and loss of eudaimonie mentioned at 1.5.3 with human injustice punished by the divine. The discontinuity between the immediate and long-range topics in the global program I have just considered are thereby to some extent filled in and resolved. Herodotus’ historie inquires about guilt (starting with Croesus) and about rise and fall (also starting with Croesus) because it inquires about the extent to which a rise in power entails guilt and a fall is caused by divine punishment.126 Before examining the crucial statements in which the narrator generalizes on this causality, I will briefly examine how he interacts with interpretations of other speakers on the role of the divine in the instability of human fortune. In the narrative about Croesus, the text engages with the evidence through a process of verification that betrays its own uncertainties. Two oracular utterances connect the demise of Croesus with the tisis (vengeance) for Gyges’ overthrow of the Heraclids and attribute to Croesus his 125. Eudaimonie is a political term (see, e.g., 7.220.2), but it also denotes personal happiness. In this case, as usual, the historical/political and the biographical/personal codes are strictly conjoined. 126. On the guiltiness and loss of eudaimonie of Croesus, see Croesus’ self-exculpation at 1.87.3, contradicted at 1.91.4. See Nagy 1990, 240– 42. On the principle of divine retribution in Herodotus see Fornara 1990 and, most recently, Harrison 2000, 102– 21.

184

Telling Wonders

own share of guilt (1.13.2, 1.91). These constitute a privileged sort of evidence. The narrator, for his part, especially seeks signs of Croesus’ personal culpability. We see it in the introduction to the narrative of Croesus’ misfortunes in the descending phase of his reign, where the verb ε ι κα ζω is a mark of the most speculative degree of opinion: “After Solon  left, great anger from the god overtook Croesus because, one imagines [ως   ε ικα σαι], he believed himself to be most blessed among men” (1.34.1). The interpretive gloss just quoted adapts the description of human experience just provided by Solon to the story of Croesus’ downfall in a moralistic way. This might seem strange. Though the Athenian sage has confirmed the narrator’s generalization on the instability of good fortune, his maxims appear to emphasize either the random aspect of the process of reversal or the amorality of divine envy, which targets not wrongdoers but especially those who have everything.127 But Solon’s words are cryptic and deliver a mixed message. In his opposition between the rich man, who is likely to experience disaster, and the fortunate man, who is not (1.32.6), the disaster is called ate, a term that traditionally denotes both the misfortune and the moral folly of one who has brought the misfortune on oneself as a result of surfeit and transgression.128 Similarly, ε πιθυµι η refers to the desire that the rich man is in a better position than others to fulfill (and from which the merely fortunate man is exempt): the word suggests an irrational impulse leading to self-detrimental action, which in the Histories is typical of monarchical rulers.129 An allusion to the personal guilt connected with the opportunities that power affords therefore intrudes surreptitiously in Solon’s interpretation of the instability of fortune. The narrator’s interpretation at 1.34.1 takes the evidence of Solon’s speech into account by attributing the cause of Croesus’ loss of eudai monie to a superior force: ε κ θεου νεµεσις [anger from god]. The word nemesis (rightful indignation) is both emotional and moralistic, thereby striking a compromise between Solon’s ethically problematic phthonos (envy) and the judicial terms timorie and tisis, which Herodotus uses  µε το θειον παν  ␸θονερον  τε  ε ον 127. See the words of Solon at 1.32.1: ε πιστα µενον  κα ι ταραχωδες [“since I know that the divine is utterly invidious and troublesome”]. Plutarch (De Malign. Herod. 15 ⫽ Mor. 857F– 858A) finds this statement insulting to the  ε στι ανθρωπος  gods. Cf. 1.32.4: παν συµ␸ορη [“man is entirely a thing of chance”] and  the reference to τυχη [chance] at 1.32.5. 128. Cf. discussion of 8.77 under “Divine Communication” later in this chapter. 129. See chap. 2, n. 232 and corresponding text.

Interpretation and Evaluation

185

elsewhere to denote retribution.130 Thus, Croesus, in Herodotus’ interpretation, is neither the embodiment of mysterious human chance (Solon’s συµ␸ορη at 1.32.4) nor the victim of a divine power that prevents men from having too many good things. In the world of nature, the divinity that prunes excessive growth is a rational principle of balance: “divine providence is, as one would expect, wise.”131 But in the sphere of human history, Herodotus looks for evidence of a divine participation that makes sense also in ethical terms. Solon and other characters who are speaking to kings or are themselves kings establish a connection between calamity and greatness rather than between calamity and wrongdoing.132 The last speaker who refers to divine envy, however, will place it in an ethical context. The Greek victory against the Persians, Themistocles says, was the work of “the gods  and heroes, who begrudged [literally “envied,” ε ␸θονησαν] that one man reign over both Asia and Europe, a man who was impious and unbearably reckless, who treated in the same way temples and private buildings, burning and overthrowing the statues of the gods; a man who even flogged the sea and lay fetters on it” (8.109.3). With respect to the historical case of Xerxes, at any rate, the divinity emerges as an ethical and rational force. Divine “envy” turns out to be the response of the divine to culpable human attempts to rival, antagonize, and replace it. Herodotus first formulates a general rule concerning divine justice in relation to the loss of human eudaimonie in his interpretation of what Greek tradition regards as the paradigm of all fallen cities. In Egypt, where issues concerning the history of humankind and the most remote past of the Greeks themselves attain special clarity, Herodotus has also learned the real meaning of the destruction of Troy. By disclosing that Helen spent the entire duration of the Trojan War in Egypt, the Egyptian priests in Memphis have suddenly provided for the fall of Troy an explanation that makes sense in both historical and ethico-theological terms (2.113– 20). The interpretive gloss by which the narrator corroborates the logos argues that if Helen had reached Troy together with Alexander, 130. See Macan 1895, 1: cxiv. Contra Chiasson 1986. On the root nemes- used to denote a reaction to a wrongdoing, see especially Il. 24.53 and the survey in Giraudeau 1984, 67– 73. !  ε στι, ε ουσα  σο␸ η.  See especially 131. 3.108.2: του θει ου η προνοι η, ωσπερ κα ι ο ι κος Immerwahr 1966, 312– 13. 132. Amasis does so at 3.40.2 (cf. 3.43.1), Artabanus at 7.46.3 and 7.10ε. So does the epinician poet in his praise of prominent individuals and tyrants. See McGlew 1993, 41; Nagy 1990, 274– 313. But Artabanus reveals his moralism at 7.18.2.

186

Telling Wonders

as Homer says she did, the Trojans would have ended up returning her to the Greeks, and Troy would not have been destroyed (2.120). Thus, if the chance arrival of Helen in Egypt was the single factor that deprived the Trojans of the means of reparation and therefore survival, it was not a chance arrival at all. The simplicity of this discovery is so compelling that it proves to the narrator that for great injustices, great are also the punishments from the gods των µεγα λαι ε ι σ ι κα ι α ι τιµωρι αι παρα  µεγα λων αδικηµα  [των  θεων].  των (2.120.5) We should note, as in the case of the maxim on the instability of human happiness, all the subsidiary issues that this generalization still leaves up in the air. If the divine always punishes human injustice, is the loss of human eudaimonie always the manifestation of divine punishment? Do cities always “become small” because of human injustice? Herodotus’ evidence in the narrative does not support the notion that divine action can be intelligible to men in every case.133 The narrator even acknowledges the often unexplained mixture of fortune and misfortune that is inherent to the human state.134 But the general rule that grave offenses will meet with great punishments represents Herodotus’ minimal interpretation of what the gods’ action at Troy intended to make manifest to men, an interpretation that Herodotus, in his turn, under(κατα␸ανες)  takes to make manifest to his audience (ε γω γνωµην απο␸α ι νοµαι). An especially important problem remains unresolved on the paradigmatic battlefield of Troy: the ethical status of the human avenger. Because the war the Greeks waged against Troy represented an instrument of divine retribution against Trojan wrongdoings, was it a just action? Menelaus’ “impious” deed in Egypt, clearly evoking the human sacrifice at Aulis in one of the Greek traditions, gives a hint that not all may be morally right with the aggressor against Troy.135 Herodotus’ second gener133. See especially the logos of Mycerinus (2.129, 133). 134. See Harrison 2000, 112– 13. Thus, though the ultimate end of Cambyses does verify the rule of divine retribution (see 3.64.3), the cause of his madness is uncertain. See 3.33 (analyzed in Munson 1991, 51– 53), where the generalization “many are the evils that are wont to happen to men” stands to the causes of misfortune, as the generalization at 5.9.3, “in the vast length of time, everything is possible,” stands to the vicissitudes of history. 135. See 2.119.2– 3 and discussion under “Revising Greek Traditions” earlier in this chapter. On Greek responsibility for the war against Troy, see also 1.4.1. Outside of Herodotus, see especially Aesch. Agam. 40– 72, 104– 39, 183– 226. See also the discussions in Nussbaum 1986 (32– 38) and Williams 1993 (132– 36).

Interpretation and Evaluation

187

alization about divine retribution precisely evaluates the justice of human revenge in the eyes of the god. This generalization occurs in the narrative about the antecedents of the Persian expedition against Cyrene during the reign of Darius. The sixth king of Cyrene, Arcesilaus, is killed in Barca. His mother, Pheretime, eager to make war against the city, obtains the help of the Persians, who use this revenge as a false pretext for the conquering of Libya (4.165– 67). Once they capture Barca, they enslave most of the citizens. Pheretime impales around the bastions of the city those most responsible for her son’s death and cuts off the breasts of their women, planting them all around the walls (4.202– 203.1). Like the excess of her revenge, the horror of Pheretime’s subsequent end gives her story the clarity of limiting cases. But Pheretime did not end her life well either. For as soon as, having obtained her revenge from the Barcaeans, she returned to Egypt from Libya, she died a bad death: still alive, she started breeding worms  out of herself, because, truly then [αρα], vengeful acts that are excessively violent are for the gods a cause of resentment/envy against men  θεων   ε πι ␸θονοι [ανθρ ωποισι α ι λι ην ι σχυρα ι τιµωρι αι προς γινονται]. Of such a kind and so great indeed was the revenge of Pheretime, the daughter of Battus, against the Barcaeans. (4.205)  The particle αρα marks the maxim as an inference based on the events narrated. As in the Trojan logos, these involve the “becoming small” of a city—in fact, its fall as a result of a siege in turn motivated by revenge. Here, however, the narrator’s interpretation focuses on the culpability of the avenger. In each case, the generalization drawn from a specific historical event involves a shift to the grammatical plural. Just as great adikiai (injustices) attract great timoriai (vengeful actions) from the gods—and these divine timoriai may be carried out by human agents—so when the human timoriai are excessive they become injustices, which will in turn deserve divine punishment because they are ε πι ␸θονοι (cause for resentment/envy) to the gods. Only in this passage does the narrator participate in the theological code of his speakers by using the phthon-stem in connection with the calamities that god sends to man. Like Themistocles in the speech we have already seen (8.109.3), he here joins the Solonian notion of divine envy of human power to that of divine anger against wrongdoing. While the timoriai of the gods are never “excessively violent” (as devastating as they may seem), the

188

Telling Wonders

excessively violent timoriai of men, here represented by Pheretime’s mutilations, constitute “monarchic” usurpations of a divine prerogative— hence invoking the phthonos of the gods.136 Punishment from the gods is a fundamental historical cause of human reversal in the Histories and is the only historical cause at any level that the narrator proclaims in general terms. This goes a long way toward explaining why Herodotus’ inquiry into human causes is largely concerned with assessing the aitiai (guilts or accusations that become causes of actions for the accuser) and who is aitios (responsible, guilty).137 From beginning to end, the logos provides evidence for the recurring role of human vengeance as the only possible motive that has a theoretical claim of legitimizing a range of aggressive actions (dispossessing, murdering, making war on others) that are otherwise adika, “unjust.”138 The narrative of the destruction of Barca, where vengeance is a pretext for the Persians (4.167.3) and a criminal impulse for Pheretime, encapsulates the criteria by which Herodotus’ logos undermines such a claim. Many timoriai of the Histories, with or without ulterior motives, are equivalent to adikiai, and as such they are vulnerable to divine punishment. From ancient history and foreign settings, Herodotus’ investigation of the workings of divine vengeance eventually reaches the context of the struggles between and within the cities of Greece. In an astounding compositional move, Herodotus gives the Spartan king Leotychides, himself a paradigm of crime and punishment, the role of asserting the ineluctability of divine retribution through his own narrative: this is the story of Glaucus, a man who presumed to obtain Delphic permission to commit perjury and whose house was subsequently obliterated.139 Much as do Herodotus’ own narratives, Leotichides’ story functions at once as an ainos and as the presentation of a piece of historical evidence. It is ad136. The monarchic connotation of ε πι ␸θονος suits the present context also because Herodotus’ history of Cyrene illustrates the way in which a royal dynasty of founders becomes transformed into a tyrannical dynasty (see especially 4.161.3– 4.164). See McGlew 1993, 172– 73. 137. The fundamental study of the various aspects of causality in Herodotus is still Immerwahr 1956. See also Pagel 1927; de Romilly 1971; Lateiner 1989, 189– 210. 138. Cf., e.g., Croesus’ aitiai against cities of Ionia (1.26.3); the role of revenge in Croesus’ decision to attack Cyrus (1.46.1, 1.71.1, 1.73, 1.75.1), in Darius’ Scythian expedition (4.1.1; cf. 4.118.3– 5, 4.119.2– 4), and in Xerxes’ Greek expedition (7.5– 9, 7.11.2– 4). See Immerwahr 1956, 253, 254– 64. 139. See 6.86α– δ. On Leotychides himself as victim of divine tisis, see 6.72.1 (where the negation reproduces the pattern of the gloss about Pheretime at 4.205).

Interpretation and Evaluation

189

dressed to the Athenians who, just as Glaucus tried to keep a deposit of money, now refuse to return certain Aeginetan hostages. Its context is the account of the hostilities between Aegina and Athens, which Herodotus pursues piecemeal, starting from the time when the Aeginetans, exalted by their prosperity and mindful of their “ancient enmity” against Athens, help the Thebans to avenge themselves against the Athenians (see 5.79.1: τει σασθαι, τι σις) by initiating a “heraldless war.” The Athenians then prepare their timorie against Aegina despite an oracle that advised them to wait for thirty years to attack; at that moment, however, the threatened Spartan invasion of Attica intervenes to suspend their plan (5.89.2– 90.1). As a result of this conflict with Athens, Aegina medizes, which induces the Spartans to seize Aeginetan hostages and confine them in Athens (6.49– 50, 6.73.1– 2). Leotychides’ parable about Glaucus has no effect: the Athenians keep the hostages, with the excuse (προ␸α σιας) that they do not “consider  to return to only one Spartan king what has been enjust” [δικαιουν] trusted to them by two (6.86). The Aeginetans feel injured by the Athenian refusal, even though, for their part, they still have not “paid the των)   penalty” [δουναι δι κας] for their own previous injustices (αδικηµα against the Athenians when they “acted with hubris” to please the Thebans (6.87: the reference is to the heraldless war). So the Aeginetans retaliate by capturing an Athenian theoric ship during a festival at Sounion (6.87); the Athenians in turn respond by conspiring with a cer tain Nicodromus, an Aeginetan who had a grudge (µεµ␸οµενος) against his fellow citizens and who now agrees to betray the city to the Athenians. The plan does not succeed, but the Athenians eventually inflict a major defeat by sea on the Aeginetans (6.88– 90, 6.92.1). In the whole intermittent narrative that centers around the hostility between Aegina and Athens (including a flashback at 5.82– 88 about the beginning of their “ancient enmity,” a unified code emphasizes a tangle of abuses and retaliations. The objective ethical status of all these timoriai and tiseis, which in turn elicit retaliation, remains remarkably ambivalent.140 Mutual conflicts pitting city against city and citizen against citizen  140. This is confirmed by the insight of Lateiner (1980) that the phrase δουναι δι κας (which appears at 6.87 in reference to the Aeginetan failure to pay the penalty to the Athenians) always reflects the claims of various parties, not the position of the narrator. For timoria words emphasizing “the key motive for heroic action” in fifth-century Athenian political discourse, see Rose 1995, 77– 79, on the Ajax and on Pericles’ statements in Thuc. 2.42.

190

Telling Wonders

leave no one free of adikemata and impiety, even though everyone acts on the basis of alleged criteria of justice, justified vengeance, procedural issues, and the like.141 In the midst of all this, the parable of Glaucus reintroduces the notion that in contrast with men, the divine operates with surefooted simplicity and unerring autonomy, cuts across legalistic pretenses, identifies each wrongful action, and correctly assesses intentions. Nothing is said about the consequences of the Athenians’ violation of a binding agreement parallel to that of Glaucus (i.e., their failure to return the hostages). But given the context, that omission does not necessarily imply that they have avoided divine punishment, nor does having the message on divine justice conveyed by an individual who will himself be a historical exemplum for it impugn its validity. In the same narrative, in fact, Herodotus identifies direct evidence for divine retribution. In Aegina, the wealthy faction executed the democrats led by Nicodromus; they even cut off the hands of one of them who was hanging on to the temple of Demeter Lawgiver. Herodotus interprets,  “From this action, even a curse [αγος] resulted for them, from which they were not able to free themselves by means of sacrifices, although they kept trying, but rather they were banished from the island before the goddess would become appeased toward them” (6.91.1). Through the connection this gloss imposes on mutually distant and causally unrelated events, Aegina is the most recent instance presented in the Histories in which a city’s loss of eudaimonie is a historical process related to guilt. Precisely the word eudaimonie is used to denote the prosperity that led the people of Aegina to initiate their heraldless war (5.81.2; the war is called an act of hubris at 6.87). The Aeginetans’ history of violations, which begins with their theft of the statues of Damias and Auxesias from Epidaurus (5.83.2), culminates in their execution of the suppliant and ends, well within the recent memory of Herodotus’ audience, when Aegina ceases to exist as a polis.142 Divine justice has been as radical in this case as for Glaucus and Troy.143 Herodotus’ interpretation of the fall of Aegina as the result of divine 141. See Immerwahr 1966, 211– 15; Lachenaud 1978, 51– 55. 142. 431 b.c. See Thuc. 2.27.1, where the statement that “the Athenians expelled from the island men, women, and children, reproaching them for being not the least responsible [α ι τι ους] toward themselves for the outbreak of the war” shows that claims of justified vengeance have not gone out of style. 143. Herodotus’ unmarked narrative, through the oracle at 5.89.2, also connects the Aeginetans’ “injustice,” represented by the heraldless war, to their loss of independence in 457 b.c. (see Thuc. 1.108.4).

Interpretation and Evaluation

191

punishment for its guilt against the gods may well stem from contemporary Athenian propaganda. Even if this is the case, those scholars who have maintained that therefore Herodotus justifies the Athenian action against the island are far off the mark.144 The ethical status of those who serve as instruments of divine tisis is for Herodotus a separate issue that he tends not to explore when he presents evidence for the tisis. The gloss at 6.91.1 does not even mention the Athenians as the agents of the destruction of Aegina. But the entire narrative of the hostility between Athens and Aegina in the Histories accumulates instances of Athenian culpability in other respects and attributes to both parties largely symmetrical guilts.145 Herodotus’ Aegina is both the double of Athens in the same narrative and the antecedent of the Athens of later times.146 The city whose cycle of rise and fall has just been completed by the time of narration does not provide a comforting analogue for one that seems bound on a similar course. With the parable of Glaucus, explicitly introduced to underline the Athenians’ bad-faith refusal to return the hostages, the issue of Athenian vulnerability in the face of the divine first comes to the surface of the text. Both Athens and Aegina survive to contribute to the subsequent resistance against Persia; by some mysterious compensation, their mutual hostility will produce the fleet that will save all the Greeks (7.144.2). But further on in the logos, Herodotus alludes to the possible consequences of Athenian injustice when he wonders what happened to Athens in retribution for throwing into the well the heralds whom Darius had sent to demand submission. What undesirable thing happened to the Athenians as a result of this action, I cannot tell, except that their land and city were 144. See How and Wells 1928, 2:100. 145. For signs of Athenian culpability in the religious sphere as well as in the secular sphere in the account of early hostilities with Aegina, see 5.85– 86, 5.87.1– 2, 5.89.3, 5.86.1, 5.87. 146. See the allusions to contemporary circumstances within the narrative of their early hostilities (5.82– 88), discussed by Figuera (1985, 66). At 5.83.1– 2, the Aeginetans are described as θαλασσακρα τορες (masters of the sea), a term that normally refers to fifthcentury Athenian sea power ([Xen. Ath. Pol. 2.2.1, 14; see Figuera 1985, 91– 92 and nn. 28– 30). The analogy between Athens and Aegina is reinforced by the way in which each of them mistakenly relies on the support of the Aeacidae, who stand for divine protection for the purposes of defense (8.83.2), not of waging or helping others to wage aggressive wars (5.79– 81; cf. 5.89).

192

Telling Wonders

devastated, but I do not think that this happened for this cause/guilt [α ι τι η]. (7.133.2) The question is particularly cogent because the narrator asks it while again in the process of verifying the rule of divine tisis. To the guilt of the Athenians corresponds an analogous guilt of the Spartans, who threw Darius’ heralds into a pit (7.131, 133.1). But while the punishment of the Athenians remains undetermined, that which befell the Spartans on account of the anger of Talthybius, the heroic patron of heralds, is a matter of record. Just as the Spartans, through Sperthias and Boulis, offered compensation to the Persian king, whose heralds were sacrilegiously killed, so the gloss just quoted tentatively mentions the Persian king’s devastation of Athens as a punishment for the Athenian crime. Herodotus immediately discards the interpretation as unsatisfactory, however, just as he shows how the artificial reparation of the Spartans succeeded only in part. The divine works at its own pace, and in both cases, it postpones tisis to a later time. When Sperthias and Boulis offered their lives to Xerxes, the king declined to break in his turn the “laws of all men” and to free the Spartans from the aitie (guilt) by making himself aitios (guilty) through vengeance (7.136.2). Divine anger, however, “flared up again much later during the war between Athenians and Peloponnesians, as the Spartans say.” The Spartans make the causal connection between the wrath of Talthybius and the later occurrence (7.137.1); but as he introduces the proleptic report, Herodotus corroborates their interpretation in his own voice. Under the circumstances especially, this event appears to me very much the work of the divine [µοι . . . θειοτατον ␸αι νεται]. The fact that the anger of Talthybius was discharged against heralds and did not abate before it found its expression—it was justice that brought  ε␸ερε]; but the fact that it should fall on this about [το δι καιον ουτω the children of these men who had gone to the king on account of [Talthybius’] anger, Nicolas the son of Sperthias and Aneristus the son of Boulis . . . well, it is evident to me that this was the work of the  µοι οτι  θειον ε γενετο divine [δηλον ων το πρηγµα]. (7.137.1– 2) This gloss, which brings about the latest datable reference in the Histories, also constitutes the most spectacular confirmation of the maxim

Interpretation and Evaluation

193

attached to the Egyptian logos about Troy that the gods punish human injustices (2.120.5). Opinion in that passage (γνωµη) is here replaced by glosses of evidence (“this event appears to me,” “it is evident to me”). Herodotus’ inquiry on the morality of the divine has found another landmark that precludes randomness (“it was justice that brought this about”). As in the case of Helen’s stay in Egypt, the lack of factual connection between human planning and the achievement of ethically intelligible results proves the effect of a higher causality that cooperates with and exploits, for autonomous ends, the political motives of men and the natural course of events. These events are finally recounted: “many years after the expedition of the king,” Peloponnesian ambassadors bound “for Asia,” including the son of Sperthias and the son of Boulis, were captured at the Hellespont and put to death in Athens (7.137.3– 138.1). According to Thucydides (2.67), this happened in the second year of the Peloponnesian War, and the ambassadors in question were on their way to seek money and military cooperation from the Persian king. The narrative of Herodotus avoids emphasizing the painful discrepancy between the mission of Sperthias and Boulis to Persia and that of their sons fifty years later. Yet the bare mention of this second set of Spartan envoys to Persia would have been enough to bring to the consciousness of the listeners the changed circumstances of their own time.147 More striking is the uncomfortable asymmetry the narrative creates between Sparta and Athens in respect to their parallel crimes. Not only does the Athenian killing of the Persian heralds remain unrequited in the presence of the Spartan evidence for the inevitability of divine retribution, but the very same action that frees the Spartans from their debt serves also to compound the guilt of Athens.148 The depopulation of Aegina in 431 b.c. and the killing of the Peloponnesian heralds in 430 owe their place in Herodotus’ history to their value as the most recent demonstrations of the rule of divine tisis.149 In both cases, the human agents of tisis are the Athenians; but in the second case, the Athenians also figure as the perpetrators of an action that is analogous to the crime for which the gods hold the Spartans accountable. By 147. Would they have known that the Athenians threw the Peloponnesian envoys into a pit [ε ς ␸α ραγγα] and that their motive was revenge (see Thuc. 2.67.4)? 148. This point is made by Georges (1994, 161– 63). 149. See also 7.233, which again suggests the idea of punishment and further guilt, this time of the Thebans. See Cobet 1971, 71.

194

Telling Wonders

430 a series of setbacks, some entirely unforeseeable, made the future of Athens more uncertain than ever before.150 Herodotus’ silence on the topics of the Peloponnesian invasions of Attica and of the outbreak of the plague is in keeping with his usual reticence. But his accumulation of historical evidence for the moral action of the gods leads him to identify Athens as an anticipated, not yet consummated paradigm at the end of a series of complete and closed cases starting with Troy. What will happen to Athens—what has perhaps already started to happen—for her past and more recent aitiai, symbolized by the old and the new heralds? This question remains implicit and suspended. At the inception of the account of Xerxes’ invasion that immediately follows, moreover, Herodotus deliberately sets it aside and counterbalances his reference to the guilt of Athens with his famous “true opinion,” a forceful assessment of Athens’ merit in the Persian War (7.139). Divine Communication The participation of the divine is constant. Only exceptionally, however, does it fail to coincide with the natural course of things, and only exceptionally can men detect it. In the affair of the heralds, the particular coincidence of events was “evidently divine” (7.137.2). In the moments before the battle of Mycale, the combination of the epiphany of a heralds’ staff on the beach and a rumor in the Greek camp about successful fighting at Plataea is for the narrator evidence of the divine nature of things in general.151 Evident from many signs is the divinity of events [δηλα δ ε πολλοι σι  πρηγµα των], if also on that occasion τεκµηρι οισ ι ε στι τα θεια των it happened that . . . (9.100.2) The miracle of Mycale stands at the intersection between divine communication and divine interference. By affecting human events, the gods, 150. See Thuc. 7.28.3 and, for the strangeness of the plague, especially Thuc. 2.50, 2.51.1. 151. See 9.101.2. Another coincidence is that the battles of Plataea and Mycale were both fought in the neighborhood of a precinct of Eleusinian Demeter (see 9.101.1). On other occasions , even one coincidence occurring by itself is noted (see 7.166, 8.15.1; cf. 6.116). See Immerwahr 1966, 254 and n. 52, 258. On divine communication, see Harrison 2000, 122– 57.

Interpretation and Evaluation

195

as in the destruction of Troy, make their criteria of regulation manifest to 2.120.5). Here they affect an event humankind in general (κατα␸ανες, through communication, because their revealing the success at Plataea increases the energy of the Greek army at Mycale (9.100.2). In most cases, however, divine communication does not entail effective action. Herodotus’ exemplary case of divine communication tout court is what happened to the Chians at the end of the Ionian revolt. In the battle of Lade, where they performed splendid deeds of valor, the Chians suffered the greatest number of losses (6.15– 16), and in the aftermath, they became the victims of a seemingly gratuitous attack by Histiaeus (6.26). Factually unrelated but symbolically analogous natural disasters preceded this series of political and military misfortunes that “brought the city to its knees.”152 To Herodotus, this is proof of a mysterious link (κως [somehow], 6.27.1; cf. 3.106.1, 3.108.2) and of a general rule of divine communication. There are wont to be signs somehow, when great evils are about to  αν δε κως προσηµαι νειν, ευτ  happen to a city or to a people [␸ιλ εει η εθνει εσεσθαι]; for before these events, η µεγα λα κακα η πολι µελλ µεγα λα] came also to the Chians. (6.27.1) great signs [σηµηια It becomes evident that the gods send great semeia (signs) in the imminence of great kaka (misfortunes), just as they punish great adikiai (injustices) by means of great timoriai (punishments). Between these two areas of divine intervention—retribution or any type of effective action, on the one hand, and prediction, on the other—it is not possible to establish a necessary or clear correlation. The text does not direct us to interpret Chian misfortunes as divine punishments. Their higher causes are perhaps imbricated with other guilts from other agents of the Ionian revolt, but they remain, and Herodotus allows them to remain, unknown.153 Here, empirical historical facts only allow him to perceive the divine as the sender of signs. To the broader sphere of prophecies, Herodotus seems intent on 152. See 6.27.2 for two separate incidents that cause the death of Chian children. The analogical link between these event signifiers and the event signifieds is that the victims here represent the future of the city, which will be more seriously compromised as a result of the war. See Immerwahr 1954, 16– 17. 153. See the oracle quoted at 6.19.2 with an unexplained reference to “evil deeds” of  ε πιµηχανε Miletus [Μι λητε, κακων εργων].

196

Telling Wonders

applying the method he attributes to the Egyptians, who write down each teras (prodigy) and what it turns out to have predicted; in this way, if a similar sign occurs, they will know what it means (2.82.2). Capable of performing a great variety of speech acts, sometimes within the same utterance, dreams and oracles are more likely than are event signifiers to indicate not only that something will happen but also, retrospectively at least, why, in terms of divine criteria of regulation, something was going to happen.154 Thus, a dream vision, not a human speaker, formulates the only clear-cut maxim concerning divine punishment in a voice other than that of the narrator. This provides some of Herodotus’ best evidence that divine verbal communication with men can be especially useful for the interpretation of history. On the eve of his murder at the hands of Harmodius and Aristogeiton, Hipparchus saw in his sleep a man of imposing and splendid appearance who “uttered the following riddling lines” [α ι ν ι σσεται τα δε τα επεα]: “Bear, o lion, unbearable things, suffering with forbearing heart. / No one among men who commits injustice will avoid vengeance  ι ς ανθρ    τι σιν ουκ  αποτ  [ουδε ωπων αδικ ων ι σει]” (5.56.1). These words exemplify both the value of verbal signs from the gods and the problems they create for the inquirer. An exhortation in the first line implies a prediction or warning, reinforced by a gnomic saying. What subsequently happens to Hipparchus renders the prediction, at least, entirely clear, ε ναργεστα την (5.55), turning the vision into yet another item of proof of the reliability of divine communication. But aspects of the dream utterance, first and foremost the specific applicability of the maxim, remain in the realm of riddle. What was the guilt that called for tisis (retribution)? Does the address “lion” point to Hipparchus’ monarchical status, to tyranny as a field of adikie (injustice), or just to the fortitude of one who must prepare for death?155 Words and facts combined do not always lend themselves to reconstructing a clear overall picture. The interjection of a great number of oracles in the logos contributes to its meaning but also turns it into a series of riddles. 154. Crahay (1956, 40) counts ninety-six oracles in Herodotus. Their function in the narrative is examined by Kirchberg (1965); see also Lachenaud 1978, 244– 305. Dreams in Herodotus are examined by Frisch (1968). See also Harrison 2000, 122– 57. On non-Greek mantic prophecy, see Klees 1965; Lachenaud 1978, 229– 44. 155. For the lion as a term of both the monarchic and the heroic codes, see chap. 4, “Wondering Why.”

Interpretation and Evaluation

197

Herodotus’ collection includes oracles impossible to decode except retrospectively (3.57.4) and “double” or ambiguous predictions that are bound to come true—in some sense—either way. There are dreams that are perhaps not divine (7.12– 18), responses obtained through corrupt means or falsified by dishonest professionals (6.66, 7.6.3– 5), oracles that have allegedly failed a test of veracity (1.46– 49), and prophecies that are part of self-serving traditions in the discourse of international disputes (7.148.3). Among the most problematic are the two oracles the Athenians receive from Delphi before the arrival of Xerxes to Greece. Both appear to discourage resistance against the Persians.156 The narrative emphasizes the stubborn determination of the Athenians and their refusal to accept the first oracle. Their ingenuity in making the second mean what they want it to mean might in other circumstances appear an exercise in false hope (7.142– 43; cf. 1.53.3). At Mycale, an omen of victory helps the Greeks defeat the enemy by offering encouragement; here, the Delphic responses are terrifying (7.139.6) and therefore potentially self-fulfilling in a negative way. If the Athenians became the “saviors of Greece,” it is because they chose to fight at Salamis despite the oracles. Against the background of this aporia-generating complex of evidence, the narrator steps out to proclaim the principle of the veracity of divine predictions. The defensive quality of the discourse has here impaired the normal form of what I have so far been able to call a “maxim.” Nevertheless, a generalization it is, firmly attached, according to the usual procedure, to the display of a particular exhibit.  1. As for oracles, I cannot contradict/object [αντιλ εγειν] that they  are not truthful [αληθ εες], not wishing to attempt to devalue them when they speak clearly [ε ναργεως], looking at facts like these: When they bridge with their ships the sacred shore of golden-sword Artemis and marine Cynosoura, having sacked lovely Athens in their mad hope [ε λπι δι η], µαινοµεν 156. See 7.140– 41. For what both of these oracles “really” mean, see Elayi 1979, 227– 30.

198

Telling Wonders

divine Dike shall quench strong Koros, son of Hubris, dreadfully furious, who thinks he rules all things.157 2. For bronze shall clash with bronze, and Ares will redden the sea with blood. Then the all-seeing son of Cronos and the lady Victory will bring over the day of liberty for Hellas. In the face of such things and against Bacis speaking so clearly [ε ναργεως], neither do I dare myself to utter contradictions  [αντιλογ ι ας] about oracles, nor will I accept them from others. (8.77.1– 2) Through the self-referential and double negative form of the statements framing the quotation of the oracle, like the reference to the “opinion that will cause resentment” at 7.139.1, the text encodes an unreceptive attitude among the audience.158 Moreover, just as at 7.139.1 the narrator represents himself as overcoming also his own resistance and compelled publicly to display what “appears [to him] to be a true thing”  so here he comes across as an objec[τ η γε µοι ␸αι νεται ειναι αληθ ες], tive investigator who must yield to the evidence of truthfulness (ε ι σ ι  αληθ εες, 1). This time, the results of historie reassign aletheia (truth) where it properly belongs—the sphere of divine utterances. Herodotus’ statement “Looking at facts like these, I cannot utter contradictions” implies a readiness to contradict, if the empirical evidence so required. Herodotus’ position with respect to divine matters is, as we have seen, in certain respects related to the agnosticism of the Sophists. What distances Herodotus from the Sophists is the recurring message that his research keeps confirming the relevancy of the divine in both history 157. I am preserving the reading of the manuscripts, though How and Wells (1928, 2:262) say it is meaningless, and though several emendations have been proposed. All of them, at any rate, refer to the relentless will to power of Koros. 158. Skepticism in the late fifth century with regard to oracles is exemplified by the attitude of Thucydides (at, e.g., 2.17.2, 2.54.2– 3, 3.96.1, 5.26.3; see Gomme 1956, 160– 61). Even in Athens, the general public would have held more traditional views (see, e.g., Thuc. 2.52– 54, 2.8.2– 3), but modern ideas, the war, the alleged Delphic oracle favorable to Sparta (see Thuc. 1.118.3; cf. 2.54.4), and the plague contributed to the erosion of religious beliefs. Jordan 1986; Forrest 1984, 7. Ideologically, Herodotus’ assertion of the veracity of oracles is the counterpart of Sophocles’ Oedipus Tyrannus, produced in a similar climate. See Dodds 1966, 47; see especially Knox 1971, 159– 84.

Interpretation and Evaluation

199

and culture.159 Here, therefore, his profession of belief takes the form of an emphatic refusal to counter with sophistic speech the speech of god. He cannot antilegein (both “utter a rebuttal” and “make antilogies,” the last a technical term of sophistic argumentation), and he will not accept antilogiai from others.160 Daring to do so would risk offending the gods (cf. 2.45.3). But Herodotus most especially wants to protect his oversecularized audience from another danger: a disregard of divine speech that is bound to entail the misunderstanding of history. Herodotus’ generalization on the veracity of oracles is in fact attached to a special item of proof, a response that both predicts intelligibly and interprets accurately. This prophecy of Bacis corresponds point by point to the historical action of the battle of Salamis, whose narrative, as if in a movement of spontaneous recognition from the part of the narrator, it suddenly interrupts. Unlike the Delphic oracles to the Athenians concerning the same event, an oracle that speaks with unmistakable and vivid clarity corroborates a view of the divine that rehabilitates all its utterances. This oracle comes from a relatively minor source rather than from the politically involved Delphic shrine, and it is found in an ancient collection rather than being solicited for the occasion. On the strength of the evidence it provides, the ambiguous and the obscure also find their place in the scheme of things. This last point leads us to the interpretive aspect of the oracle as a part of that enargeie (clarity: 8.77.1– 2) that makes Herodotus identify it as decisive proof of the aletheie (truthfulness) of the divine word. In the case of numerous prophecies recorded in the logos, the historical context does not call for a moral judgment; if it does, the god may express it unclearly or even appear to go out of his way to avoid providing moral guidance. When people inquire about what they should do, the god likes to put the ball back in their court, leaving them to decide on the basis of their judgment of circumstances, their values, and their moral sense.161 By contrast with these cases, the oracle of Bacis concerning Salamis clearly formulates divine criteria of justice and even amounts to a direction for reading history in ethical terms. The Persian invader here becomes the embodiment of koros (surfeit/insatiability/excess), the son of hubris (violence/arrogance/transgression), which δια ∆ ι κη, “divine justice” or 159. See “Equal Knowledge” earlier in this chapter. 160. On the sophistic uses of the term antilegein, see Kerferd 1981, 60– 67, 84– 107. 161. Consider, e.g., the oracles to the Cymaeans (1.158– 59); to the Thebans and the Athenians concerning Aegina (5.79.1, 5.89.2), and about the fall of Miletus (6.19.2).

200

Telling Wonders

the justice of Zeus, finally suppresses.162 In the archaic poetic tradition,   (Pind. the word κορος, strictly associated with υβρις, as here, or with ατη Ol. 1.55– 57), denotes the negative repercussion of olbos in the sense of “material prosperity.” Koros is an undesirable excess of good things (food, wealth, power), the state of being glutted with them. It causes blind action and its disastrous consequences.163 The tyrant in Otanes’  formulation of the monarchical model is υβρι κεκορηµενος [glutted with  arrogance] (3.80.4). The word υβρις is applied twice more to the typical monarch in the Constitutional Debate, and the same stem recurs to describe either the criminal acts of kings or individuals with monarchical leanings or a people’s military aggressions.164 In the expression ε λπι δι η (8.77.1, line 3), elpis denotes the unfounded expectation of a µαινοµεν king in Herodotus’ logos, (1.80.5), and its metaphorical qualifier mainomene (crazed) recalls Herodotus’ attribution of literal madness to the most extreme of his monarchical paradigms.165 The oracle of Bacis interprets the Persian defeat of Salamis in a way that agrees with the generalization of divine retribution formulated by the narrator (2.120.5) and by the dream of Hipparchus (5.56.1). The transgressor is here the monarch in a literal and metaphorical sense, and the transgression is an imperialistic attack. Within the limited geography of Salamis, the oracle represents the battle in the same terms by which Herodotus himself throughout the logos visualizes the whole idea of transgression. The phrase “When they bridge with their ships the sacred shore of golden-sword Artemis and marine Cynosoura” (8.77.1, lines 1– 2) represents an image equivalent to the bridging of the Hellespont by Xerxes and all the other violations of physical boundaries, not as real geographical subdividers, but as the symbols of ethical laws. When the Persians bridge the strait of Salamis with the encircling maneuver, this constitutes, in the words of the oracle, an act of hubris that has as its goal 162. 8.77.1, line 4. For an early testimony of retributive justice, connected with Zeus, see Il. 16.384– 92. The formulation of the concept is discussed by Lloyd-Jones (1971). In Herodotus, cf. Hermotimus’ mention of the νοµος δι καιος [law of justice] of the gods at 8.106.3. 163. The intimate association of koros and hubris is confirmed by Solon frag. 5 Diehl3, 9, and Theog. 153– 54, where the second is the offspring of the first rather than, as in Herodotus’ oracle and in Pind. Ol. 13.10, the other way around. Cf. Aesch. Agam. 766. See Nagy 1990, 131, 281– 82, 291– 92; for the Bacis oracle in Herodotus, see 327. 164. See 3.80.2, 3.81.1, 1.114.5, 3.118.1, 3.126.2, 3.127.3, 6.127.3, 8.3.2, 9.27.2,  τυραννον 9.73.2, 5.77.4, 6.87, 7.16α2. Cf. Soph. O.T. 872: υβρις ␸υτευει (see end of n. 158 in the present chapter). 165. See chap. 2, “The Monarchical Model in Sparta,” text and n. 71.

Interpretation and Evaluation

201

to enslave the Greeks. The Persian defeat by Dike on that occasion, therefore, brings “the day of freedom” as a gift to the Greeks (8.77.2, line 8). This corresponds to Herodotus’ own account of the naval battle of Salamis as the crucial event that, as he says, allowed Greece to “survive in freedom” [περιειναι ε λευθερην] (7.139.5). It also corresponds to the idea advocated in the Histories that the ultimate act of hubris is the attempt to expand one’s rule by enslaving others. The oracle of Bacis, then, not only proves the veracity of divine utterances in view of the precision with which it predicts the subsequent factual course of events— what happened, when and where, and for whom. It also agrees point by point with Herodotus’ own understanding of the events’ meaning and encapsulates in a short utterance all the most important principles of his historical logos and of the message it wants to convey. From the point of view of the apodexis (performance), this oracle serves to validate Herodotus’ interpretation, just as from the point of view of the historie (inquiry), it may well constitute its source.166 Herodotus’ understanding of history and his interpretation of divine signs affect one another in the case of one natural phenomenon, less eloquent than an oracle, but unique and specially timed: the earthquake of Delos. Thucydides must reflect a widespread contemporary belief in mainland Greece when he says that the event, unprecedented in the history of the Hellenes, occurred “shortly before” the outbreak of the Peloponnesian War and was interpreted as a semeion (sign) of that imminent trouble.167 But Herodotus, perhaps correcting this notion, cites the authority of the Delians to assign the earthquake to 490 b.c., exactly when the Persian general Datis and his force, on their way against Greece, pass the island at the middle point in the Aegean and thereby cross the ideal boundary between Asia and Europe.168 On that occasion, he specifies, “Delos shook . . . for the first time and the last down to my day” (6.98.1). Herodotus does not contradict the public perception that the prodigy predicted the Peloponnesian War, but he rather uses it to broaden the significance that it would be more natural to attribute to the earthquake if it occurred when the Delians say it did. He interprets it, in other words, 166. For the view that the story of the Persians’ encircling maneuver at Salamis derives from the oracle, see Immerwahr 1966, 278– 79. 167. Thuc. 2.8.3. For discussion of the discrepancy between Herodotus and Thucydides, see How and Wells 1928, 2:104; French 1972, 21; Stadter 1992, 788– 99. 168. 6.98.1. See Stadter 1992, 785– 95, especially 787.

202

Telling Wonders

as a sign not merely of the misfortunes of the Persian Wars but also of the subsequent turmoil from the wars of Greeks against Greeks down to the moment of narration. The earthquake of Delos indicates that the campaign of Datis and Artaphrenes in 490 b.c. was opening for Greece a new epoch of about one hundred years, scanned by three generations of Persian kings. κου] this was a prodigy [τερας] 1. And no doubt [µεν that the  that were about god manifested to men for the evils [κακων] to happen. 2. For in the time of Darius, the son of Hystaspes, Xerxes, the son of Darius, and Artaxerxes, the son of Xerxes, during these three consecutive generations, more evils happened to Greece than during the twenty other generations that came before, some [of these evils] deriving to Greece from the Persians, and some from the leaders themselves fighting over the rule/empire/hegemony  αυτ  ων  των  κορυ␸αι ων περ ι της αρχη   ς πολεµεοντων]. [απ  εν  . . . αεικ  that Delos 3. So, it was not at all out of order [ουδ ης] be shaken, having previously been unshaken. (6.98.1– 3) The gloss verifies the maxim I considered earlier that the gods send “great signs” in anticipation of “great evils” (6.27.1). Such an extraordinary phenomenon is justified, according to the narrator, by the unprecedented dose of kaka (evils) that followed, brought about first by external and then by internal war. As we have seen happen in the course of Herodotus’ verification of cases of retribution, so also here the chronological progress of the events in the logos, coupled with the special timing of the divine, causes the narrative to rejoin the time of the narration, thereby establishing both analogy and continuity between the past and present. Several fifth-century sources look at the Persian Wars and the subsequent wars among Greeks as parallel entities with respect to size, strategy, and the ethical status of the participants. According to the official Athenian position, the achievements of Athens after the Persian Wars reproduce Athens’ achievements against the barbarians.169 On the other side of 169. Pericles compared Agamemnon’s Trojan War, as the archetypal war against barbarians, and his own Samian War (Plut. Per. 28.7, quoting Ion). The paintings of the Stoa Poikile may have set battles of the Greeks against Amazons, Trojans, and Persians side by side with one between Athenians and Spartans (Paus. 1.15). See Holscher ¨ 1998, 173– 76.

Interpretation and Evaluation

203

the ideological struggle, the “tyrant city” tradition of political discourse casts Athens, the onetime liberator of the Greeks, in the role that had formerly been played by Persia.170 For Herodotus in the passage I am considering, the analogy between the two sets of wars is based on the kaka that they both have brought to Greece. Just as the Persians bring evils on Greece for the sake of increasing their arche (empire) or of turning Greece itself into their arche (satrapy), so the leading Greek states bring evils on Greece by fighting with one another for the sake of the arche.171 Athens is not here the keystone of the analogy, except insofar as it was the only city of Greece that, after the Persian Wars, had fought for and acquired what was in fact called an arche (empire). Many interGreek wars, including the Peloponnesian War, had started as fights over Athens’ arche. Herodotus’ special understanding of the divine sign establishes continuity between the two sets of kaka by bracketing off as a unit the onehundred-year period of the reigns of the three kings Darius, Xerxes, and Artaxerxes in contrast with the previous twenty generations.172 As additional proof that this time constitutes a unity and is all therefore covered by the omen, he ends by pointing out the ominous significance of the kings’ names (6.98.3). The epoch of kaka begins with the earthquake of Delos at the time of the Marathon campaign (490 b.c.) and, implicitly, with the advent to the throne of Darius, the first of the three kings (521 b.c.). In For the continuity of Athenian achievements in the tradition of the epitaphios, see Loraux 1986, 132– 71. 170. See Elpinice’s reproach to Pericles in Plut. Per. 28 and the parallel made by the Corinthians in Thuc. 1.69.5. Implicit comparison between the Persian Wars and the Peloponnesian War is pervasive in Thucydides. See especially Connor 1984, 155– 57, 175– 76, 198– 200. Herodotus’ narrative describes Persian actions against the Greeks that suggest future Athenian actions: the plan to occupy Cythera, an item of Athenian strategy from 454 (implemented in 424), is discussed by the Persians (7.235); Persian imperialistic speeches contain terms of a code that at the time of narration were applied to Athens (7.5– 9, 7.45– 53); the report of the Persians’ establishing democracies in Ionia after the Ionian revolt (6.43.3) may allude to the Athenian practice to establish democratic governments in the states of its league. See Raaflaub 1987, 227– 28. 171. See Nagy 1990, 308, on this passage. 172. Going back twenty generations before Darius brings us to the time of the Dorian settlement in the Peloponnese, which marks the beginning of the Spartan king list. For a general scheme of Herodotus’ chronology, see Lloyd 1975, 171– 94, especially 177– 82. It would be attractive to be able to interpret the twenty-generation period as starting right after the archetypal event of the fall of Troy; but since the three subsequent generations here mentioned cannot amount to more than about one hundred years, we cannot make twenty generations reach back to ca. 1330– 1250 b.c., the approximate time of the Trojan War according to Herodotus’ chronology.

204

Telling Wonders

between these two events, the narrator elsewhere identifies another beginning, the Athenian expedition in support of the Ionian revolt (499 b.c.): Herodotus calls the Athenians’ twenty ships with a pointed allusion to the  η κακων]  for both paradigm of all wars, the “beginning of evils [αρχ Greeks and barbarians.”173 The three-king “century” includes the wars against Persia, the fighting that went on during the Pentecontaetia, and the Peloponnesian War down to the time of performance. The events of the Pentecontaetia are telescoped together with those of the Peloponnesian War. Under the heading of kaka deriving from the leaders of the Greeks fighting for the arche, both are presented as a natural continuation and extension of the kaka imported by the Persians.174 To Herodotus, in other words, the wars of the past, the Persian Wars, have perpetuated themselves. And in fact, according to his narrative, the ships built for the war against Aegina, ships that “saved Greece” (7.144.2) and allowed the Athenians to be “saviors of Greece” (7.139.5), become available again very soon for use against other Greeks. Herodotus puts no interval between the battle of Salamis and Themistocles’ attack on the islanders from whom he exacts monetary contribution by force.175 The defensive war of the Greeks against the Persians is still going on, and the tables have already turned. Immediately after Salamis, the Greeks also raise the issue about their pursuing the war against Persia offensively, turning it into a war “about the king’s own country,” as the  expression goes (8.108– 9, especially 8.108.4: το ε ν␽ευτεν δ ε περ ι της   ποιεεσθαι  ωνα;  ε κει νου ηδε τον αγ cf. 8.3.2). Similarly, after the godgiven victory of Mycale, the Greeks reject the option to move the Ionians out of Asia, the Athenians choose to preserve the “Ionian arche,” a league is founded with the Athenians in charge, and the Greek fleet departs for the Hellespont.176 These clear elements of continuity with the future, 173. 5.97.3 (see also 5.28.1, 5.30.1). Cf. Il. 5.62– 63. See Cobet 1971, 63. 174. For the Pentecontaetia as an intermediate period and a fighting prelude to the Peloponnesian War, see Thuc. 1.89– 117. Herodotus mentions the Peloponnesian War as a separate entity only at 9.73.3. 175. See 8.111– 12, a passage affected by contemporary rhetoric and referring to fifthcentury oppressive Athenian diplomacy toward smaller states. The resemblance to Thucydides’ Melian Dialogue has been noted by many (see, e.g., Aly 1929, 99; Strasburger 1955, 21; Gigante 1956, 136 and n. 1). See Stadter 1992, 795– 98. The idea that, historically, Athenian policy became very aggressive soon after 478 is argued by Fornara and Samons (1991, 76– 113). Herodotus makes the change start even earlier. 176. See 9.106.2– 4. The divine signs at Mycale (which I discuss earlier in this section) prevent interpreting the Greek crossing of the Aegean beyond Delos to liberate Ionia as a violation, as Stadter does (1992, 892– 94; see rather Immerwahr 1966, 288); on the other

Interpretation and Evaluation

205

especially the smooth transition to the offensive stage of the war and to the already ambivalent leadership of Athens, rob the Histories of a clearcut triumphant closure: the narrative simply stops shortly afterward, precariously balanced at the Hellespont and full of negative signals.177 The open-endedness of Herodotus’ logos, the suggestion that the outcome of the second rebellion of Ionia is only preliminary to further fighting, and the proleptic references to the later wars of Greeks against Greeks scattered through the last part of the history of the war against Persia178 agree with the gloss interpreting the earthquake of Delos as the announcement of a continuous chain of evils, with no relief or divide between the war he is narrating and subsequent internal struggles. The “beginning” of the three-king period is staggered over a number of years and marked by several events (the advent of Darius, the Ionian revolt, the expedition of Datis with earthquake). Its completion must be similar. The narrator positions himself still within that period, though at a point when the trend of events in Artaxerxes’ reign could already be evaluated. By then, the outbreak of the Archidamian War and all the accompanying catastrophes we learn about from Thucydides must have appeared to contemporaries to signal the beginning of the end of something.179 If we accept this as the extratextual context of Herodotus’ performance, his interpretation of the earthquake of Delos has a prophetic side: it contains an implicit prediction that a change is imminent and that a new epoch is about to begin in the next generation of Persian kings. What this change will be, what the wars of the koruphaioi of the Greeks will finally produce, remains unknown. In search for guidance for reading the future, Herodotus has canvassed the past. From the mass of sometimes uncertain evidence, he is able to derive three maxims: that hand, the aggressive operations against Andros, Paros, and Carystus, on this side of Delos, are obviously represented as unjust acts. This is another demonstration that continental boundaries as such should be taken not too rigidly but merely as symbols of ethical boundaries in specific transgressive situations. The Greek actions after Mycale are, however, presented as worrisome; the very existence of Greek cities in Asia emerges as problematic. Compare 9.106.2 with 1.164– 68 and 1.170: at 1.170, if they had followed the advice of Bias the Ionians, according to a gloss of the narrator, “could have been most prosperous among the Greeks.” 177. See chap. 2, “The Monarchical Model of Athens.” 178. See 9.35.2, 9.73, 9.75, 9.105 (Athenian siege of Carystus; cf. Thuc. 1.98.3); Schmid and Stahlin ¨ 1934, 1/2.590 n. 9; Cobet 1971, 59– 82. 179. Fornara (1971b, 32– 33; 1981, 150) argues that this passage was composed after the death of Artaxerxes (424 b.c.) and even after the end of the Archidamian war. But see Cobet 1977, 5.

206

Telling Wonders

human eudaimonie is unstable, that the gods punish wrongdoings, and that divine messages are true. Of only five explicit references to what we may broadly regard as the narrative now of the logos (i.e., the time after the outbreak of the Peloponnesian War), at least three are brought in for the purpose of proving with facts the validity of these general rules.180 Athenian Antilogies The Histories are a collection of worldwide historical evidence that ultimately seeks and offers guidelines for an objective assessment of the contemporary kaka of the Greek city-states at the time of performance and their future prospects. The histor takes an outsider’s position, and the work is directed not against any particular polis but to all the Greeks. The interpretations I have examined make clear, however, that the image of Athens as an unfinished paradigm of rise and fall and the issues of Athens’ power, merit, and guilt loom large in the second part of Herodotus’ logos. The sure sign of Athens’ prominence first occurs at the point in the narrative when that city acquires freedom from tyranny. By virtue of its new internal order, Athens “becomes great” and is the only state in the Histories that prompts the narrator to generalize in his own voice on the issue of government. While the metaphysical maxims attempt to derive absolute cosmic, ethical, and divine laws from the data provided by the visible world, when the narrator observes political realities, he acknowledges them to be inherently mixed. His two maxims on democracy occur twenty chapters apart and in the same narrative framework. They are structurally parallel and antithetical in substance. 1. And it is clear [δηλοι ], not only in one single respect but in general, that equal right to address the assembly is a serious asset [η ι σηγορι η . . . ε στ ι χρη µα σπουδαιον], if also the Athenians while they were being ruled by a tyrant were no better in war than any of their neighbors, but once they were rid of the tyrants they became by far the first. This then demonstrates that when they were being held down, they fought badly on purpose as working for a master, but once they became free,  εκαστος  ος  each man was eager to achieve for his own sake [αυτ ε ωυτω  προεθυµεετο κατεργα ζεσθαι]. (5.78) 180. See 6.91.1, 7.137, 6.98.2– 3 (discussed earlier). For 7.233.1, see n. 149 in the present chapter. The last occurs at 9.73.

Interpretation and Evaluation

207

 2. It seems that it is easier to deceive many men than one [πολλους  οικε ειναι ευπετ   γαρ εστερον διαβα λλειν η ενα], if Aristagoras could not deceive Cleomenes of Lacedaemaon, who was alone, but managed to do so with thirty thousand Athenians. (5.97.2) This antilogical set is related to the contrasting arguments of the Constitutional Debate, in which three characters analyze the advantages and disadvantages of the three basic forms of government: democracy, oligarchy, and monarchy (3.80– 83). There, each speaker holds a fixed subjective position in favor of one form of government and opposes one or two of the remaining ones. In the overall picture from the point of view of the text, however, none of the three forms appears immune from dangers for the commonwealth. The derivation of the debate from a relativistic discourse of Protagorean stamp is evident.181 The most conspicuous part of the contest is played out between monarchy and democracy, which appear mutually antithetical and parallel at the same time. In a monarchical regime, one man is permanently in charge and does what he wants without being accountable (3.80.3), while democracy is characterized by accountability, the rotation of offices, and common deliberations (3.80.6).182 Both the monarch and the demos, however, are susceptible to hubris (3.80.3; cf. 3.81.1– 2) and wrongdoing (κακοτης, 3.82.4; cf. 3.82.4). As the narrator takes stock of the contradictory character of the historical evidence, the phenomenon of the Athenian democracy prompts him to transfer to himself the antilogic functions that the Constitutional Debate distributes among different characters. The duality of Athens that emerges from other late-fifth-century texts (pure and impure, innocent and guilty, liberator and enslaver) is a central feature of that city’s portrayal in the Histories.183 Because Athens is double, it can only be described, as here, antilogically, in a form of discourse that is particularly 181. Lasserre (1976, especially 81) argues that Protagoras was the author of a constitutional debate also in a Persian setting. On the partial affinity of Herodotus’ thought with that of the Sophists, see discussions in earlier sections of this chapter. 182. The democracy of the debate (called isonomie at 3.80.6 and 3.83.1, democratie at 6.43.3) is described in terms suitable to Cleisthenes’ democracy (called democratie at 6.131.1 and 5.78). See Ostwald 1969, 96– 160; Fornara and Samons 1991, 41– 56. 183. E.g., the Spartan ambassadors at 8.142.3 contrast Athenian responsibility for the enslavement of the Greeks (α ι τι ους δουλοσυνης τοισι Ελλησι) and their role as liberators   (πολλους ε λευθερωσαντες ανθρ ωπων). Herodotus offers contradictory evaluations of Pisistratus’ reign (1.59.1, 6.). See Strasburger 1955, 10– 15. For Themistocles as the embodiment of Athens, see his antilogic arguments at 8.108– 9; see also Munson 1988, 103– 4.

208

Telling Wonders

congenial to that city and its perception of things. These glosses point in fact to another Athenian trend. Unlike most of the narrator’s generalizations we have seen so far, and unlike the discourses on government in the Constitutional Debate, they employ the vocabulary of strategy and the  useful (χρη µα σπουδαι ον, ευπετ εστερον διαβα λλειν), in preference to the ethical code used to evaluate something being good or bad.184 Both generalizations assess the advantage of the opportunities for deliberation enjoyed by the Athenian demos in the assembly, both evaluate this advantage by its results in the field of war, and both contrast it with the concentration of power in the hands of one man. In the first generalization (5.78), the term isegorie (equal speech) underlines the psychological effect on the ordinary citizen of his being able to stand in the midst of his peers and participate in the debate about policies. This effect extends beyond the field of deliberation (“not in one thing only, but in a general way”) and is here measured by performance on the battlefield. In the case of the Spartans also, Herodotus’ narrative conveys the notion that freedom makes men more eager to fight than does subjection to a ruler (7.135.3, 7.101– 4). But whereas Spartan arete, according to Demaratus, partially derives from an unquestioning obedience to the city’s nomos-despotes, stronger than any human despot (7.104.4– 5), the Athenian brand of freedom replaces the human despotes not with “the law” but with the individual η ε ργαζοµενοι). citizen (ε ωυτω   . . . κατεργα ζεσθαι vs. δεσποτ Since the Athenian citizen fights in wars that he has personally deliberated on through his right to speak in the assembly (isegorie), or, in other words, since he makes the laws, he “works” in his own interest, “each one for  εκαστος  ος  himself” [αυτ ε ωυτω   ]. Individual self-interest is the ideological foundation of popular sovereignty in the democratic state.185 The positive evaluation of isegorie is based on the specific evidence of a success in battle: the Athenian defeat of an aggressive attack by Boeotians and Chalcidians (5.77; cf. 5.74.2). The narrative that illustrates the 184. The predominance of the useful over the just in political deliberations is often mentioned by Athenian speakers in Thucydides (see, e.g., Thuc. 1.75). The theme of selfinterest recurs obsessively in the evaluation of the Athenian democracy by the “Old Oligarch” ([Xen.] Ath. Pol. 1.6– 9, 2.20, and passim). In Herodotus, the Athenians make the transition from an ethical to a utilitarian discourse in their speeches at 8.143– 44 and 9.7, 11; see particularly 9.7α1– 2.  185. See Munson 1988. The term εκαστος (each) is an index here of individualism, elsewhere of the particularism of states (1.169.1, 7.219.2, 8.57.2). The distributive form distinguishes Herodotus’ description of democracy from that of the Hippocratic author of  . . . κινδυνους  περι κινδυνεουσι Airs, Waters, Places: τους ε ωυτων (16; cf. 23).

Interpretation and Evaluation

209

successful implementation of isegorie itself is rather the report of the assembly deliberations on military policy after the oracle of the wooden wall (7.142– 43): here, the self-interest of the Athenians, empowered by freedom from tyranny and right of speech, leads to the “Good” decision to resist the Persians. The system proves again to be a “serious asset” for the commonwealth as a whole.186 The antithetical counterpart of this assembly scene is another assembly scene, which leads to a “Bad” war decision and provides the textual context and the historical evidence for Herodotus’ second, and negative, generalization on the democratic state, at 5.97.2. This evaluation occurs in the narrative of Aristagoras’ journey to mainland Greece for the purpose of obtaining support for the Ionian revolt. The narrative itself in turn provides the frame for the analeptic sequence on Athens’ liberation from tyranny and first successes, concluded by the narrator’s praise of isegorie. Thus, the coming of Aristagoras looms on the horizon, both historically and narratively, during the decisive moment of Athens’ democratic beginning and consequent rise to prominence. When he arrives holding out hopes of plunder, Athens is powerful (ε δυνα στευεν µεγιστον) and ready for an imperialistic war.187 In Herodotus’ narrative of Aristagoras’ visit, however, the representation of the Athenians’ new democratic energy, restlessness, and self-interest remains subsidiary to a less grandiose portrayal. In Athens, Aristagoras repeats the tendentious ethnographic information he has given at Sparta (5.49.3– 8), with the aid of a map such as the histor Herodotus has elsewhere judged oversimplified and inadequate.188 He adds the idealistic argument that the Milesians, as colonists of the Athenians, should be able to count on their help.189 Whereas he has failed with Cleomenes at Sparta, Aristagoras persuades the Athenians by talking of the wealth of Asia and the Persian method of warfare, in which “they use neither shield nor spear  . . . χειρωθη ναι].”190 In Herodotus’ so as to be easy to defeat [ευπετ εες 186. Similarly, the report that follows, on the earlier deliberation concerning the building of a fleet, shows that the Athenians are able to understand when private and public interests coincide (see 7.144). 187. See 5.97.1. For the connotations of dunasteia, see chap. 2, n. 32 and corresponding text. 188. See 5.49.1. Cf. 4.36.2 (see chap. 2, “The Texture of the Earth”). 189. 5.97.2. This replaces the argument of Aristagoras in Sparta on the “shame and grief” that the enslavement of the Ionians brings on the Spartans as the leaders of Greece (5.49.2– 3). That he makes this point before the representation of Eastern wealth reflects his assessment of what would be persuasive to each audience.   χειρωθη ναι). 190. 5.97.1. Cf. 5.49.3 (ευπετ εως), 4 (ευπετ εες

210

Telling Wonders

explicit interpretation at 5.97.2 (passage 2 quoted earlier), the people’s decision to support the revolt in Ionia—in turn interpreted with the famous “beginning of evils” gloss (5.97.3)—is first and foremost a demonstration of the people’s collective ignorance. As one of the speakers in the Constitutional Debate observes in stronger terms (3.81.2), this is bound to put a democratic state at a disadvantage. Throughout the Histories, Herodotus thematizes together with Athenian sophie (cleverness) the paradox of Athenian euethie (simpleminded της, ness). After Aristagoras, Miltiades deceives the Athenians (απα 6.136.1), and then Themistocles does too (διεβαλλε, 8.110.1). In their earlier history, the Athenians are deceived twice into putting a tyrant in ε ξαπατηθει ς, 1.59.5– 6; µηχανωνται   place (µηχανη, . . . πρηγµα ευηθεστα τον, 1.60.3). They take a local maiden for a goddess (1.60.3), are blind to tyrannical schemes, and do not know the world. They are  persuaded that grandiose enterprises will be easy (ευπετ εως, 6.132; cf.  χειρωθη ναι 5.97.1), because they are themselves, like all multiευπετ εες  tudes, easy to deceive (ευπετ εστερον διαβα λλειν, 5.97.2). To the ambivalence of Athens in the ethical sphere corresponds a contradiction at the level of knowledge and intelligence. This factor cuts the image of Athens down to size. We are reminded of the besotted Demos in Aristophanes’ Knights or, more strikingly, of the assembly that in Thucydides deliberates on the Sicilian expedition—sovereign, vociferous, and ready to go, but not competent or truly in charge.191 At the same time, Herodotus’ general evaluations serve a broader purpose than describing their primary referents. Just as he proclaims that freedom from tyranny and the citizens’ right to make policies through speech is a great advantage for all, so he notes that vulnerability to speech makes democracies blunder. This is due less to monarchical hubris than to the simplemindedness of the ordinary Greeks who run the state. When it comes to euethie, the people of Athens in the logos has much in common with the audiences Herodotus’ logos addresses, both Athenian and not (cf. 2.45.1). Other than communicating a more abstract moral message, the histor takes it on himself to display and to cure through his own, non-Aristagorean brand of speech this shared na¨ıvet´e about the reality of foreign peoples and lands, the shape of the world, the motives of leaders, 191. See Thuc. 6.8– 26. Athenian ignorance of the realities of Sicily (see Thuc. 6.1.1) prompts an exceptionally lengthy ethnographic and geographical passage in Thucydides (6.2– 6.5.3). See also Eur. Suppl. 410– 25, with its positive counterpart in the praise of isegoria at 432– 41.

Interpretation and Evaluation

211

and the correct and falsified signs of divine support. Ignorance in these matters affects public decisions and brings about the “evil” of unnecessary wars. The Evils of War The two interpretive glosses I have just considered assess the repercussions of a political regime on war performance and war deliberation, respectively, in a way that is consistent with Herodotus’ overarching message about these topics. Among the “achievements great and wonderful” that the apodexis histories promises to save from oblivion, acts of courage and prowess are, as in Homeric poetry, among the most conspicuous objects of the narrator’s praise.192 In the experience of the narrator, those qualities that ensure a high level of performance in battle are tied to the accomplishment of important actions that affect whole communities and represent the morally neutral “great deeds” of history.193 Equally conspicuous, however, is the idea of the intrinsic moral value of arete (excellence) in war, as the performance of a public duty and as a display of personal and national merit. A people’s struggle in war on behalf of freedom is especially “worthy of description” (see, e.g., 1.177, 2.157). But armies, contingents, and individuals, be they Greek or barbarian and whatever their cause, are regularly evaluated for the extent to which they rise to the occasion in the moment of battle. If arete is an absolute moral value, however, war is just as absolutely an evil, as the glosses on the three generations of kaka and on the beginning of kaka indicate (6.98.2, 5.97.3). This is again a Homeric idea, as is the attribution of kleos to war deeds. In Herodotus, however, the evaluation of arete competes with the assessment of responsibility and dike (justice), so that since war is an evil, making war is a culpable action, not at the level of fighting, but at the level of deliberation. Thus, Croesus, the only character in the Histories to utter a gnomic saying on the evils of war, also raises the question of who is responsible for bringing it about: “It was the god of the Greeks who, having incited me to make the expedition, was responsible [αιτιος] for these things. For no one is so out of his mind as to 192. See “Specific Glosses of Interpretation” in this chapter; chap. 1, “Referential Glosses.” See also Immerwahr 1966, 262– 65, 308– 9; Sa¨ıd 1980. Dillery (1996) examines the pattern of the duel as a contest of worth. 193. See the generalization at 7.153.4. The case that contradicts it suggests divine intervention.

212

Telling Wonders

choose war over peace. For in peace, children bury their fathers, and in war, fathers bury their children” (1.87.3– 4). Of course, Croesus is wrong on two counts. Many rulers choose war over peace in the Histories, and
 ρξαντα like the rest of them, Croesus is aitios for the wars he initiates (υπα

ι κων εργων, 1.5.3; cf. the Dephic oracle at 1.91). αδ In the distinction between providing for national security and waging a war of aggression, Herodotus’ didactic message makes one of its main points by countering the contemporary tendency to view aggression as a function of defense.194 Embedded in the evaluation of the special strategy that makes the Scythians immune from external enemies is the generalized statement that to possess means of defense is essential for any nation:  το µεγιστον   “The single most important of all human things [ε ν µ εν των

ανθρωπη ι ων πρηγµατων] has been devised by the Scythian nation in the most intelligent way we know about” (4.46.2). By the same token, Herodotus’ narrative indirectly criticizes the Ionians because, for one reason or another (lack of unity, organization, discipline, or mobility), they are unable to provide for themselves and safeguard their freedom (see, e.g., 1.169.1, 6.11– 12, 1.170). The symbolic significance of the Scythians in the Histories is to a great extent connected with their exemplary ideology of war. Though warlike by custom and unconquerable, the Scythians manage to make war as little as possible. In the episode of the Scythians and Amazons, the crossing of the Tanais signifies, as we have seen, a renunciation of aggression.195 The Scythians’ imperialistic exploit, in which they “started a war” and “began the injustice,” and for which they have received divine retribution, lies far in their past.196 They are henceforth firmly autarchic, with no impulse to expand, and are even able to avoid battle if their land is attacked. If reduced to an extreme position of defense, they are ready to fight (4.127.3). Among the Greeks of the Histories, a similar attitude toward war is attributed to the Spartans, whose imperialism has also mostly spent itself at an earlier stage of their history.197 For the benefit of Xerxes, who has 194. See the Athenians in Thuc. 1.75 and the Spartans in Thuc. 1.23.6, as well as the preemptive aggressions of Croesus and Xerxes in Herodotus (1.46, 7.11.2). See Payen 1997, especially 79– 93, 247– 319. 195. See chap. 2, “The Other Is Same.” 

  196. Cf. the words of the Scythians’ neighbors: προτερον αδικ ησαντες Περσας κα ι   αρξαντες πολεµου, etc. (4.119.2– 4). On the Scythian invasion of Asia, see 1.103.3– 106.1, 4.1.1– 2. 197. See 1.66– 68, especially 68.6; Immerwahr 1966, 200– 206.

Interpretation and Evaluation

213

challenged the Spartan claim to superior arete, Demaratus differentiates between readiness and initiative in warfare: “I do not promise to be able to fight against ten men at once or against two; if it were a matter of  τε ειναι], I would not even fight against one. But if there choice [ε
κων

were necessity [αναγκα ι η] and if some great trial urged me on, I would like nothing better than to fight against one of those men who say that they are each a match for three Greeks” (7.104.3). The anankaie Demaratus is here talking about compels men to defend their freedom with spears and even with axes and drives the Spartans at Thermopylae to fight with their hands and teeth when they lose their swords (7.135.3, 7.225.3). If a war imposes anankaie on others, by contrast, though it may be undertaken for a variety of more or less subjectively valid motives, it lies outside of the realm of necessity. In objective ethical terms, it is an unjust war, and in utilitarian terms, it is unwise because, as Artabanus says to Xerxes, it exposes one to danger “without necessity” [µηδεµιη ς 198

 γκης ε ουσης].  ανα Sparta’s reluctance to make expeditions almost leads that city to underestimate the Persian danger. At least twice in the sixth century, Sparta plans or undertakes hostile invasions of Attica but does not carry them out in the end.199 Herodotus’ portrayal would have reminded his audience of the striking change in policy that determined the outbreak of the Peloponnesian War. Previously, says Thucydides, the Spartans would avoid getting  ζωνται, Thuc.  µ η αναγκα

into wars unless compelled by necessity (ην 1.118.2). But in 431, everyone was eager to fight. With a much lower

 γκη, Thucydides attributes threshold than Herodotus for invoking ανα compulsion to both sides.200 For Herodotus’ Athenians, perceiving the difference between defense and aggression has been difficult since the beginning of democracy. Their newly acquired excellence is described in terms that suggest an agonistic effort vis-a-vis ` their fellow Greeks.201 “Extinguishing the hubris” of 198. 7.10δ1. Artabanus reiterates the thought in positive terms at 7.18.3. On Herodotus’ notion of a necessary war as a defensive war, see Munson forthcoming. 199. These Spartan missions include the abortive invasion of Cleomenes, both “unjust” (5.75.1) and “inglorious” (5.77.1); and the projected reestablishment of Hippias (see 5.91– 93, especially 5.93.1, with allusions to later events). The list of 5.76 recalls three other interventions, one mythical and two beneficial. For Spartan reluctance to go out against the Persians, see 1.82– 83, 152; 5.50; 6.106; 9.6– 8. 200. Thuc. 2.8.1; cf. 1.80.1 with 1.23.6, and see Ostwald 1988, especially 1– 5. 201. Herodotus says at 5.78, “they became by far the first.” Cf. Miltiades’ words at 6.109.3. See Stadter 1992, 801.

214

Telling Wonders

Boeotians and Chalcidians, who had attacked them first, entails “crossing over” [δια βαντες] to Euboea and establishing a cleruchy in Chalcis, an expansionistic move.202 The subsequent expedition of the Athenians in support of their Ionian kin, who in turn were fighting for their own freedom (5.97), blurs the distinction between defense and offense; according to the previously examined gloss on the three generations of kaka (6.98), it also leads from war against Persia to war among the Greeks. Herodotus’ generalized statement that war is an evil occurs precisely in a gloss that refers proleptically to the transition from the defensive to the offensive stage of the Persian Wars, after 479. The passage also represents the Athenians as causing a concomitant transition from war against a foreign enemy (polemos) to struggle within Greece (stasis). A. For there had been a discussion at the beginning, even before sending the embassy to Sicily concerning the alliance, that the Athenians should be in charge of the naval force. But when the allies objected, the Athenians yielded [εικον], because they were extremely concerned that Greece should survive and knew that  if they quarreled about the command [ε ι στασιασουσι περ ι τ ην η
γεµονι ης], Greece would perish,  and right they were [ο ρθα νοευντες]: A1. for internal struggle is a greater evil than a war fought in agree ment by as much as war is a greater evil than peace [στα σις γαρ  ε στι οσω    ! κα κιον " ! εµ␸υλος πολεµου ο
µο␸ρονεοντος τοσουτω   πολεµος ε ι ρηνης]. Being aware precisely of this fact, they did not make opposition, but yielded [εικον], B. so long as they very much needed the others, as they [subse
διεδε " 

ων,  ως  quently] showed [µεχρι οσου κα ρτα ε δεοντο αυτ ξαν]: for when, having repelled the Persian, they began to fight  for his territory, putting forward as a pretext [προ␸ασιν] the

ι λhubris of Pausanias, they snatched away the leadership [απε  η
γεµονι ην] from the Lacedaemonians. οντο τ ην But this happened later. (8.3.1– 2)  202. 5.77.1– 4. On the verb διαβαι νω, see chap. 2, n. 257. The term κληρουχους may be an anachronism (see French 1972, 18), but it evokes contemporary Athenian imperial policy.

Interpretation and Evaluation

215

Herodotus’ discourse here appropriates and modifies the contemporary anti-Athenian topos, such as we find in the speech of the Thucydidean Sthenelaidas, comparing the “good Athenians before” the Greek victory against the Persians and the “bad Athenians after.”203 Section A complements the praise at 7.139.5, where the Athenians “decided that Greece should survive and be free,” with the added element of Athenian compliance for the purpose of avoiding stasis (εικον, repeated in section A1). The transition between “good” and “bad” Athenians begins in section B,  " 

ων  [so long as they very much where µεχρι οσου κα ρτα ε δεοντο αυτ needed them] isolates within the altruistic motive just attributed to the Athenians (the safety of Greece) the motive of self-interest (the safety of
δι εδεξαν,  Greece and therefore their own). ως [as they showed], is a gloss of evidence marking the retrospective interpretation that the Athenians’ Panhellenism at the time was strategic and that their behavior after the Persian danger had passed proves their loss of concern in avoiding stasis. This behavior is connoted as a high-handed usurpation of the command

ι λοντο]; cf. 1.14.1), achieved by exploiting a situa(“snatched away” [απε tion that provided the Athenians with a legitimate official claim  (προ␸ασιν). But the passage is less clear-cut than this exegesis makes it sound. First, “the hubris of Pausanias,” though mainly focalized by the Athenians through their insincere claim, is presented as an uncontested fact. It brings the ethical code to bear to the detriment of the Spartans. Second, the fuzziness of the text is evident in the controversies of modern critics, who (not unlike Herodotus’ listeners, one presumes) have not all been equally inclined to read anti-Athenian sentiments into this passage. Thus, Pohlenz  " and Immerwahr translate µεχρι οσου at the beginning of section B not with “for so long as” but with “until”; they argue that the subject of  ε δεοντο and all the other verbs that follow must be “the allies” (implied

ων,  the object of from “the allies” in section A); and they regard αυτ

ε δεοντο,  as referring, not to the allies in 480, but to the Athenians. In this rendering, it is the allies (i.e., the members of the alliance later on, even though evidently not the same states as the allies in section A) who at one point “very much needed” the Athenians because the behavior of Pausanias had caused a crisis in the leadership. The allies therefore publicly 203. Thuc. 1.86.1. Because of its comparative element, this topos is an important subtext in Herodotus. See, in the present chapter, n. 109 and corresponding text, as well as discussion under “Disputes, Arbitration, and the Subjectivity of Opinions.”

216

Telling Wonders

 cited the hubris of Pausanias (προ␸ασιν, then, would merely point to the public nature of the protest); they took the leadership away from the Spartans and gave it to the Athenians instead.204 This is not the most natural interpretation of the passage.205 One would be almost certain that a contemporary audience would have understood it in the other way, if it were not for the words Thucydides uses when he reports how the Athenians justified the origin of their imperial rule on the basis of this change of leadership in the war against Persia. In a passage cited by Pohlenz in support of his interpretation of Herodotus’ gloss, the Athenians in Thucydides say that they obtained their empire (arche) not by

ων  force but because “the allies came to us and asked us themselves [αυτ 206   δεηθεντων] to become the leaders [η
γεµονας]. In light of a familiar contemporary pro-Athenian argument that the Ionians had asked/needed/  wanted (verb δεοµαι) the Athenians to replace the Spartans in the leader  ship, Herodotus’ κα ρτα ε δεοντο, with implied subject and object, does not express all that clearly who needed whom, and the obscurity tempers the directness of the anti-Athenian tone of the passage.207 The ambiguity is very much in keeping with the discourse of Herodotus, who, when it comes to certain issues and cities, tends to praise openly (as he here praises the Athenians for yielding the command of the fleet in the war of resistance) and to blame allusively. This gloss, at any rate, stresses not so much the issue of subsequent Athenian wrongdoing as the generalization on war and stasis, the value of which the Athenians might have continued to recognize but evidently did not. The narrator proffers it in his own voice to establish a bridge between the past of the narrative and the present of the narration, between past war and present stasis. The generalized statement is not here based on specific evidence presented in the Histories, as in other cases; rather, it derives from the current ex204. See Pohlenz 1937, 170– 72; Immerwahr 1966, 220– 21 and n. 87. 205. See especially Strasburger 1955, 20 and n. 4. 206. Thuc. 1.75.1. Here the story of the hubris of Pausanias is a subtext; cf. 1.77.6 in the same speech. Badian (1993, 130) calls this “the foundation myth of the Athenian empire.” Fornara and Samons (1991, 84) follow the regular translation of Hdt. 8.3.3 and remark on its points of contact with Thucydides’ account of the foundation of the Delian League (1.96.1).  207. Immerwahr (1966, 220– 21 and n. 87) maintains that καρτα ε δεοντο in Herodotus means “[the allies] needed very much” and not “[the allies] asked eagerly,” but I do not think that the shift in meaning between Herodotus and Thucydides is crucial here. The point is that, according to Immerwahr’s interpretation, Hdt. 8.3.2, like Thuc. 1.75.1, suggests that the idea of Athenian leadership came from the allies, not the Athenians, and that the Athenians did the allies a favor at the time.

Interpretation and Evaluation

217

tratextual experience of narrator and audience. In “if they quarreled about the leadership, Greece would perish” in section A, the verb στασια σουσι refers to the quarrel that would cause the allies to withdraw from the resistance, as they had threatened to do (8.2.2); the η
γεµονι η (leadership) that causes the quarrel is the military command in the war against Persia. The ruin of Greece will come at the hands of the Persians if the Greeks do not keep a united front. In the generalization, however, στασις (quarrel) heightens its impact because the comparison places it  within the semantic field of πολεµος (war): stasis emphulos is worse than polemos by as much as polemos is worse than peace. The ruin of Greece that derives from stasis is therefore brought about not by the Persians on account of Greek divisiveness but by the Greeks themselves fighting with one another and fighting not simply with words and withdrawals. The term η
γεµονι η similarly assumes a larger meaning: the divisiveness over a military command stands for an internal war for arche or supreme power. The transition from the defensive war of the Greeks “of one mind” against Persia in the past to the war of the Greeks against each other in the present passes through the offensive stage of the war against Persia, when the Greeks [turned the conflict into one about [the king’s] terri αγ 

ωνα   tory” [περ ι της ε κει νου ηδη τον ε ποιευντο] (8.3.2, in B). This transition is characterized by an escalating degree of kakon (evil) as well as by a decreasing degree of necessity in the Herodotean (or “Spartan”) sense of the term. As at 6.98.2, the kaka deriving from the Persians lead to the kaka caused by the Greek leading states fighting for primacy. Disputes, Arbitration, and the Subjectivity of Opinions The gloss on the evil of war and internal struggle shows that the notion of stasis encompasses verbal and diplomatic infighting as well as armed internecine conflict. Herodotus’ text rarely mentions the contemporary ideological or violent struggles of the Greeks at the time of performance, but it either symbolically represents or literally echoes them by reporting, at different narrative levels, a number of verbal controversies. A final group of gnomic statements in the Histories requires that we examine the way in which the narrator positions himself in the midst of different claims and how he is able to apply his experience as histor/researcher to a role of histor/arbitrator and composer of quarrels.208 208. Herodotus’ stance as judicial arbitrator has been examined especially by Nagy (1990, 250– 73, 314– 22).

218

Telling Wonders

Instances of stasis in the Histories fall into two main categories, narrative and metanarrative. The first include, besides war among or within Greek city-states (e.g., that between Athens and Aegina), several episodes of Greek divisiveness and disputation. These primarily occurred during the Persian Wars and concerned four related topics: Medism, strategy, behavior in battle, and command or other privileges.209 The various areas of friction overlap. Moreover, since in some cases, nonbelligerent stasis in the past either results in military conflict or raises the specter of future armed conflicts, it serves as the overarching symbol of stasis in the belligerent sense at the time of narration.210 A second category of stasis is represented by the conflicting traditions at the time of narration concerning actions of the time of the narrated. These metanarrative quarrels connect past and present in a literal way, because they constitute the verbal backdrop of active contemporary struggles. As invisible narrator of the past, Herodotus often paints an uncomfortable picture of the highhandedness, self-interest, and agonism of the Greeks and even implicitly dispenses blame. But when he appears in the text as researcher of current traditions or as judge of past actions, he records the quarrels, presides over them, absolves the actors involved, and dispenses praise. The passage I last considered establishes the continuity between diplomatic stasis of the Greek states at the time of the wars against Persia and belligerent stasis in the subsequent period; it also makes an implicit reference back to the episode of the Greek attempt to enlist Gelon of Syracuse in the cause of Greek freedom (see 8.3.1, section A: “even before sending the embassy to Sicily”). Among the narratives of quarrels, this one contains the most striking signal that in the symbolic code of the logos, the divisiveness of the Greeks in the world of the narrated stands for their subsequent state of war against one another. When the Spartan and Athenian envoys both refuse to give Gelon a share of the command of the Greek forces, he declines to participate in the resistance altogether. Just as Spartans and Athenians have argued their respective rights to leadership going back to the heroic age on the basis of ancient texts,211 209. On the theme of divisiveness among the Greeks, see Immerwahr 1966, 225– 35. 210. When the Athenians prospect an alliance with Xerxes, they allude to attacks on the other Greeks (see 9.11.2). The pact of the confederate Greeks included the imposition on the Medizing states of a tithe for Delphic Apollo to be collected after the war (see 7.132.2). Collections of war indemnities tend to lead to military aggression (see, e.g., 8.111– 12). 211. See 7.159 (reproducing Il. 7.125; see How and Wells 1928, 2:197), 7.161.3 (cf. Il. 2.552).

Interpretation and Evaluation

219

so Gelon replies with an unself-conscious quotation to a (for him) future text. Apparently the Greeks have only people who want to be 

commanders/rulers (αρχοντας) and no one who will be ruled (αρξ οµενους), he says; very well, then, the envoys should go back and report to Greece that “the spring has been taken out of the year” (7.162). The narrator interprets “the spring” as referring to Gelon’s army, which the Greeks have forfeited by their inflexibility, but his gloss merely draws attention to the oddity of the metaphor in this context (7.162.2). More appropriately, Pericles used the phrase, according to Aristotle, to eulogize the young men of Athens fallen in war and snatched from the city like the spring from the year.212 We do not know what was the occasion of this funeral oration—the Samian War of 440 has been suggested as a possibility. By appropriating the metaphor, at any rate, Herodotus guides the listener to establish vertical analogy with an extratextual context, where Gelon, who is here at the center of the dispute and plays his thematic role within the monarchical model, is no longer relevant. The point of the narrative, at the level of meaning I am talking about, is that the mythical and rhetorical themes of hegemonic propaganda, here mobilized by the various contendants to establish their primacy over the others, are an ominous symptom of discord; that they will recur in the political discourse, fueling later inter-Greek conflicts; and that these conflicts, in a much more serious sense than political speeches can express, will “take the spring out of the year” for Greece.213 Herodotus’ unmarked allusions to the public discourse of his time are likely to be more extensive than we are able to realize, given the incompleteness of our evidence. Thucydides’ narrative and speeches represent, nevertheless, a valuable source for capturing the cultural codes shared by the narrator of Herodotus’ history and his audience. Thus, for example, in one of the metanarrative disputes, the way in which the Athenians tell the story of their expulsion of the Pelasgians from Attica seems to reproduce their self-justifying argument that their empire deserves praise for not being as oppressive as it could be.214 In another narrative, a “great 212. Arist. Rhet. 3.10.7 (1411a) and 1.7.34 (1365a). How and Wells (1928, 2:198), following Wesserling, consider the gloss an interpolation on the grounds of its inappropriateness to the Herodotean context. 213. See Fornara 1971a, 84. 214. See 6.137.4; cf. the Athenians in Thuc. 1.76.3– 4, applicable to specific cases of the type of Thuc. 2.70.3 or 3.50.

220

Telling Wonders

wrangle of words”215 breaks out at Plataea concerning whose contingent has the right to occupy the left wing of the army, Athenians or Tegeans. The two contendants enumerate “deeds old and new” (9.26.1; cf. 9.27.1) that make them worthy of being deemed the best. The Athenian speech mentions past services to Greece as a whole and actions taken in defense of particular groups of Greeks against the alleged hubris of other Greeks. The list conforms to the manner of fifth-century hegemonic rhetoric.216 Also, the rhetorical topoi of this speech reproduce current conventions. The Athenians begin (I am paraphrasing), “We are gathered here to fight, not to make speeches, but since the Tegeans have challenged our primacy, we must answer” (9.27.1). Similarly, on the eve of the outbreak of the Peloponnesian War, the Athenian ambassadors at Sparta say, according to Thucydides: “We are not here to argue with your allies but on other business. However, since they have raised an outcry, we wish to show that we have a right to our possessions” (Thuc. 1.73.1). The dismissal of remote deeds as secondary to recent merits in the Persian Wars is also typical (cf. the Athenians at 9.27.5 and at Thuc. 1.73.2). In the speech of Herodotus’ Athenians, moreover, the transition between the mythical and the historical exploits ironically suggests the possibility of an ethical change: “But it does no good to recall these things [i.e., the defeat of the Amazons and so on]: for men who were excellent [χρηστοι ] then could  be no good [␸λαυροτεροι] now, and men who were no good then could be better now” (9.27.4). These words parallel those of the Thucydidean Sthenelaidas that, as I have already conjectured, seem to be based on a current anti-Athenian argument: “Yet, if they were good/courageous

[αγαθο ι ] then against the Medes and are bad [κακοι ] toward us now, they deserve double punishment because from good they have become bad” (Thuc. 1.86.1). By combining different themes of later political rhetoric in his representation of hegemonic quarrels among Greeks during the battle of Plataea, Herodotus’ logos makes reference both to the continuity and to the reversal of circumstances in the historical transition from the polemos of the narrative to the contemporary stasis. In other cases, the evocation of later events emerges from the way in which the narrative describes the actions of the specific protagonists. On  is also applied to the verbal quarrels of the Greeks at

215. 9.26.1. The word ωθισµ ος 8.78, but it denotes the melee of combatants in a physical battle at 7.225.1 and 9.62. See Immerwahr 1966, 274. 216. See 9.27.2– 5; Loraux 1986, 67– 75. One of the deeds is the Athenians’ defeat of the Amazons, on which see chap. 2, “The Other Is Same.”

Interpretation and Evaluation

221

the eve of Xerxes’ invasion, the Corcyreans contrive to remain neutral even though, as they themselves claim, they possess “a not inconsiderable number of ships, in fact the greatest number after the Athenians” (7.168.3). The emphasis on Corcyra’s selfish neutrality and its sea power here recalls how, according to Thucydides, Corcyra became a cause of discord in Greece later on: when the city saw fit to abandon its traditional policy of isolation, the size of its fleet gave it the bargaining power for obtaining an Athenian alliance.217 The point of Herodotus’ narrative, in other words, is that Corcyra’s fleet, which recently did so much to exacerbate the internal conflicts in Greece, could have helped the Greeks at the time of their defensive war against Persia but deliberately failed to do so, just as the city that had been fence-sitting in the hour of need for Greece later came out of its famous neutrality only to contribute to the outbreak of the Peloponnesian War. In the stasis narratives I have examined, the invisible narrator links instances of past controversies among the Greeks with the extratextual developments close to the time of narration.218 In a “metanarrative stasis,” by contrast, the dispute that historically and symbolically corresponds to other manifestations of enmity does not or does not only belong to the world of the narrated but has already spilled over into the world of the performance. These cases are brought to the surface of the text as the histor/researcher visibly attempts to find out and report what happened: “From this point on [in the narrative], I cannot exactly describe [present tense] which of the Ionians were [past tense] cowardly or brave men in this naval battle; for they accuse one another [present tense:

ηλους   καταιτιωνται]”  αλλ γαρ (6.14.1). The narrator’s lack of knowledge with respect to distant events reveals the interconnectedness between the characters of the logos and its sources, both apparently the  subject of καταιτιωνται, “they accuse.” The disunity of the Ionians at Lade continues in the mutual accusations of their descendants. The display of historie (inquiry) throughout Herodotus’ text communicates the important teaching that a variety of subjective factors motivates 217. For the Corcyrean affair, see Thuc. 1.24– 55, especially 1.32.4 (policy of isolation); 1.33.2, 1.36.3, 1.44.2 (Corcyrean fleet). 218. See also 8.56– 64, 75 (Themistocles vs. the Greeks: see Munson 1988, 101); 8.11– 12 (see n. 175 in the present chapter); 8.141– 9.11 (Athenians and Spartans: see Fornara 1971a, 84– 86; Raaflaub 1987, 240). Also, 6.108 should be read against the background provided by Thucydides’ Plataea narrative (Thuc. 2.71– 78, 3.20– 24, and especially 3.52– 68).

222

Telling Wonders

people’s logoi. Characteristic of the last tract of history treated by Herodotus, however, is the fact that the controversies are internal to Greece, and the problem of uncertain or contradictory reports is connected with the contemporary ideological struggle that accompanies the military struggle of the Greek city-states. Herodotus’ role as investigator and recorder of the past coincides more than ever, therefore, with a role of political arbitrator and judge. His choice between competing versions is almost bound to be equivalent to a verdict on what are the aitiai (causes/accusations/guilts) and who is aitios (responsible) in a judicial and moral sense. The defendants in these cases are not paradigmatic figures of the past, like Croesus, but Herodotus’ fellow Greeks at the time of performance.219 This perhaps explains why a negative verdict so rarely occurs. When Herodotus displays to his audience which logoi are available concerning events of contemporary political significance, he does one of two things: either he reports discrepant versions with glosses of sources and lets them stand or fall on their own merit, or he rejects or avoids corroborating those logoi that amount to indictments.220 Acquit, praise, or suspend judgment, but never convict: we may call his explicit and official judicial practice. The categorical conviction of Ephialtes at Thermopylae is the exception that proves the “no explicit conviction” rule and helps to illuminate its meaning. The judicial character of the indictment illustrates the distinction between utterances that “make the words match the world” by representing what is the case and those that “make the world match the words”: “It was Ephialtes who led the Persians around the mountain by  the path, and this man I write as the guilty one [τουτον αιτιον γρα␸ω]” (7.214.3). Herodotus’ interpretation of culpability does not represent facts as much as it establishes the record, whose concreteness and permanence he conveys by the performative verb γρα␸ω.221 As Herodotus’ own report of a variant version reveals (7.214.1), and as we can infer from the historical situation at Thermopylae, the identification of Ephialtes as the 219. Nagy (1990, 250– 73, 314– 21) examines ancient evidence concerning interpolis arbitrations and shows the juridical vocabulary and correspondence with Herodotus’ vocabulary of historie. 220. See, e.g., 5.85– 86 (logoi of Athenians and Aeginetans about their early hostilities); 6.137 (logoi of Hecataeus and Athenians about expulsion of the Pelasgians from Attica); 6.121– 31, especially 6.121.1 and 6.123.1 (accusation against the Alcmaeonids rejected: see chap. 4, “Wonder and Disbelief,” in the present book). See also 5.44– 45, especially 5.45.2. On 1.70.2– 3, see Lateiner 1989, 80; Cooper 1975.  221. See Searle 1982, 3– 4. Cf. the judicial meaning of γρα␸ η.

Interpretation and Evaluation

223

traitor would have been far from definitive.222 But the narrator singles out in the logos an ordinary individual whom an impartial judicial body representing different cities of Greece had already officially convicted (7.213.2); he had subsequently fled under the weight of the aitie and died. His death, though unrelated on the human level, had suitably struck him after his crime, with the sort of timeliness Herodotus elsewhere regards as divine.223 By excluding all other possibilities, Herodotus contains the damage, resolves the controversy, and closes the case. His indictment of Ephialtes stems from the same stance as a reconciler that he outwardly assumes when his historie calls on him to arbitrate quarrels among the Greek city-states. Yet the text of the Histories encodes two different and competing messages: it implicitly exhorts all Greeks to put aside their mutual aitiai (accusations) at the same time as it forces them to face up to the various aitiai (guilts) attributed to them. Herodotus’ task of reconciling quarrels goes hand in hand with his role as their inexorable reporter in a discourse that entails pitting the Greeks against each other. The Salamis narrative includes a remarkable Athenian rumor according to which the Corinthians fled at the beginning of the battle and were only persuaded to return by a divine apparition after having remained absent for the entire time of the action (8.94). The narrative is in indirect speech with intrusive oblique infinitive224 and is followed by the tradition on the Corinthian performance from two other sources. The Corinthians themselves, first of all, do not agree with the Athenians and believe they have been “among the first” in the naval battle; the rest of Greece, moreover, bears witness  to the Corinthians’ claim (µαρτυρεει, 8.94.4). The interrogation of witnesses, in other words, discredits the Athenian story and acquits the Corinthians of cowardice. But what is the function of this type of inclusion in the Histories? We have seen contexts when Herodotus reports logoi he vigorously refutes to expose the ignorance or bias of his sources.225 Here, however, in light of Adeimantus’ bitter opposition against fighting at Salamis (8.59– 61), and in the absence of more detailed evidence of Corinthian prowess in 222. See How and Wells 1928, 2:215; Burn [1962] 1984, 413. 223. See “Divine Retribution” earlier in this chapter. 224. This is a syntactical device by which the narrator distances himself from the received logos. See Cooper 1974. 225. See, e.g., Herodotus’ refutations of Greek traditions concerning the Egyptians (discussed under “Strategies of Evaluation” earlier in this chapter.

224

Telling Wonders

Herodotus’ narrative of the battle, the Athenian story does damage the memory of their performance.226 At the same time, the story implicitly convicts the Athenians of slander and alludes to their current enmity with Corinth. The entire Salamis narrative bristles with episodes of disunity that establish a continuum between past and present discord.227 Though no one is explicitly accused or blamed in the voice of the narrator, the logos implicitly underlines the shortcomings of all the Greeks. With the story about the alleged Corinthian cowardice in mind, we should revisit the episode of Corcyrean Medism.228 The Corcyreans promise help to the Greeks with a “fair-faced” speech, but they have “other plans” (7.168.2). Though they send sixty ships, they hold them off the coasts of the Peloponnese, watching to see how the war will turn out. Visa-vis ` the Greeks, they plan to claim that the Etesian winds prevented their fleet from rounding off Malea (7.168.4). This mind-reading narrative is enriched by a speech introduced by a purpose clause and is never in fact uttered, but reported in direct discourse (an unparalleled device), with which the Corcyreans intended to court Xerxes’ favor if he had been victorious (7.168.3). What makes the Corcyrean episode especially instructive is the deliberate way in which it excludes the form of discourse that puts historie on display. The sequence on the Corinthians at Salamis explicitly mentions two versions and three sources: a protagonist source, a minority source (the Athenians), and “the Greeks.” Similarly, the Corcyrean narrative embeds three sources and two different versions of what happened: (1) the Corcyreans say that they tried to reach the rest of the Greeks but were frustrated by the Etesian winds (called an “excuse” at 7.168.4); (2) the Greeks accepted the Corcyrean version of events (the Corcyreans “misled the Greeks”; see 7.168.4); (3) a nameless source (or Herodotus himself)  still considers the Corcyreans aitioi (cf. α ι τιωµενων, 7.168.4) and tells a different story (corresponding to the direction of the whole at 7.168.1– 4). Because in this form of discourse the unequivocal condemnation of the Corcyreans would have depended on the histor’s endorsement of 226. Plutarch (De Malign. Herod. 39 ⫽ Mor. 870B– 871B) quotes several inscriptions that celebrate Corinthian valor at Salamis and says that Herodotus’ practice of telling damaging stories only to discredit them is in fact a device for slander. See, e.g., 3.56; cf. De Malign. Herod. 27 ⫽ Mor. 863A. 227. See 8.59– 62 (quarrel over strategy); 8.84.1– 2 (metanarrative controversy on who started the battle); 8.85 (Ionians vs. mainland Greeks); 8.92 (Themistocles vs. Polycritus). 228. See 7.168, discussed earlier in this section.

Interpretation and Evaluation

225

version 3, he withdraws from this capacity and is replaced by a semi– invisible narrator who reports motives and intentions as facts. Throughout the last four books of the Histories, the fluctuation of the discourse between a minimally narrated narrative and the form of historie/dispute allows the text to show Greek failings while preserving the visible narrator as benevolent arbitrator, refuter of vicious gossip, praiser of good deeds, and reconciler of the Greeks. This histor is present and conspicuous in a group of narratives concerning the Greek embassy to Argos for the purpose of requesting support against the Persians (7.148– 52). Here the story includes several elements that are also present in the reports of similar unsuccessful Greek missions to Gelon (7.153– 67), Corcyra (7.168), and Crete (7.169– 71). The past divisiveness described by the narrative is moreover concomitant with a current controversy, which emerges metanarratively from the report of four variant versions. A small sampler of Herodotus’ methods for handling sources, this is also the fullest instance of judicial arbitration in the Histories.229 It leads to the formulation of a general principle on the subjectivity of controversial positions. Version A is the protagonist version, which “the Argives say about themselves” (7.148.2). Despite a discouraging oracle, the Argives promised to participate in the Greek resistance on the conditions that they obtain a thirty-year peace with Sparta and at least half of the command of the armed forces (7.148.2– 4). The peace was especially important to them, because their city had lost many men during Cleomenes’ attack and needed time for its children to grow to adulthood; they feared that if  would befall them at the hands of the Persians, another disaster (κακον)
ηκοοι)  they would end up subjected (υπ to Sparta (7.149.1). To the Argive demands, the Spartan envoys replied that the matter of peace would have to be decided by their assembly at home; with regard to the hegemonie, they could only offer the Argive king a share equal to that of the two Spartan kings. Hearing this, the Argives found the greed of the Spartans unbearable; considering that it was preferable to be ruled by the barbari
ιξαι) to the Spartans, they bid the ans than to subject themselves (υπε heralds to depart (7.149.2– 3). Version B is “another logos current through Greece” about an alleged previous event (7.150.1– 3): Xerxes had apparently sent an embassy to urge the Argives to remain neutral and to promise them special regard if 229. See Nagy 1990, 315.

226

Telling Wonders

he were victorious. For this reason, when the Greeks tried to enlist them in their cause, the Argives demanded the arche: they knew the Spartans would refuse, and the Argives wanted to have an excuse (προ␸α σιος, 7.150.3) for remaining at peace. A third story (B1), introduced as “another logos that some of the Hellenes tell about something that happened many years after,” is offered next as a corroborating proleptic appendix to version B. When the Athenian Callias and his colleagues happened to be in Susa “on another mission,” Argive heralds were also present. The Argives asked Artaxerxes whether the friendship they had formed with his father, Xerxes, was still valid, and they received warm reassurances from the king (7.151). What follows is a composite gloss containing the generalization and ending in yet a fourth version (C). 1. Whether Xerxes sent a herald to Argos to say these things [B], and whether Argive heralds went up to Susa and asked Artaxerxes about the friendship [B1], I cannot say exactly, nor do I  express any other opinion [γνωµην] about these things except that which the Argives themselves say. 2. But this much I know, that if all men were to bring together all  κακα] in the middle, with the intentheir own evils [τα ο ι κηια tion of making an exchange with their neighbors, after bending down to inspect the evils [κακα] of their neighbors, each of them would gladly bring back home those that they had brought over to begin with. 3. So, not even the Argives behaved in the most shameful way. And I have the obligation to report the things that are said, but I do not in the least have the obligation to believe them, and let this rule be valid for my entire logos. C. Because also the following is said, that it was the Argives who called the Persians into Greece, because the conflict with the Lacedaemonians had gone badly for them and they wished to have  anything rather than their present grief [λυπης]. (7.152.1– 3) The first statement (7.152.1) begins by declining to corroborate the indictments in versions B and B1. In light of the relative transparency and complexity of the Argives’ own version A, the narrator makes their gnome his own. Among various factors that may have contributed to their (at least potential) Medism, the most compelling appear to be their weakness resulting from the defeat they had suffered at the hand of

Interpretation and Evaluation

227


ηκοοι)  Cleomenes and their fear of becoming subjects (υπ of the Spartans (7.149.1). As we have already seen, to avoid stasis, the Athenians yielded on the issue of the hegemonie (εικον, 8.3.1). For the Argives, however, to
ιξαι, 7.149.2), would yield to the Spartans, even in a limited way (τι υπε have been uncomfortably close to submission. The point of view of the Argives forms the specific context of the maxim at 7.152.2 and of the added-on version C that closes the gloss. Herodotus’ lack of knowledge with respect to the veracity of the various versions is compensated by his experience (ε πιστα µενος) of a more general nature, which he now brings to bear on his interpretation of this case. The passage is difficult, because it combines the caution of historie with the obscurity of the ainos, as Nagy saw.230 At the most immediate level of meaning, the kaka that men compare in the gnomic saying are “misfortunes,” like the kaka Herodotus elsewhere attributes to war.231 The specific primary referent of the phrase oikeia kaka (domestic misfortunes) is represented by the Argives’ defeat by Sparta (called a kakon at 7.149.1) and their current vulnerability.232 The narrator implies that whatever the Argives did, they were acting under the pressure of their misfortune, a pressure made more severe by the subjective (and exaggerated) perception that people have of their own misfortunes in the absence of a suitable “marketplace” opportunity for revising that perception. Seen in this light, the behavior of the Argives is not as culpable (or “shameful” [αισχιστα]) as it could have been (7.152.3). This evaluation leads the researcher/ arbitrator to illustrate in version C a theoretically possible behavior from the side of the Argives that would be even more shameful than anything prospected so far but that would be similarly motivated by their previous  misfortune and their own perception of it (λυπη). But again, because something is circumstantially plausible and is said does not mean that it really happened. To the subjectivity of people’s motives for action corresponds the partiality of the accusations. Between the reports of versions B and B1 and the even more damaging version C, Herodotus recalls two rules that govern both his scientific/historical inquiry and its judicial counterpart. First, sources need to be heard. Second, 230. Nagy 1990, 315, citing Aesop’s fable “The Two Packs” (266 Perry) as a parallel to Herodotus’ formulation. 231. See especially 6.98.2, 8.3.1. This interpretation is supported by Legrand (1946, 7.151) and others. 232. Cf. especially 6.21.2. Threat of war and conquest is also implied in πα␽εα . . .  παθεα at 1.153.1. ο ι κηια

228

Telling Wonders

the logoi have only inconclusive evidentiary value, for they are likely to be politically biased in a historical context, just as they are culturally determined in an ethnographic one.233 In this particular case, then, the researcher/arbitrator’s self-referential glosses acquit the Argives by appealing to the unreliability of the logoi, just as his referential glosses of interpretation and evaluation at least excuse their behavior on the basis of the subjective validity of their own perception. The reading I have just given is, however, one-sided. Understandably, Plutarch appears more exasperated than usual in the face of this passage
ες

εν  υγι  αλλ

α παν  and applies to it the unforgettable verdict
Ελικτα κουδ 234  περιξ [all twisted, nothing sound, all back to front]. One reason why our interpretation does not fully satisfy is that it requires that the main point of the generalization at 7.152.2 must be the implicit idea “[People’s behavior should be excused because] men have an exaggerated perception of their own misfortunes,” rather than the explicitly described stage when one would hypothetically compare one’s kaka with those of one’s neighbors and decide to keep them after all. For the comparative operation to confirm or produce, rather than correct, the subjective perception of “all men,” the kaka each person brings to the middle must be “wrongdoings.”235 Then, the meaning of the generalization for the specific case is, first, that the Argives do not consider what they did all that bad in comparison to the bad actions of others. Second, those ready to convict the Argives of the most serious wrongdoings are under the same subjective perception: they are excessively harsh toward others and underestimate their own wrongdoings. The subjectivity of the source-protagonist is again counterbalanced by that of the accusing logoi. From Herodotus’ objective standpoint, “not even the Argives did the most shameful things,” with αισχιστα in the evaluation picking up on κακα (in the sense of α ι σχρα , “shameful”) in the maxim.236 The suggestion that others besides the Argives have kaka (wrongdoings) to account for applies both to the world of the narrated and to the extratextual world of the performance. In Herodotus’ narrative, The233. Cf. in an ethnographic context, 2.123.1 and 4.195.2, with similar glosses of noncorroboration. 234. Plut. De Malign. Herod. 28 ⫽ Mor. 863E, quoting from Eur. Andr. 448. The translation is by Bowen (1992, 57). 235. See especially Macan 1908, 1.1.209, followed by How and Wells 1928, 2:191. 236. Cf. the sophistic Dissoi Logoi, 90 DK 2.18 and 2.26, where τα α ι σχρα (shameful things) is what is being compared and found καλα (beautiful/honorable) by different people. See “Funeral Customs and Other Nomoi” earlier in the present chapter.

Interpretation and Evaluation

229

ban actions certainly appear more “shameful” than those of the Argives. In the real world of the narration, Argos had settled down to peace and inactivity, in the least culpable position imaginable at least since 451 b.c.237 By contrast, each of “the leading states warring with one another about the arche” (6.98.2) had been exerting itself in the hope of obtaining the help of Persia against the other.238 The Spartan king Archidamus in Thucydides perhaps formulates a typical argument when he says, “For all those who, like us, are the objects of designs from the part of the Athenians, it is not a matter of reproach to provide for their own survival [διασωθη ναι] by acquiring for our side not only Greeks but also barbarians” (Thuc. 1.82.1). For Herodotus, to seek the cooperation of Persia for the purpose of making war against other Greeks is a bad action.239 But in a world where different groups of Greeks rationalize this type of behavior for the sake of their mutual wars, even the most unlikely worst actions of the Argives, who feared Spartan domination, were not αισχιστα, that is, without parallel in their baseness. The maxim at 7.153.2 contains one of three instances of the phrase “all  men” (πα ντες ανθρωποι) in Herodotus’ metanarrative. In one of the other cases, as we have already seen, it appears in the very similar passage that envisions people comparing nomoi (customs) and finally choosing their 237. Argos was at peace with Sparta (see Thuc. 5.14.4) as well as with Athens (from 462: see Thuc. 1.102.4). Cf. Thuc. 5.28.2. Herodotus’ leniency toward Argos contrasts with his negative (though, as usual, unmarked) representation of Thebes (see, e.g., 7.205.3, 222, 233). Correspondingly, on the Athenian tragic stage, Argos, in contrast to Thebes, emerges as a city that “can be saved”: see Zeitlin 1990. 238. See Thuc. 2.7.1; cf. 4.50, perhaps later than Herodotus’ Histories. See How and Wells 1928, 2:190. See also Aristoph. Acharn. 61– 134. For the fateful Spartan embassy to Persia mentioned by Herodotus at 7.137 (cf. Thuc. 2.67), see “Divine Retribution” earlier in the present chapter. 239. Callias’ embassy to Susa mentioned by Herodotus at 7.151 was almost certainly part of the negotiations that eventually led to peace with Persia in 449 b.c. According to Samons (1998, 135), on that occasion Athenians and Argives went to Susa together after stipulating their alliance with anti-Spartan intent in 462/461. If Herodotus were implying as much, his mention of that mission would be another point in favor of the equivalence between the actions of the Argives and those of other Greek powers. However, Herodotus 7.151 perhaps mentions Callias and the Athenians with him to provide a clue to the source of this story, but otherwise clearly implies that their meeting with the Argives at Susa was coincidental. We can also speculate that Herodotus would have approved a cessation of the offensive war against Persia in and of itself (pace Badian 1993, 134), and that his statement that the Athenians were there “on another matter” is designed to dissociate Callias’ mission from the self-serving Argive embassy. On this difficult passage, see also Macan 1908, 1.1.209. The role Herodotus attributes to the Callias family is problematic also at 6.121, a passage we will discuss in chapter 4.

230

Telling Wonders

 own (3.38.1; repeated with παντων [ανθρ ωπων] at 3.38.4). The third case is the generalization that “all men know equally about the gods.”240 We should consider the interconnections among these three generalizations on the topic of beliefs and opinions held by different people as the sign of the contribution of ethnographic historie to Herodotus’ political ideology. The use of the phrase “all men” also links historie in the sense of “inquiry” and historie in the sense of “arbitration.” In each of the three cases, the phrase “all men” refers to a human community whose overall identity both depends on and transcends its internal differentiation. While at 3.38.1 and 2.3.2 the internal differentiation of the entity “all men” from the point of view of practices and beliefs coincides with the subdivision of humankind into different ethnea, the generalization at 7.152.2 implicitly replaces membership in an ethnos with membership in a Greek polis to account for differences in perception. Whether they are misfortunes or wrongdoings, the kaka are chosen, discarded, and evaluated in different ways on the basis of men’s political outlook as citizens of Argos, Athens, Sparta, and so on. The narrator’s statement of cultural relativism at 3.38 is designed to promote a cosmopolitan attitude in the listeners; it encourages them to overcome their cultural subjectivity in the only way it can possibly be overcome, namely, by recognizing its compelling universality as objective proof of the equivalent value of the different nomoi. The maxim at 7.152.2 follows the same model for the purpose of mediating disagreements among the Greeks and to provide the audience with a higher vantage point for judging themselves and others. The histor pursues these goals by means of a verdict that, in a characteristic way, explicitly acquits and implicitly convicts. The narrative of the Histories spares none of the Greeks. In this context, to apply the norm of subjectivity to a political quarrel is to invite the different parties to rise above that norm and reexamine with a more impartial outlook their own “bad actions.” Conclusion The generalizations in the voice of the narrator represent sporadic but precious indicators of the substance of the message that the Histories as a whole wants to communicate to the Greeks. At the highest level of inquiry, we find a tenacious attempt to verify and demonstrate certain truths that 240. 2.3.2. See “Equal Knowledge” and “Funeral Customs and Other Nomoi” earlier in the present chapter.

Interpretation and Evaluation

231

belong to the supracultural theological sphere: the fluctuations of human fortune with respect to individuals and states, the moral participation of the divine in these processes, and the willingness of the gods to share some of their knowledge with humans. The central problem of human misfortune in the overarching context of the Histories is represented by war and the painful confusion between the need of any state to cultivate the values and the resources that will enable it to maintain its freedom, on the one hand, and the agonistic impulse to fight for the acquisition of rule, on the other. Concomitant with and symbolic of the unnecessary armed conflicts of Greeks against Greeks are the verbal and ideological quarrels in which claims, counterclaims, self-justifications, and accusations are all a part of a rhetoric of mutual aggression. Here Herodotus establishes an implicit parallel between Greek ethnocentrism and Greek polis particularism. Just as the Greeks cannot seem to find a balance between the awareness of their own cultural worth and the systematic misunderstanding and disparagement of foreign cultures, so they define their political identity on the basis of a hegemonic polis ideology. The connection Herodotus draws between antagonism toward the barbarian other and antagonism toward other Greeks may seem paradoxical if we consider that Panhellenism “in itself accentuated rather than softened the distinction between Greek and barbarian.”241 The connection is, however, fully justified, for example, in Athenian uses of Panhellenic ideology as an imperialistic tool vis-a-vis ` other Greeks and in the confusion of the contemporary Greek political discourse in general concerning who the other might be.242 Herodotus’ authority rests on his knowledge of the world and on his lack of allegiance to any particular state. From this vantage point, he teaches his audience the ambivalent character of reality and provides them with a model of arbitration for negotiating their differences.

241. Guthrie 1971, 162. 242. Cf. Antiphon DK 87 B44, frag. B: “we are barbarians to each other.” For the relation between Panhellenism and imperialism, see Perlman 1976.

Chapter 4

Thoma

Comparison and interpretation are associated with Herodotus’ overarching goal of explaining. The first sentence of the Histories, however, situates  α ι τι ην [for what reason]) within the explanation (represented by δι ην much broader context of a celebration of achievements great and wondrous (θωµαστα ) across the inhabited world. Celebration and narratability overlap with explanation but refuse to be restricted by it. Explanation provides direction without entirely controlling either the pace of the logos or the paths it will take. Among the advertisements that throughout the work keep reestablishing the autonomy of narratability vis-a-vis ` explanation, words of wonder occupy a special place: they alone name an emotional response the narrator wants the recipient to share.1 This is not to say that in Herodotus’ speech act, words like thoma are always strictly    rationed and precisely targeted.2 But θωµα, θωµα . . . µοι, θωµα µεγιστον, and the verb θωµα ζω are the most emphatic and mysterious directions the narrator provides. They tend to announce special semiotic challenges and occasionally mark the highest philosophical level of the inquiry. The wonder the Histories are designed to inspire is not (at least not openly) self-referential, that is, directed at the logos of Herodotus. Aside from one or two cases when wonder is directed at someone else’s logos, the object of wonder belongs to the world of the narrated, which in turn imitates the real world. Nevertheless, mention of the wonder response may create, from the point of view of the recipient, a wondrous discourse event, something that surprises primarily within the discourse. Thus, 1. On the issue of what wonder is and whether it constitutes an impression registered “in the heart” or “in the brain,” see Greenblatt 1991, 15– 22. For the notion of wonder in Herodotus, see especially Barth 1968; Dewald 1987, 154– 55, 165; Hartog 1988, 230– 37; Payen 1997, 117– 28.   2. E.g., the expressions αξιος θωµατος or θωµα σαι αξιος used as metanarrative qualifiers within narrative statements (see, e.g., 1.185.3, 3.47.2, 3.113.1, 4.199.1, 4.53.3) are    often as casual as αξιοθ εητος or αξιαπ ηγετος, αξιος λογου, and so on. See Barth 1968, 98– 99.

232

Thoma

233

when I examined Herodotus’ description of the leather boats of the Assyrians, I was proceeding backward.3 In cases of this sort, the phenomenon itself perhaps causes less wonder than the text’s injunction that we,  along with the narrator, should wonder (θωµα µεγιστον µοι). This in turn leads to interpretation and, retrospectively, to a redirection of our wonder response to the object. Herodotus presents to his audience both ethnographic and historical items as wonders. In the history, appeals to wonder draw attention especially to exceptional actions or agents—onetime behaviors of individuals and animals or occurrences that appear to reveal the intervention of the  divine. In the ethnographies, a θωµα tends to be a tangible foreign artifact, a phenomenon of the landscape or a feature of the flora and fauna of distant lands. Metanarrative thoma is rarely used to describe either the activities of foreign peoples in the ethnographies or those of foreigners in the history.4 In the context of their respective cultures and through Herodotus’ rhetoric of similarity or equivalence, exotic actions and customs want to be interpreted as normal. Unlike the narrator, characters are frequently in wonder at the behaviors, utterances, or appearances of foreigners, because they are different.5 On one occasion, Herodotus ex pects that some of his listeners will experience great wonder (θωµα µεγιστον) at a Persian action that is too similar, that is, reminiscent of the contemporary policy of a Greek, not a barbarian, superpower.6 He implies no doubt that some may not even believe it; but for those who do, this unexpected event will invite reflection. Whatever is or appears to someone exceptional or strange represents, 3. See my examination of 1.194 in my introduction. 4. At 2.35, the abundance of thomata in Egypt is paralleled, rather than illustrated, by the list of customs that are “opposite to those of the rest of the world.” The song of Linus (2.79.2– 3), the collection of spices in Arabia (3.111.2, 112), and the boats of the Assyrians (1.194) are the only other items termed thomata in metanarrative that have something to do with customs. As Payen observes (1997, 118), none of the Scythian customs, e.g., as strange as they may seem to the reader, belong to the semantic register of thoma: Herodotus denies that there are many thomata in Scythia, and customs are not included among the few (4.82). 5. E.g., the Fish Eaters wonder at the long life of the Ethiopians (3.23.2), Heracles at the Mixoparthenos (4.9.2), the Scythians at seeing the Amazons (4.111.1), Darius at the actions of the Paeonian woman (5.13.1), and Xerxes’ scout at seeing the Spartans combing their hair and doing exercises at Thermopylae (7.208.3). 6. The action is Mardonius’ establishment of democracies in Ionia after the revolt (6.43.3, with back reference to 3.80.1). See chap. 2, “The Sameness of the Lydians”; chap. 3, “Explicit Evaluation,” n. 170.

234

Telling Wonders

at any rate, a stimulus for further thinking. Whether in the ethnography or in the history, once the metanarrative labels any particular fact a thoma, that fact tends to jump out from its narrative context but nevertheless needs a context in which to make sense.7 It is not explained but demands somehow to be explained and to participate, in its turn, in the text’s network of explanations. It provides an impulse to mental inquiry, much as, in Aristotle’s formulation, wonder provides the impulse to philosophy.8 It is an inquiry, however, that the text declines to actualize but implicitly identifies as the task of the recipient. Herodotus and the Conventional Code of Ethnographic Wonders Scholars have not sufficiently explored Herodotus’ notion of wonder. The tendency has been rather to regard wonder as a self-justifying response in the face of what is unusual according to commonsense Greek assumptions of normalcy. The use of wonder words in the Histories even represents, according to Hartog, not so much the narrator’s signal for the audience to share his response to an object as Herodotus’ own response to audience expectations of the most conventional sort. This topos of traveler tales of all ages, he argues, was commonplace also in ancient ethnography. Because people like to be amazed, the narrator adopts a rhetoric to bolster his claim that he is qualified to amaze them.9 Hartog’s own analysis of narrative sections of the Histories, including reports of wonders, is clever and profound; but it proceeds against the text and with little regard for Herodotus’ communication of meaning to his audience. Wonder words in the Histories are often signals that something has a special meaning. This is not to deny the existence of a previous ethnographic wonder 7. See Greenblatt 1991, 2– 25, on the “anecdotal” character of all literature of marvels. 8. Arist. Metaph. 1.2.8– 9 [982b]; see also Plato Thaet. 155d. Cf. Redfield (1985, 103) on Herodotus: “Wonder is the beginning of wisdom when it leads to further thought.” 9. “Thoma may be reckoned among the procedures used by the rhetoric of otherness. Generally speaking, the impression it conveys is one of trustworthiness, for the narrator cannot fail to produce this rubric, which is expected by his public. To omit it would be, at a stroke, to ruin his credibility. It is as if it were postulated that far away, in these other countries, there were bound to be marvels/curiosities” (Hartog 1988, 231). Cf. Strabo’s accusation (11.6.3) that accounts of marvels by Herodotus and others pander to the desires of the audience. For wonder as a conventional topos of ancient ethnography, see Jacoby 1913, 331– 32; Jacoby maintains that ethnographic logoi were organized to include a description of the nature of the country, a survey of customs, the mention of thomasia, and finally a section on political history.

Thoma

235

tradition. Though we know too little about it to make a proper comparison, Herodotus’ awareness of this tradition especially emerges from three metanarrative statements that refer to audience expectations by saying that this or that country has or does not have many thomata or thomasia.10 The most expressive of these, for our understanding of Herodotus’ adaptation of the conventional code of wonders to his own celebratory code, is the following introduction.  the land of Lydia does not have that As for wonders [θωµατα δ ε], much suitable for description [ε ς συγγρα␸ ην], compared to other countries, with the exception of the dust washed down from Mount Tmolus. But it offers one building that is by far the biggest, not counting those in Egypt and Babylon. (1.93.1) Placed at the beginning of the sentence and followed by the name of the term θωµατα the country and the programmatic word συγγρα␸ ην, here functions as one of the summarizing elements in a title more or less equivalent to “Description of Lydia: Wonders.” But as it turns out, this announcement belongs to a programmatic correction. Not much else fits the topic in the normal sense, says the narrator, and not all available items are interesting. The absence of thomata in Lydia corresponds to this country’s cultural similarity to Greece. Difference from Greece is, then, one expected re quirement for θωµατα of foreign lands.11 “The dust from Tmolus” would qualify in this sense, but Herodotus mentions the phenomenon as notorious and immediately moves on. By the standards of ethnographic narratability, this golden sand washed down by the river Pactolus presumably belongs to the same order of phenomena as other natural treasures of the East. Herodotus gives a lengthy description of the harvesting of gold in India.12 This follows a pattern similar to that of the harvesting of Arabian spices by various methods, each more “wonderful” than the 10. See 1.93.1, 2.35.1, and 4.82; Jacoby 1913, 331; Hartog 1988, 231. θωµα σιος, even and θωµα,  more than θωµαστος seems to be a technical term in the ethnographic tradition. See Barth 1968, 108. But in Jacoby’s classification (see n. 9 in the present chapter), it is not clear, e.g., to what extent the rubric thomasia would have overlapped with the others in ethnographic writing before Herodotus. 11. See Hartog 1988, 231– 32. For Lydian customs similar to those in Greece, see 1.94.1. 12. 3.102.2– 105. Cf. Herodotus’ interest in the gold of Ethiopia (3.114) and northern Europe (3.116.1).

236

Telling Wonders

other.13 At the edges of the earth, where people live apparently untrammeled by societal structure and in direct contact with their environment, the ethnographer is closest to an anthropological mode of inquiry. The Indian-Arabian sections focus on the ingenuity and labor of man in his natural state, on the coexistence of opposite extremes, and on the relations and correspondences that appear to exist between animals and human beings.14 According to Herodotus’ interpretation, these ethnographic and zoological data indicate that the world of nature in general and particularly the customary behaviors of animals are regulated according to a balance that is intelligent and providential;15 this leads to the implicit suggestion that, at some level, the customary behavior of men translates into the area of nomos by virtue of this providential design. The methods for collecting rare products are perfectly good wonders in conventional terms. But for Herodotus in this “ends of the earth” passage, they also provide additional evidence for defining custom in its essence as paradoxically both natural and divine. This confirms the unitarian view, previously theorized, of “Custom king of all.”16 Lydia, by contrast, is a land of the center. The lesson it ethnographically teaches is not anthropological but political and historical. In this context, Herodotus rejects as not significant the natural phenomenon of the gold from the Tmolus. He even fails to discuss it as the source or emblem of the famous Lydian wealth that impressed the Greeks and that 13. See 3.111.1, 3.112; cf. 3.113.1. On the structural similarity of the methods of collecting spices, see Detienne 1977, 14– 20. 14. The last is a concern that, for different reasons, Herodotus was also able to pursue in Egypt. See chap. 2, “The Texture of Nomos.” 15. See 3.108; cf. 3.106.1. See Immerwahr 1966, 312. 16. See 3.38, again based on a testing case provided by a population of the southeastern edges, the Callatian Indians. See chap. 3, “Funeral Customs and Other Nomoi.” The ingenuity of the primitive harvesters of spices is at one point denoted with the cultured term σο␸ ι ζεσθαι (3.111.3). This term also describes the behavior of cats in the section on Egyptian animals (2.66.2). These two uses of the verb bridge the gap between cultural activity (“devices”) and elementary impulses. Similar theoretical interests govern Herodo tus’ “great wonder” [θωµα µεγα] (3.12.1) at the softness of Persian skulls and hardness of Egyptian skulls on the battlefield of Papremis. He attributes this contrast to the opposite customs of wearing a head covering or shaving the head: whereas his observation of primitives allows him to verify the ultimate derivation of culture from nature, here he notices changes in nature as a result of culture. Herodotus’ most theoretical glosses on anthropological subjects tend to occur in book 3 for two reasons, both connected with the progress of the historical narrative: here the actions of Cambyses, especially in relation to Egypt, raise the question of the absolute value of nomos (and therefore of its connection to phusis); and here the conquests of Darius bring the narrative to the extremities of the earth, a suitable field for such discussion.

Thoma

237

represents an important theme in his own historical narrative about  item the narrator agrees to valorize in this Lydia.17 The one and only (εν) section is the monumental tomb that the people built for one of their kings. This oversized building both qualifies as a thoma in the conventional sense and serves, as we have seen, as a more analytical symbol of Lydian society and its structure, of the single disturbing custom in which Lydia differs from Greece, and of the oppressive monarchy that will ensure the country’s eventual loss of freedom.18 Thus, the introduction to the first properly ethnographic section of our text indicates that Herodotus accepts a canonical aspect of ethnographic narrative, the report of thomata, but will subsume and adapt it to his representational needs. When Herodotus berates those Ionians who maintain that the Nile’s floods are due to its origin from Ocean (2.21), his polemic may be related to this redirection of the notion of ethnographic “wonders.” He calls the theory “more fit for an account of wonders”  [θωµασιωτερη] but also “more inept” [ανεπιστηµονεστ ερη] than a theory refuted earlier.19 Other than factually inaccurate, it belongs to those geographical speculations about the shape of the earth that Herodotus considers ideologically misleading.20 He therefore declares the “won αξι  ω µνησθηναι, drous” logos itself as “not worth mentioning” (ουδ 2.20.1) except to reject it. The oxymoron registers the narrator’s objection to an indiscriminate literature of wonders. The theory that the Nile derives from Ocean has the flaw of bringing  2.23) with no the account into the realm of the invisible (ε ς αφαν ης, possibility for verification. But just as a verifiable thoma may not be a useful tool of representation to Herodotus (the golden dust of the Tmolus is real but uninteresting), so the fact that a wonder is unverified does not automatically disqualify it from being meaningful. The oracle shrine of Apollo and Artemis at Buto, which is “worthy of logos” as a whole, includes two orders of thomata, visible and unseen. The temple in the precinct of Leto, made of a single block of stone, is programmatically ˆ 17. See, e.g., 1.14.3; 1.50.1– 3; 1.51.1– 3, 5; 1.52; 1.92.1. A heap of this golden ψηγµα is presumably what Alcmaeon falls on when he visits Croesus’ treasure house (6.125.4). For gold as an index of royalty, see Kurke 1995, 45– 51. 18. 1.94.7. See chap. 2, “The Sameness of the Lydians” in the present book. 19. Cf. 2.23. For the attribution of the theory to Hecataeus, see FGrHist 1 F 302; Lloyd 1976, 100. 20. Cf. 4.36.2– 45 and especially 4.36.2; at 4.42.1, Herodotus’ rejection of the Ionians’ geographic constructs is phrased in terms of his “being in wonder” [θωµα ζω]. See chap 2, “The Texture of the Earth,” in the present book.

238

Telling Wonders

advertised (␸ρα σω) as “as far as I am concerned, among the things that  can be seen, . . . the one that causes the greatest wonder” [το δ ε µοι των   ␸ανερων θωµα µεγιστον παρεχοµενον] (2.155.3). The combination of wonder “to me” and visibility “to me” is repeated in the conclusion of  των  νυν ο! νηος  ␸ανερωνµοι   the very brief description (ουτω µεν των ε στι θωµαστοτατον,  το ι ρον περ ι τουτο 2.156.1). After the temple, the nearby island of Chemmis, on a lake of the Delta, is a second wonder (2.156.1– 2). Here one finds a great temple of Apollo, three altars, and also “palm trees and many other trees, some bearing fruit, some not” (2.156.3). Although this is still all part of the visible, we also are told with that is, it great verbal redundancy that the island is supposedly πλωτη; floats. The narrator has not seen this phenomenon, but the Egyptians say it does float (2.156.2). They tell a story to explain this, namely, “that Leto came to this island at the time when it did not float, and having received Apollo in trust from Isis, she saved him from Typhon by hiding him in this island that now reportedly floats” (2.156.4). The sanctuary of Buto testifies to the prominence in Egypt of three gods who are also major divinities in the Greek pantheon and are similarly connected to one another in myth and cult. The flight of Leto to the (perhaps floating) island of Chemmis reproduces the wanderings that lead Leto to the floating island of Delos in the Greek tradition.21 The palm trees of Chemmis are singled out for mention because they expose what is for Herodotus the ultimate proof of the mutual interface between different religions, the correspondence of specialized signs.22 No glosses of comparison appear in this passage, but the subtext of the description of the Buto sanctuary has to do with the horizontal and synchronic similarity between Egypt and Greece that Herodotus pursues throughout the Egyptian ethnography. In a context of preexisting relations, palm trees and other features at Buto and Chemmis fulfill the same function as the existence in Egypt of the song of Linus. The latter is hard to fathom as an ethnographic thoma in conventional terms but is for this narrator ζειν µε).23 Buto, however, also encodes  occasion for wonder (αποθωµα the otherwiseness of Egypt, even its oppositeness from Greece. Over 21. See Hom. H. Ap. 3.25– 138. In a late version of the story (Hyg. Fab. 140), Leto is pursued by Python, which would increase the similarity with the Egyptian story told by Herodotus. The correspondence between Greek and Egyptian tradition here accounts for Herodotus breaking his own rule of silence concerning “divine things” (2.3.2). 22. See Hom. H. Ap. 3.18, 113. For the Greek association of the palm tree with Artemis, see Sourvinou-Inwood 1991, 99– 123. 23. See 2.79.1– 2 and chap. 2, “The Texture of Nomos,” in the present book.

Thoma

239

there, palm trees are “among many other kinds of trees, some bearing fruit some not”; here, a single palm tree sprouts in a rocky land.24 In Egypt, Leto is the nurse of Apollo and Artemis, not their mother, and these two are the children of Dionysus and Isis.25 The floating nature of Chemmis contributes to this tension between what is same and what is different: whereas Delos used to float but no longer does (it is in fact by definition akinetos: see 6.98.3), Chemmis did not float but is said to do so now.26 Unlike the Nile’s origin from Ocean, therefore, the supposedly floating Chemmis is a wonder to Herodotus, as it perhaps was to Hecataeus.27 Herodotus’ polemic against Hecataeus, if polemic is implied here, limits itself to an emphatic gloss of noncorroboration that acknowledges the scientific distinction between phenomena that are verified by eyewitness, phanera, from those that are learned through oral reports: “I personally did not see it either move or float, but, hearing this, I am AMAZED [τεθηπα], if it truly is a floating island.”28 In the description of Scythia, a land with no θωµα σια except for the number of its rivers and the vastness of its plain, something that is indeed visible but cannot be what people say it is counts nevertheless as worthy σαι . . . αξιον)   of wonder (αποθωµα and preserves its meaning. They show [␸αι νουσι] the footprint of Heracles, stamped on the rock; it resembles a man’s footprint but has a size of two cubits . . . (4.82) 24. The infertility of Delos is an essential element of the myth of the birth of Apollo (see Hom. H. Ap. 3.48, 55, 60, 72). 25. An opaque gloss of testimony recruits Aeschylus to the system of interconnections  (see 2.156.6). Here the verb ηρπασε both recalls the famous abduction of the Demeter and  Persephone myth (cf. ηρπαξεν in Hom. H. Dem. 2.3, etc.: by snatching the Egyptian logos that Artemis was the daughter of Demeter, Aeschylus has abducted Artemis to replace the Kore of Greek tradition) and participates in Herodotus’ polemic against Greeks who exploit Egyptian cultural knowledge without giving it due credit (2.123.3). For the view that Aeschylus’ version in this gloss has something to do with the substance of the Eleusinian mysteries, see Mazzarino 1966, 97 and n. 124. 26. See especially Pindar frag. 33d SM on the transformation of Delos as a nonfloating  ι νητον τερας). island (ακ On the symbolic significance of the immobility of Delos, see Wood 1972, 141 n. 55. 27. Hecataeus describes Chemmis in remarkably poetic terms: εστι δ ε η! νησος µετα  ρσι η κα ι περιπλει κα ι κινεεται ε π ι του υδατος (FGrHist 1 F 305). 28. 2.156.2. There is no evidence that Hecataeus expressed disbelief of the fact that the island floated. The verb τεθηπα appears only here in the Histories. Like θωµα ζω, it follows  See Smyth 1956, the most frequent construction of verbs of emotion with ε ι rather than οτι. 2247. The narrator is amazed at the fact that the island floats, if it does, not wondering whether it does.

240

Telling Wonders

Scythia’s scarcity of wonders means poverty of material culture in contrast to the powerful kingdoms of the East. The excepted items are natural wonders, the plain and the rivers, elsewhere described at length and representing the elusiveness and insularity of the Scythians, in whose land Darius’ army will roam unproductively and risk being trapped.29 But in relation to the Greeks, as we have seen, what makes Herodotus’ Scythia most insular and foreign does not preclude and sometimes even favors the notion of a special similarity between the two peoples. Elsewhere Herodotus even reports an uncorroborated story “the Black Sea Greeks say,” according to which the Scythian nation was born from the marriage of their wandering hero Heracles to the Mixoparthenos, a local maiden half woman and half snake.30 Here the footprint “they show” (appropriately huge to match the landscape and the hero’s stature) is not a sort of memorial but rather the visual symbol equivalent to that fictional event. In Herodotus’ description of Scythia, it is a pure narrative sign, and as a sign, it represents, from the point of view of the text, a thoma. The traditional topos of ethnographic wonders could provide Herodotus (and no doubt Hecataeus and others, though in ways we do not know) not with material for amazing his audience “ineptly” but with a rich reservoir of symbolic forms. One of the conventional attributes of thoma must have been µεγαθος, size, which is hardly devoid of significance, including in a political sense.31 Already in the archaic period, the Greeks perceived foreign lands, especially the East, as “big” and themselves as “small,” with the related invidious problem that bigger is evidently and devastatingly better—and has the better of you—though in the end “small” turns out to be better and stronger.32 Bigness in nature is dwarfing and is sometimes related to wilderness and fear of the unknown, as in the case of the Scythian rivers and plain. In cultural artifacts, it often signifies wealth, power, or both at once.33  ντων . . . θωµασιωτα 29. See 4.47– 59, 4.99– 101. So the Pontus is πελαγεων . . . απα τος (4.85.2). 30. 4.8– 10. See the indirect noncorroboration at 4.10.3– 11.1. See Vandiver 1991, 172; Georges 1994, 1– 9. 31. Hartog (1988, 234) observes that the quantitative aspect of thoma is confirmed by the frequency with which it is described in terms of measurements. $ κατα κοσµον/ο  σµικρη 32. Phocylides writes (frag. 4 Diehl), πολις ε ν σκοπελω ι κευσα  κρεσσων Νι νου α␸ραινο υσης [A city that is small but on a lofty promontory and well ordered is stronger than foolish Niniveh]. 33. Consequently, it signifies the “greatness” of the builder. See Immerwahr 1960, 265 on physical erga.

Thoma

241

Numerous large buildings in the Histories are recruited in the service of the monarchical code, and a few of these are designated with words of the thoma family. But Herodotus’ signs are more subtle and varied than this simple correspondence. Babylon and its dam built by Nitocris—this last an achievement that is wondrous specifically because it is big and tall     ι , 1.185.3)—are coun(αξιον θωµατος, µεγαθος κα ι υψος οσον τι εστ tered by the θωµα µεγιστον of the Assyrian boats, which are also, peacefully, oversized (1.194.3). In Egypt, the gigantic proportions of Eastern art and architecture are adapted to the grandiosity of nature and testify to the high degree of technical skill of this ancient culture.34 Here Herodotus insists on measurements as part of his polemic against what Froidefond calls the “malevolent snobbism” of the Greeks, who minimize Egyptian things and even shrink Egypt’s territory.35 The pyramids, a product of monarchic oppression, are outranked on the scale of thomata by the labyrinth, a collective monument of twelve just kings, and by the yet more wondrous Lake Moeris; one is the cultural match of the land of Egypt, the other of the Nile.36 Lloyd remarks how many aspects of Herodotus’ attitude toward Egypt are revealed in his description of the labyrinth (2.148)—in his display of sources, admiration of size, and disparagement of the Greeks.37 From a narratological perspective, we should notice how conspicuously the persona of the narrator has invaded his subject. He celebrates, groups, excepts, compares, compares again, and ranks. The labyrinth is “too great for logos” [λογου µεζω], and indeed there is relatively little logos here in the sense of description or “story.” Predominant are glosses of historie, that is, the narrative of the journey of the researcher/narrator, who soon multiplies into “we” for maximum interpretive and narra ι τε torial authority.38 Using all the available verbs of seeing (ε'ιδον, αυτο ! ωµεν, ! ωµεν),   ι θεησα µενοι, αυτο  ι ωρ  ωρ αυτο Herodotus reports that he/ 34. The propylaia of the temple of Athena at Sais built by Amasis are θωµα σια (2.175.1), but Herodotus especially wonders at Amasis’ achievement in transporting the monolithic chamber from Elephantine (µα λιστα θωµα ζω, 2.175.3). 35. On Greek diminutives for Greek things, see Froidefond 1971, 122– 23; Lloyd 1976, 310. For the size of Egypt, see especially Hdt. 2.6– 11, 15– 18. 36. See 2.148– 49; Benardete 1969, 63. 37. Consider 2.148.2: “for if one were to gather all the walls and display of works of the Greeks, they would appear to be inferior in labor and expense to this labyrinth.” See Lloyd 1993, 2:148. 38. See Chamberlain, forthcoming, on the use of “we” not in reference to an ethnographic group but as a projection, operating only on the surface of the text, of the narrator’s definitive voice and authority.

242

Telling Wonders

“they” personally saw at least the upper chambers, fifteen hundred in number and “superior to all human works”; he was not allowed to see the lower chambers but learned by hearsay (λογοισι ε πυνθανοµεθα,  η παραλαβοντες) that they contained the royal burials and sacred ακο$  crocodiles. Going through (διεξιοντες, διεξιουσι) “from the courtyard into the rooms and from the rooms into the columned porches, and from the columned porches into other vestibules, and from the rooms into other courtyards” (just as the narrator walks through cities great and small and along the paths of the logos), he/“they” experienced infinite 39  wonder (θωµα µυριον). The labyrinth represents the limiting case of glosses clustering around the term thoma to connote Herodotus’ ownership of the wonder response and consequently also his independence from the wonder tradition of ethnographic writing before him. The object of wonder is first and foremost something he sees, hears, crosses, learns, and tells.40 It is a wonder to him (µοι), and he is the one who experiences wonder (θωµα ζω). Even in the face of what I have called “conventional” wonders, Herodotus’ impulse to wonder is by definition idiosyncratic and proceeds from the one who is in charge of the logos—my wonder/my logos. Wonder theoretically frees Herodotus from tradition, chronology, his task to explain, and various other constraints; in fact, it competes with these factors as one of the requirements or appetites of the logos. Hartog (1988, 234) noticed the “connection between thoma and digression” in the following passage, where advertisement and program introduce an ethnographic thoma that lies clearly outside the expected range of marvels in distant lands. So, there [i.e., in Scythia] these phenomena occur because of the cold. But I wonder [θωµα ζω]—for indeed my story sought digres  δη µοι ο! λογος  ς sions from the beginning [προσθηκας γαρ ε ξ αρχη ε δι ζητο]—for what reason in the whole territory of Elis mules are not able to be born, when the place is not cold and there is no other visible cause. The Eleans themselves assert that it is as a result of a curse that mules are not born among them. (4.30.1) 39. 2.148.5– 6. The participles could imply an indefinite subject (“anyone going  through”), but the main verb in the past tense (“they caused wonder” [θωµα παρει χοντο]) and the previous first-person plural finite verbs suggest “to us as we were going through.” As Payen remarks (1997, 120), “la vue est associ´ee a` la marche de l’enquˆete . . . elle se confond mˆeme avec le proc`es de l’´ecriture.” 40. See Dewald 1987, 155, n. 21.

Thoma

243

Wondering Why The semantic field of thom- words, especially the verb thomazo, includes an interrogative mode. The question being asked is about the meaning of the thing in a broader, real-world context. In most cases, it remains implicit and ill defined, but at 4.30.1, it is explicit and formulated in terms of why: why are there no mules in Elis? There are no mules or asses in Scythia because they cannot withstand the cold. This belongs to a natural state of affairs, whereas the absence of mules (not asses) specifically in Elis (not in neighboring regions: see 4.30.2) remains a mystery. The divine, which ultimately rules nature (3.108), operates in predictable ways so that phenomena can be seen to derive from material causes. An  αλλου  effect that breaks the pattern and has no visible cause (εοντος ουτε ␸ανερου α ι τι ου) leads one, by default, to the immediate and extraordinary agency of a transcendent force.41 The local inhabitants of Elis resort precisely to this explanation by maintaining that there are no mules in their country because of some curse. The vagueness of their answer makes it all the more valuable. Because the historical and ethical circumstances of the Elean curse remain obscure, the hypothesis of supernatural causes rests primarily on the observation of the irregularity itself.42 This constitutes, therefore, an entirely disinterested and independent item of evidence, a tool of persuasion Herodotus can use at a distance to deal with an issue of great importance to the overarching message of the work. If direct divine intervention can be deduced ethnographically from chronic irregularities in the natural world, that assumption can be perhaps extended to sporadic historical occurrences, even when the laws of nature are not violated so drastically or at all.  It is a wonder to me [θωµα δε µοι] that when they were fighting next to the grove of Demeter, not even one of the Persians apparently went into the sacred precinct or died there, and it is around the sanctuary that most of them fell, on unhallowed ground. And I am 41. So, people in Scythia “are in wonder” when it thunders in winter and regard it as a prodigy (τερας, 4.28.3). A unique case of rain in Upper Egypt was a great portent (␸α σµα µεγιστον) concomitant with the Persian conquest (3.10.3). 42. This is unlike, e.g., the ethical and historical reason of the permanent ethnological abnormality of the Scythian enareis (see 1.105.4). Animals as well as human beings are momentarily affected by transcendent causes as a result of human wrongdoing at 1.167.1, 6.139.1, and 7.171.2 (cited by Smith 1992, 9, 53, and n. 5). For the absence of mules in Elis as a taboo rather than an exception in nature, see Nagy 1990, 336.

244

Telling Wonders

of the opinion, if one must have at all an opinion about divine things, that the goddess herself did not let them in, because they had burned her temple at Eleusis. (9.65.2) There are no mules in Elis, no Persians in Demeter’s precinct. The answer to the question of cause must both times remain provisional, but cases of the first type are scientific corroboration for the more ideologically charged cases of the second. From a scientific perspective, animals embody elemental normalcy in nature as projected by the divine and provide a standard to measure the more complicated world of humans. Animals display a range of customary behaviors in the same way as humans do, except that these animal behaviors are not subject to the cultural processing that results in human nomoi. As a field on which god operates more directly, the animal world represents an intermediary between the divine and the human realms.43 Translated into historical terms, this view encompasses the idea that, on the one hand, animal events reflect human events and, on the other hand, the ways in which they do so mysteriously register divine reaction. From the point of view of Herodotus’ narrative, animals are also mediators between ethnography and history; they participate, moreover, in both the theological and the symbolic codes of explanation. In an episode of the narrative of Xerxes’ march against Greece, lions in Thrace attack the camels of the Persian army (7.125). Whereas in the case of the absence of mules in Elis the discourse needed to justify an abrupt narrative transition to a different place, here it works just as hard, in a way that appears entirely optional, to break the logical continuity of the narrative for the purpose of bringing the event into prominence. It starts, While Xerxes was passing through, lions attacked the camels carrying the provisions. This sentence is pure nonnarrated narrative; as a moderately interesting entry in the journal of the march, it could have been left at that. But a gloss follows, attached by γα ρ, with added narrative details:44 43. See Smith 1992, 7, 32– 33, and passim. Smith has counted 804 references to fauna of some kind in the Histories, 600 if we exclude horses. 44. The structure of the discourse in the first two sentences of 7.125 is analogous to the sequence of summary introduction followed by narrative; but see chap. 1, “What Is Metanarrative,” in the present book.

Thoma

245

 . . . for, coming down at night and leaving their homes [ηθεα], the lions touched no other animal or man and worked havoc only on the camels. Here, the phrases “no other animal” and “only the camels” incline toward the interpretive. But the best is yet to come, in the form of a concluding gloss where explanation is replaced by interrogation in an advertisement of narratability with the narrator’s first person. I am in wonder as to the cause [θωµα ζω δ ε το αιτιον], whatever it ζον] the lions to stay away from the  was that compelled [το αναγκα rest of the army and attack the camels, animals that they had never seen before and of which they had no experience [ε πεπειρεατο]. By repeating the substance of the story three times, the discourse implies that these lions did not behave like normal animals. The gloss “I am in wonder as to the cause” [θωµα ζω δ ε το αιτιον] formulates an enigma parallel to that of the absence of mules in Elis (“I wonder why . . . in the  αλλου  absence of other visible cause” [θωµα ζω ο τι . . . . εοντος ουτε   φανερου α ιτ ιου], 4.30.1– 2). If this event also is supernatural, given the context, it perhaps constitutes an omen; if so, we are invited to decode it appropriately. What is its significance? The phrase “camels carrying the provisions” is bound to recall the prominence of the issue of supplies for Xerxes’ army in the preceding narrative, including Artabanus’ warning on this matter.45 His prediction will have no fulfillment, but it underlines Persian vulnerability abroad. More importantly, camels are the most exotic part of the Persian army: Cyrus uses them, for example, to frighten the Lydian horses and defeat Croesus at Sardis (1.80). They are a synecdoche for the strange invader himself. Lions symbolize lethal behavior and a display of strength, with polyvalent connotations.46 In a political and social setting, a lion signifies kingship and dominance or, in a negative sense, tyranny and any destructive power that acts from within.47 In response to a prophecy, Meles, king 45. See 7.49.5, 7.50.4. See also, in the voice of the narrator, 7.83.2, 7.118– 20, 7.187. 46. The lioness at 3.108.4 is “strongest and most daring” [ ι σχυροτατον κα ι θρασυτατον]; her cub, when still unborn, destroys her womb with its claws. 47. In the fight between the lion and the dog at 3.32, the fighting dog represents Cambyses’ brother Smerdis, the dog’s brother is the opposite of Cambyses, and the lion is

246

Telling Wonders

of Lydia, tries to make Sardis impregnable by bringing a lion around the walls. Hipparchus and Cypselus are represented as lions—the first in a dream, the latter in an oracle.48 Lions in the face of an external enemy, however, connote a valorous offensive stance and fierceness in war.49 The lions who slaughter the camels in Xerxes’ army leave their haunts (called  ηθεα) and take on an exotic opponent of which they have no experience  (ε πεπειρεατο). They are the short-range precursors of the Spartans at Thermopylae, who leave their Peloponnesian homes and customary ways  (i.e., their ηθεα) to fight against an army of which they are “inexperi enced” [απειρο ι ].50 Just as the lions are “compelled” to attack, so the Spartans have undertaken a “compulsory” war: this means a defensive war, but their valor at the time of the battle turns defense into legitimate offense.51 For they have the “courage of bulls and lions,” according to the oracle at 7.220.4, and are led by their king Leonidas, “son of lion,” the Heraclid heir of the Pelopids of Lydia—traditionally and heraldically connected with lions—who will be immortalized with the statue of a lion.52 The gloss of identification of Leonidas at Thermopylae, which introduces his Heraclid genealogy, qualifies him as ο! θωµαζοµενος µαλιστα [the object of greatest wonder], from the point of view not exclusively of the narrator but of all, then and now (7.204). In a nearby and thematically related section, metanarrative thoma praises the courageous actions and words of the two Spartan ambassadors, Sperthias and Boulis (7.135.1). The Persian scout at Thermopylae is in wonder at the Spartans combing their hair and doing exercises before the battle his analogue. See also Knox’s 1952 analysis of the parable of the lion cub in Aesch. Agam. 717– 36. 48. See 5.56.1 (Hipparchus), 5.92β3 (Cypselus), 1.84.3 (Meles: unfortunately, the lion was born to him from a concubine, and Meles neglected to bring him all the way around, so the device ultimately did not work. Did the king fail adequately to provide for his succession?). Pericles is represented by a lion in a dream at 6.131.2; see “Wonder and Disbelief” later in the present chapter. 49. See, e.g., Il. 11.113– 21, 170– 78. 50. 9.46.2. Their inexperience is due to the Spartans’ absence at Marathon, as Pausanias says. I am grateful to Deborah Boedecker for suggesting the double meaning of  the term ηθεα in this passage, with allusion to the Spartans’ exceptionally giving up their insularity. This is not the first instance of this wordplay: see chap. 2, “Identification with the Other,” in the present book. 51. Cf. Il. 11.473– 86, where the lion’s offensive stance is used to characterize the defensive battle situation of Aias moving against the Trojans who crowd around wounded Odysseus. For Spartan aggressiveness at Thermopylae, see 7.211.3 and 7.223.2– 3. On  γκαι η in reference to the Spartans and “compulsory” war, see Demaratus at 7.104.3, ανα discussed in chap. 2, “The Evils of War,” in the present book. 52. See 7.225.2; Immerwahr 1966, 260– 61, n. 69; Georges 1994, 141– 42.

Thoma

247

(ε θωµαζε , 7.208.3). Herodotus’ wondering in the lion episode is related to and foreshadows the broader sense of wonder that he wants the listener to experience at the almost numinous epiphany of heroes in the first battle of this Persian war.53 In the passage about the absence of mules in Elis and other cases, perhaps including the lions episode at 7.125, wondering about cause is equivalent to being struck by the possibility that the divine is manifesting itself through an irregularity of nature.54 What is unambiguously a miracle—that is, a τερας, φα σµα, or σηµηιον—can be termed, from the point of view of the emotional reaction it elicits, a thoma.55 The attack of the Thracian lions, however, represents less of a violation of natural laws than do other animal omens in the Histories, especially the two reported in this same narrative of Xerxes’ march (7.57.1– 2). We should perhaps think of the lions phenomenon as what Homer would call a πελωρ: this occurs whenever animate beings, animals or heroes, reveal themselves pervaded by the presence of the divine, a notion that would well agree with Herodotus’ scientific and ethnographic observation of the continuity among the animal, the human, and the divine realms.56 But the question of why such and such happened and the answer to that question are in this particular case secondary. The story about the lions and the camels is first and foremost a narrative event, a metaphor. The narrator’s intensely subjective and disproportionate wonder at this minor incident in the world of the narrated first and foremost creates a mise en abˆıme for the larger picture of the Thermopylae narrative and what that narrative represents in the larger picture still. It invites the listener not so much to speculate on cause as, less specifically and concretely, to interpret meaning. Why Wonder Another minute instance illustrates the extent to which Herodotus’ communication relies on the symbolic code. This passage also foregrounds 53. See chap. 3, “Specific Glosses of Interpretation.”  54. Cf. 7.153; Herodotus mentions that the feat of Teline is a “wonder to me” [θωµα µοι] in light of the man’s effeminate nature (πε␸υκεναι); the implements of the goddesses in the narrative provide the clue. 55. See 6.117.2– 3 (blinding of Epizelus at Marathon), 8.135 (Mys receives oracle in Carian). The miracles at Delphi (8.37– 38) are objectively called τερατα, ␸α σµατα, and   θεια, but Herodotus also glosses the narrative with wonder words (θωµα, αξια θωµα σαι, 8.37.2). See Nenci 1957, 281– 89, for the various terms for miracles in Homer. 56. Cf., e.g., the “divine behavior” of cats (θεια πρηγµατα) at 2.66.3. See Nenci 1957, 189– 293, on the Homeric concept of πελωρ.

248

Telling Wonders

the behavior of animals: during Darius’ campaign against the Scythians, the mules and donkeys in the Persian army caused disarray among the Scythian horses (4.129). Once again, the mode of narration forces the recipient of the narrative to perform interpretive operations beyond the literal meaning of the text. The summary introduction to the episode contains an embedded program with an advertisement of narratability. I.

One thing that was helpful to the Persians and of hindrance to the Scythians when they were assaulting Darius’ camp—I am  going to tell a great wonder [θωµα µεγιστον ε ρεω]—was the voice of the asses and the sight of the mules. (4.129.1)

An explanatory gloss follows, with a back reference to the ethnographic section about Scythian winters I have already mentioned (4.28). The gloss repeats the same thing twice. II. For the land of Scythia produces neither ass nor mule, as I have explained before. In the whole country of Scythia there is neither ass nor mule because of the cold. (4.129.2) Next we find another summary introduction, resumptive of the first and in ring composition after the gloss. ! ι σαντες], they threw into confuIII. So, when the asses brayed [υβρ sion the Scythian horses. Now comes what we may call the narrative core of the episode, with factual details and in the mode of iteration.57 The negative causal clause  at the end (actually participles preceded by ατε) constitutes yet another gloss that essentially rephrases the idea, already expressed in sentence II, that there are no mules or asses in Scythia. IV. Often, in the middle of an attack against the Persians, when the horses would hear the voice of the asses, wheeling around they were thrown into confusion and were in wonder [ε ν θωµατι εσκον], pricking up their ears, for they had neither heard such a voice nor seen that sight. 57. Iteration occurs when the narrative represents once what happened many times. See Genette 1980, 116.

Thoma

249

The whole passage is capped by a conclusion that reduces the import of the interpretive statement “helpful to the Persians and of hindrance to the Scythians” in the initial introduction (I). V. Because of this, then, they gained a small advantage in the war. (4.129.3) This brief narrative is even more redundant than the one about lions attacking camels. The notion of Persian advantage is repeated twice (I, V), as is the central function that the sound of asses and the appearance of mules troubled the horses (III, IV—both times with ταρα σσω). The background information about the absence of asses and mules in Scythia appears three times (II twice, IV). As he is emphasizing the event, the narrator makes clear that it is small and inconsequential to the outcome (ε π ι τι ε ␸ εροντο). σµικρον This time it is most definitely not an omen; it violates no natural law. There is no wondering why, since the rational cause of the event is clear and reiterated in the glosses. Both the story about Cyrus exploiting the strategic advantage of the camels in his battle against the Lydians (1.80) and the case of the Thracian lions, presented as an exception (7.125), lead us to expect precisely what happened in this case, namely, that animals will be fearful of other animals unfamiliar to them. If the unexpected is an essential feature of thoma, the question is, rather, why should the horses’ being “in wonder” at unfamiliar animals constitute a “great wonder” to the narrator or to us.58 The anthropomor! ι σαντες, ε ταρα σσοντο) is an addi phism of the animals (ε ν θωµατι, υβρ tional sign that the narrative is metaphorical and designed somehow to illuminate the historical context at a higher level of meaning.59 The Per! sians, the only nation in the Histories who are called υβριστα ι by nature (by Croesus at 1.89.2), have symbolic associations with the mule, especially at the beginning of their history. Two mule prodigies refer to Persian actions, but more importantly, Cyrus, the founder of the Persian empire, is called a king-mule by the Delphic oracle.60 Mules (as well as 58. For the notion that the expected is “no wonder,” see 7.187.1. ! ι ζω (behave insolently) has the sense of “braying” at 4.129.2 only; the 59. The verb υβρ root is elsewhere applied to animals at 1.189.1 only (see n. 63 in the present chapter). The verb ταρα σσω is frequently used of armies throwing each other into disarray in this campaign (see especially 4.125.1– 5). 60. For the mule prodigies, see 3.151.2 with 3.153, 7.57.2. For Cyrus as mule, see 1.55.2; cf. 1.91.5– 6.

250

Telling Wonders

donkeys) connote poverty and an inferior social condition; they appear in Herodotean biographies of social upstarts who rise to power.61 Horses connote very different things. On the one hand, they are the sign of a “hard” culture, primitive or nomadic.62 On the other hand, horses can also signify power, wealth, and luxury. By the time of Xerxes’ expedition against Greece, the Persian is a “horse” in this second sense. This is clear from the omen of a mare giving birth to a hare. Hereby it was shown clearly that Xerxes would lead forth his host  with mighty pomp and splendor [αγαυρ οτατα κα ι µεγαλοπρε πεστατα] but that in order to return home, he would have to run for his life.63 In the episode of mules and donkeys, it is mildly ironical that the wealthy Persians gain their only advantage against the rude Scythians thanks to the most modest of their resources. At the symbolic level, however, the narrative suggests that mules and donkeys had what it would have taken for the Persians to be successful vis-a-vis ` the Scythians.64 In the narrative sequence that immediately follows, the bird, mouse, and frog the Scythians send to Darius are negative symbols signifying what the Persians are not (4.132.2). In the same way, here the Persians are not “mules,” and will shortly become “hares” (see 4.134.1; cf. 7.57.1, mentioned earlier). They have lost their original simplicity, such as the Lydian Sandanis described to Croesus (1.71.2– 4). No longer a “hard” people, as they were in the early time of Cyrus, they are a “soft” people attacking a “hard” people, and they will lose the war.65 The proliferation of animal incidents in this group of narratives and the metaphorical value of the animals in the Scythian message constitute one 61. See 1.59.4 (Pisistratus), 6.68– 69 (illegitimacy of Demaratus). Donkeys have equivalent connotations (e.g., at 1.194, 2.121δ, 5.68.1), and at 4.129, the fact that they can bray makes them more essential to the plot. 62. See 1.215.1– 2, 216.4. For the connection of horses with nomadism, see Hartog 1988, 18, on Hdt. 2.108. 63. 7.57.1. At 1.189.2, Cyrus’ white horse steps into the river “out of hybris” when Cyrus is attempting to cross (διαβαι νειν, the verb of violation of boundaries). On Persians and horses, see also 1.192.3, 7.40.2– 4, Smith 1992, 120– 22. Persian riding at 1.136.2 is, however, a trait of a “hard” culture. 64. In fact, the braying donkeys subsequently provide at least the means of Persian escape: see 4.134.3– 135.3. 65. This is an implicit rule of history: see chap. 2, “Implicit and Explicit Difference in the Ethnographies.”

Thoma

251

guarantee of a metaphorical understanding of 4.129, particularly by an audience more accustomed than we are to the functioning of the symbolic code. An additional guarantee is the disproportionate advertisement of  θωµα µεγιστον. Herodotus’ scientific attention to the animal world, which is part of his work as an ethnographer, joins the multifarious traditions that associate animals to humans. Poetic metaphor, such as we find in Homeric similes and animal metaphors in tragedy, verbally signifies specific human behaviors through animals; it is exploited in the religious tradition, which reads ominous meanings in real animal occurrences. Ionian storytelling includes “historical” anecdotes in which the cooperation of animals with humans shows either human natural ability to control lower forms of life or the granting of divine assent to human enterprises.66 Not least important, the animal fable features animals for the purpose of conveying a moralistic message about humans.67 Familiarity with the cultural conventions on which these different types of animal stories are based allows Herodotus’ listeners to integrate into an ethical structure individual events that do not in themselves need to signify anything beyond their literal meaning. Herodotus’ narration does its share to make these events fulfill the function of narrative omens or signs. The word the discourse uses to celebrate them, thoma, is a sign of signs. Vertical Analogy: Wondering Because The animal narratives I have considered show the overlap between the ethnographic notion of wonder and its historical counterpart and illustrate the rules of Herodotus’ pursuit of wonder in the account of past events. The thoma may raise a question concerning its cause or encourage reflection on its meaning. In either case, it often is a small item that the discourse retrieves from the side or magnifies on the way because of its potential to illuminate larger issues. This potential is fully realized by a spectacular image that springs up fully formed in the midst of the narrative of campaigns undertaken by one 66. See Charon of Lampsacus (FGrHist 262 F 1), analogous to Herodotus 1.80. Cf. 3.85– 86. 67. In Herodotus, an animal fable appears at 1.141. The fight between the lion cub and the puppy at 3.32 is an enacted ainos, decoded in the text and demonstrating the “natural” character of solidarity based on blood kinship. In the case of Cambyses, violations of custom entail violations of nature.

252

Telling Wonders

 of the Lydian kings. This is the θωµα µεγιστον once half-witnessed by Periander of Corinth (who was a friend of the enemy of the king), namely Arion of Methyma on a dolphin being carried off to Taenarum. Arion was a harp player, second to none of his contemporaries, and the first we know about who composed the dithyramb, gave it its name and taught the genre in Corinth. (1.23.1) The transition to the story epitomizes the freedom of Herodotus’ logos to pursue wonder on the lateral paths opened up by factual connec tions.68 The glosses of sources and cautionary λεγουσι (governing a narrative all in indirect discourse) contradict Periander’s verification in the narrative of the fact as true.69 Both off track and unbelievable, the story ranks high in the logos for the force with which the narrator applies the celebratory code to its protagonist and to its central event. Modern readers—and at least one ancient—have responded to these signals. The variety of available interpretations testifies to the depth and indeterminacy of what Herodotus regards “a great wonder.”70 The connection of narratability and symbolism, achieved in the previous examples through the cultural code of animals, emerges even more clearly in this case, owing to the narrative’s mythical cast. Attacked by sailors who want to steal his gold, Arion dons his professional robes and sings the nomos orthios (a technical term in the musical code, but meaning “correct law/custom” in ordinary language); he then jumps into the sea and is rescued by a dolphin, which carries him to Taenarum. In the poetic tradition, Apollo jumps on a Cretan ship in the guise of a dolphin and leads it to Taenarum and then on to Crisa, where he initiates the dumbfounded sailors to his priesthood by bidding them, among other things, to 68. See Pagel 1927, 4. Unlike most other anachronic narratives in Herodotus, the Arion episode cannot claim the function of an explanatory gloss in relation to its surroundings. 69. See 1.24.1, 6, 8. Packman (1991, especially 400) argues that Periander, whose initial incredulity is corrected by inquiry and verification ( ι! στορεεσθαι, 1.24.7), is analogous to Herodotus vis-a-vis ` his sources and that Arion, who is on the receiving end of Periander’s disbelief, is analogous to Herodotus vis-a-vis ` his audience, so that “on one level, the Arion story can be read as a plea for a specifically historical suspension of disbelief.” 70. See, e.g., Benardete 1969, 14– 16; Cobet 1971, 145– 51; Flory 1978; Munson 1986; Packman 1991, 399– 401. Plato exploits the metaphorical character of the story in the Republic (454d), by making Socrates say that when his positions come under the wave of an attack, he will “hope for a dolphin to take us on its back or some other impossible    means of rescue.” Here the phrase τινα αλλην απορον σωτερι αν echoes απορ ι ην in Hdt. 1.24.4.

Thoma

253

sing the paean (Hom. H. Ap. 3.388– 544). Dionysus turns into dolphins the Tyrrhenian pirates who have kidnapped him, but he saves and rewards with prosperity the righteous helmsman (Hom. H. Dion. 7). The legend reported by Herodotus uses similar themes to heroize Arion into an almost sacral figure, whose immunity from danger represents the triumph of the fundamental values he embodies: intellectual worth, adherence to nomos, loyalty to the gods, deliberate pursuit of one’s assigned task in the face of danger, and refusal to be subjugated.71 The gloss of testimony appended to the conclusion records the existence at Cape Taenarum of a tangible representation of the rescue, the bronze statue of a man riding a dolphin (1.24.5). The offering is “not  large” [ου µεγα] and is antithetical, therefore, to the impressive monuments of the East. It does less to corroborate the veracity of the legend than to confirm its meaning:72 Arion is also “small” in the face of a stronger opponent, who is ethically (though not ethnically) barbaric. The wonder of his survival joins the mystery of a prodigy to the significance of a natural event. In the context of Herodotus’ scientific and historical work, the agency of the dolphin points again to the participation of nature in a divine plan that is ethically rational according to the standards of men. The ainos of Arion is inserted in the first detailed report in the Histories of the aggression against a small Greek state by a large Eastern power. It independently confirms the evidence provided by that narrative for the causality of success and failure.73 Its message, however, also reflects preoccupations that will come into better focus later on with the rescue, almost miraculous but natural and rational, of the Greeks from the Persian danger.74 The symbol of Arion himself will in fact return at that point, this time on the main path of the logos, just before the Greeks’ first confrontation with the Persian fleet, and in a more realistic incarnation. Gone are the dolphin, the “best musician of his time,” the sacred robes and accoutrements of his profession, the nomos orthios sung among the rower’s benches, and the prodigious reappearance on dry land 71. See Flory 1978. As Benardete remarks (1969, 15), “nowhere else in Herodotus does  νοµος mean ‘tune.’” 72. See Bowra 1963 on the actual religious background of this statue. 73. See Cobet 1971, 149. Alyattes’ failure to conquer the Milesians, due to a divinely induced sickness (1.19.1– 21.1), is just as unpredictable in human terms as is the failure of the Corinthian sailors. 74. See Munson 1986, 99.

254

Telling Wonders

of the rescued musician, utterly intact and “just as he was when he jumped” (1.24.7). The new Arion is Scyllias of Scione, “the best diver of his time” (8.8.1). He deserts from the Persian camp and does what he is accustomed to do and does best: he dives into the sea. Having reached the Greeks at Artemisium, he gives them intelligence of the enemy side. A predominantly metanarrative passage in the mode of historie replaces the mythical narrative of Arion (8.8.2– 3). In what way [Scyllias] at that point arrived among the Greeks, I cannot tell precisely, but I am in wonder if the things that are said   ε στι αληθ   are true [θωµαζω δ ε ε ι τα λεγοµενα εα]. For it is said that from Aphetae, having dived into the sea, he did not reemerge until he arrived to Artemisium, having crossed a distance of about eighty stades by sea. Indeed, other things are said about this man   ικελα . . . that are similar to lies, but some are true [αλλα ψευδεσι    τα δ ε µετεξετερα αληθ εα]. On this matter, however, let me express my opinion that he arrived at Artemisium by boat. Scyllias is almost a contemporary and an Arion translated into history. The narrator intertwines the code of refutation with that of celebration and still expresses the narratability of the unverified event in terms of wonder. He does not so much “wonder whether” as he is “in wonder if”—the formulation he applies to the allegedly floating island of Chemmis.75 The phrase “other things similar to lies, . . . but some true” assigns Scyllias’ “swim” to Artemisium to an uncertain zone between these unbelievable but true facts and the antithetical “lies similar to truth” of epic poetry, meaningful fictions with the power to communicate what is essentially true.76 True or untrue, prodigious or merely exceptional, the rides to safety and freedom of Arion and Scyllias easily become the subject of ainoi that replay the rescue in the larger plot of the Histories as a concluded whole. These passages are not symbolically univocal, for they can accommodate references to more recent extratextual in75. See 2.156.2, discussed under “Herodotus and the Conventional Code of Ethnographic Wonders” earlier in this chapter. 76. Cf. the Muses in Hes. Theog. 27: “We know how to say many lies similar to truth,   but we also know how to sing true things when we want to” [ιδµεν ψευδεα πολλα λεγειν  ε θελωµεν   ηθεα   ε τυµοι σιν οµοι α, / ιδµεν δευτε αλ γηρυσασθαι] (cf. Od. 19.203, Theognis 713). See Bowie 1993, 17– 23. Herodotus’ quotation is a nod to Scyllias’ heroic status among his contemporaries, on which see Masaracchia 1977, 161, How and Wells 1928, 2:238.

Thoma

255

stances of abuse and self-interest among the Greeks. Thus, Arion’s aggressors are not barbarian pirates, as are those of Dionysus in the Homeric hymn; they are Corinthians whom Arion“trusted more than anyone else” to convey him safely home (1.24.2). The gloss of identification of Scyllias informs us that he made use of his diving ability after the shipwreck off Mount Pelion to rescue valuable property for the Persians and gain a great deal for himself as well (8.8.1). This is a nonidealized narrative element whose connotations clash with the symbolism of Scyllias’ feat at Artemisium.77 In the case of Arion, the symbolic indeterminacy is enhanced by the implicit self-referential aspects of the protagonist: he is, like Herodotus, a skilled performer who must eventually confront hostile audiences.78 One of these listens to him for the sake of pleasure but plans to kill him after the song (1.24.5). The other meets his narrative with the same disbelief the narrator of the Histories experiences from his listeners.79 Herodotus’ patterns from small to large or from far to near and across narrative levels create an open-ended system of competing associations that we modern readers have only begun to reproduce. We can only do so by taking interpretive risks. This brings us to another object of wonder, the mysterious Artemisia of Halicarnassus. More daringly than in his narrative on Arion, here Herodotus combines a self-referential element with a special exploitation of the vertical analogy between individuals and states.80 Let us begin with the second aspect first. As an ally of Xerxes in his expedition against Greece, Artemisia was for fifth-century Greeks the historical reincarnation of the mythical Amazon, the enemy of the civilized male world of the polis, the invader of Attica, where she found defeat and obliteration at the hand of the Greeks, specifically the Athenians.81 As we have already 77. Cf. the story of Ameinocles of Magnesia (7.190), criticized by Plutarch (De Malign. Herod. 30, 39 ⫽ Mor. 864C, 871C). 78. See Bernadete 1969, 14– 16; Packman 1991; Thomson 1996, 151 n. 20, 167. Benardete quotes Plato Rep. 454d (see n. 70 in the present chapter), where Socrates also applies the Arion persona to himself in the face of opposition from his listeners. I call a character “self-referential” when it emerges as a double of the narrator Herodotus. But I have been considering a term or a sentence self-referential when it speaks about the narrative either of Herodotus or of his source (see especially my definitions in chap. 1). 79. See Packman 1991 (see n. 70 in the present chapter). Cf. Payen 1997, 58. For reactions attributed to the audience of the Histories, see chap. 1, “Self-Referential Glosses,” in the present book. 80. See especially chap. 2, “Analogy as an Interpretive Tool.” 81. On the connection between Artemisia and the Amazons, see Aristoph. Lys. 671ff.

256

Telling Wonders

seen, in book 4 Herodotus reverses this politically significant tradition by inserting into his Scythian narrative the paradox of marriageable, reasonable, and peaceful Amazons.82 For the purposes of his narrative of the battle of Salamis, he creates a different Amazon paradox, one that turns same into other rather than, as in the Sauromatian logos, other into same. Unlike Herodotus’ Amazons in book 4, Artemisia is aggressive, eager for war, and equipped with excellent ships.83 Unlike the Amazons of tradition, she is Greek, not barbarian; cultured, not wild; and renowned for strategic ability. She is also a winner, and she is a winner over her own side. The Amazon of tradition, defeated by Athens, is here reborn as a symbol of the Athens that was born with the victory of Salamis.84 The triumph of Artemisia is the opposite of that of Arion and is enough to make the narrator throw up his arms in a very different sort of amazement. The narrative has its comic side: reduced in a tight spot during the battle, Artemisia undertakes to save herself in a flash by ramming an allied ship; she gets lucky and wins, and in the process, she gets doubly lucky and increases her power and prestige (8.87– 88). As an open-and-shut case, at least, Artemisia provides evidence that contradicts so much other evidence Herodotus has been accumulating for an ethically rational order of things. Can ill-gained success possibly be permanent? Artemisia is a symbol and an enigma, one of several that punctuate the beginning of the end, and the end of the end, of Herodotus’ investigation.85 We are now ready to complicate the picture further and examine the peculiar way in which the narrator signals this thoma and his own involvement with it. The following is the programmatic introduction to a long gloss of identification for Artemisia. Of the other taxiarchs, I make no mention/memorial, on the  grounds that I am not compelled by necessity [ου παραµεµνη    µαι . . . ως ουκ αναγκαζοµενος], but [I do] of Artemisia, whom   especially I regard with wonder [της µαλιστα θωµα ποιευµαι], a woman marching against Greece, who, since the death of her hus82. See chap. 2. “The Other Is Same.” 83. See 7.99.3. The inexperience in seafaring that the Sauromatian narrative attributes to the Amazons (4.110.1) becomes more intelligible at the symbolic level in light of their opposite and analogue Artemisia: see chap. 2, n. 245. 84. See Munson 1988 for a detailed argument. On personification of poleis in Greek art, see Holscher ¨ 1995, 174. 85. See, e.g., 7.133.2 and chap. 3, “Divine Retribution” in the present book.

Thoma

257

band herself holding the tyranny [τυραννι δην] and having a young son, participated in the expedition out of daring and manly cour ς age, there being no necessity/compulsion for her to do so [ουδεµιη    ο ι ε ουσης αναγκα ι ης]. (7.99.1) The last clause brings into focus Artemisia’s aggressiveness. As a fe  male τυραννος (and, we might add, as a symbol of the city-τυραννος), she does what despotic powers normally do, waging an “unnecessary”— that is, offensive—war.86 At a more immediate level, however, the expres  ς ο ι ε ουσης   sion ουδεµιη αναγκα ι ης bears positive connotations: unlike other allies compelled by Xerxes, Artemisia is, for unspecified reasons (and again like Athens), free.87 This special position is part of her paradox as a wonder and ostensibly the first reason for her inclusion in the logos. In a previous negative program, Herodotus has stated that since the local infantry commanders in the Persian force were not free but   slaves, he was “not constrained by necessity” [ου γαρ αναγκα ι η  ε ξεργοµαι] to mention them (7.96.1– 2). The combination of this passage with 7.99.1, which closely follows, yields the sequence “I am not compelled to mention other commanders who were slaves [i.e., who were under compulsion], but [I am compelled to/will mention] Artemisia, whom I consider a thoma and who was not under compulsion.” The rhetorical figure, underlined by the repetition of metanarrative selfreferential anank- words, creates an antithesis between those other foreign subjects of Xerxes and the free narrator (the prominent ε γω of 7.96.1), who is not compelled to say in his logos what he does not consider worth telling. By the same token, Artemisia of Halicarnassus, at least to the extent that she is free and “not under compulsion,” becomes here (independently from whatever else she is already beginning to symbolize) implicated symbolically in the world of the narration also as an analogue of “Herodotus of Halicarnassus,” the histor.88 86. See, e.g., Artabanus at 7.10δ1 and chap. 3, “The Evils of War,” in the present book.  87. In the historical narrative, words of the αναγκfamily most frequently refer to compulsion applied by a monarchical ruler, especially in the narrative of Xerxes’ expedition (7.103.4; 7.108.1; 7.110; 7.132.2; 7.136.1; 7.139.3; 7.172.1; 8.22.2; 8.140α2; 9.17.1). See Munson forthcoming. The unique position of Artemisia as a free agent is evidenced in the subsequent narrative of her advice to Xerxes (8.67– 69). See Munson 1988, 95– 98. 88. Georges (1994, 301 n. 93) reminds us that she may have been the great-aunt of Herodotus the real author.

258

Telling Wonders

From what sort of compulsion does Herodotus declare himself free? Implicitly it must include something similar or vertically analogous to the monarchical compulsion that the local commanders in the Persian army experience and Artemisia does not. That means all external pressure exercised by persons, for political or other reasons. Herodotus’ immunity from that type of pressure consists here, as many other times, in his being free to magnify the minute and meaningful narrative element that is a thoma to him. This will entail, as it turns out, robbing the real protagonist at the battle of Salamis of a starring role, in favor of a minor participant in the enemy camp, with five ships to her name. Episodes of Greek— and especially Athenian—skill, valor, patriotism, and love of freedom are conspicuously absent in the battle report, in striking contrast with Aeschylus’ version of the same battle and with Herodotus’ own narratives of Thermopylae or even Marathon. First, Herodotus’ narrative of Salamis attributes the Greek victory in great part to the hopeless strategic disarray of the Persian force. Second, but more importantly, Herodotus narrates this battle to make it mean what it meant as seen from the perspective of later stasis developments among the Greeks.89 The central Artemisia section (8.87– 88) contributes to the first purpose in its literal import; by constructing a metaphor for later Athenian unscrupulous behavior, it contributes to the second. Herodotus’ narrative of Salamis, in other words, adds two implicit qualifications to his earlier explicit praise of the Athenians as the saviors of Greece, largely owing to that victory (7.139). This is a radical choice, almost as surprising as Artemisia’s turnabout maneuver on the battlefield. Although daring and free from external compulsion, however, the narrator of the Histories is, unlike Artemisia, subject to an ananke of a higher order. In voicing his praise of Athens at 7.139.1, he counters   political pressure by declaring himself compelled (αναγκα ι η ε ξεργοµαι) to express an unpopular opinion. Here the compulsion is the moral duty  to tell the truth.90 In two additional cases of self-referential αναγκin the passive voice, the compulsion to tell is exercised by the logos and has to do with the narrator’s self-imposed task to provide a didactically effective 89. Cf. chap. 3, “Disputes, Arbitration, and the Subjectivity of Opinions.” The suggestion that the Persian failure was largely due to Persian strategic errors would be part of the argument of those who wished to minimize Athenian merit. See the Corinthians in Thuc. 1.69.5.   ταληθ   90. See 7.139.5: αληθ ες, εος. See also chap. 3, “Specific Glosses of Interpretation,” in the present book.

Thoma

259

account.91 Herodotus “is” Artemisia, but he also “is” Arion. The requirement of his logos, what we may call the nomos of the logos, is his obligation as well as his choice.92 This includes, with explanation, the pursuit of wonders, which are signs for things that need to be explained. Warned by the marker thoma, a sign of signs, his listeners will take note, stop and think, and explain on their own. Wonder and Disbelief The interrogative mode of the root θωµ-, as we have seen, usually does not concern the existence of the object of wonder (is it really true?) but rather has to do with its meaning. Even when the expression of wonder combines with glosses of noncorroboration, Herodotus implicitly encourages the audience to make sense of the wondrous phenomenon as if it were true. A few times, however, θωµ- is more closely connected with disbelief. If some Greeks consider the Persian support of democracies in Ionia a thoma megiston, that means they will find it hard to believe (6.43.3). In the two passages where Ionian theories are a cause of wonder because they are implausible and absurd, θωµ- (exceptionally self-referential in the sense that it identifies the reported logos) functions as a term of the code of refutation, not of the code of celebration.93 The last case I will consider perversely combines the reading directions “thoma, absurd logos, and untrue fact: disregard” and “thoma, untrue fact, interesting logos: pay attention,” perhaps even adding (according to some interpretations) “thoma, unbelievable logos, but perhaps true fact.” The last possibility is hard to accept because it entails understanding the text as saying the exact opposite of what it clearly says. That several readers have advanced it testifies to the greater than usual discordance of what the text communicates at different levels of metanarrative. If there is  ο του λογου   91. See 2.3.2 (υπ ε ξαναγκαζοµενος) and 2.65.2. See also chap. 2, “The Texture of Nomos,” in the present book. For other expressions of obligation, see chap. 1, n. 38 and corresponding text. 92. In Herodotus, moral obligation and the compulsion of nomos often conflict with monarchic obligation and are strictly related to voluntary choice. See, e.g., the case of  and declares himself compelled to tell the Prexaspes, who both decides voluntarily (ε κων) truth (3.75.1– 2). 93. See 2.21, 4.42.1 In the first passage, θωµασιος almost functions as an advertisement of nonnarratability: the theory is wondrous and therefore not worth reporting. See discussion under “Herodotus and the Conventional Code of Ethnographic Wonders” earlier in this chapter.

260

Telling Wonders

a performance in Herodotus that we would most like to observe live, complete with body language and tone of voice, this is certainly it. The passage in question ranks among the quarrels of the Histories, where the histor/arbitrator sets forth an accusation and evaluates the merits of a case. Here the accusation is presented not as a current logos but as a charge belonging to the world of the narrated, made in Athens at the time of Marathon or shortly thereafter (α ι τ ι η . . . εσχε ε ν Αθηναι οισι, 6.115). This aitie claimed that after the fighting was over, the Alcmaeonids arranged for a shield to be flashed from Athens to signal to the Persians at sea that they should round Sounion and sail to the city, which in fact they did. Nothing came of the alleged incident, however, because the victorious Athenian army raced back to Athens by land and arrived before the Persians. So the enemy gave up their designs and, from Phalerum, sailed back to Asia (6.115– 16). After the narrative of other events in the aftermath of the battle (6.117– 20), Herodotus goes back to the Alcmaeonids issue in a long gloss of refutation introduced by a rejection of the charge. Here, wonder at the fact reported leads to disbelief in the report and to its rejection based on unlikelihood:  It is a wonder to me and I do not accept the rumor [θωµα δε µοι κα ι  λογον]   ε νδεκοµαι  ουκ τον that the Alcmaeonids would have ever lifted up a shield to the Persians as a signal according to an agreement, wishing the Athenians to be ruled by the barbarians and Hippias. . . . (6.121.1) The participial clause “wishing the Athenians . . .,” reported as a logical implication of the charge, has the effect of underlining its absurdity. The refutation is attached directly here in a crescendo of rhetorical animation: “they who [ο$ιτινες] were tyrant-haters, just as much or more than Callias, the son of Phaenippus and the father of Hipponicus.” An explanatory gloss follows, demonstrating the extent of Callias’ hostility toward the Pisistratids (6.121.2); this concludes with the restatement of the comparison that had caused the mention of Callias in the first place: “So also  the Alcmaeonids were tyrant-haters [µισοτυραννοι] no less than this man” (6.123.1). So far the refutation seems to express straightforward outrage, even though the comparison that brings the family of Callias into the discourse employs a cultural code of Athenian politics that is not entirely transpar-

Thoma

261

ent to us.94 The second movement of the defense, introduced with a rejection closely analogous to the first (“It is therefore a wonder to me and I do not believe the slander,” 6.123.1), takes off again with the animated “they who” [ο$ιτινες] of the first movement. This time it adduces as proof of innocence the Alcmaeonids’ exile during the whole period of tyranny and their success in bringing the regime to an end: “it & %  Αθηνας   was they who freed Athens [τας ουτοι ησαν ο ι ε λευθηρωσα ντες] more than Harmodius and Aristogeiton, in my judgment [ως ε γω κρι νω].” Herodotus then anticipates the objection that the family might have turned to treason against their fatherland because they had some grudge against the Athenian people, but he counters this point by arguing that no one enjoyed more renown and honor among the Athenians than the Alcmaeonids (6.124.1). So not even the report that they would have   ε λογος made signals with the shield for that reason makes any sense (ουδ   α ιρεει). “A shield was raised and that cannot be denied; it happened” Herodotus concludes, “but as to who did it, I cannot say anything further” (6.124.2). The rhetoric about the Alcmaeonids as liberators of Athens recalls the encomium of the Athenians, “saviors of Greece” (7.139.5); the conclusion of the refutation has the same definitive tone of the indictment of Ephialtes (7.214.3), except that here Herodotus generalizes the indictment and acquits the individual culprit (far more prominent than Ephialtes!) instead of the other way around.95 But the mention of the Alcmaeonids’ renown generates a long explanatory gloss on the history of their wealth and status, where eager argumentation gives way to amiable mimetic narrative and, at the same time, political history gives way to folklore, or what Thucydides would call “the mythical/romantic ele ment” [το µυθωδες] (Thuc. 1.22.4). This gloss consists of three connected narratives, the first of which tells the story of how the family’s eponymous ancestor Alcmaeon enjoyed 94. Cf. the ambiguity of 7.151 (see chap. 3, “Disputes, Arbitration, and the Subjectivity of Opinions,” text and n. 239, in the present book). Here the point may simply be that, like the Alcmaeonids, they were loyal democrats and had long enjoyed good relations with Persia: the identification of Callias as “father of Hipponicus” recalls the son of Hipponicus named Callias, the negotiator of the peace. Callias II may have earned Herodotus’ approval. Probably a religious conservative (diadouch at Eleusis), he was related to Aristides (see Plut. Arist. 24.4– 8; Davies 1971, 257), the only Athenian politician whose portrayal in Herodotus does not contain elements connoting “tyranny” (see 8.79.1). 95. He acquits the Alcmaeonids but lays charge on others, complains Plutarch (De Malign. Herod. 27 ⫽ Mor. 862E).

262

Telling Wonders

friendly relations with Croesus and visited him at his court in Sardis, where he was given a large amount of gold (6.125). The second is about Cleisthenes, the tyrant of Sicyon, choosing a husband for his daughter, Agariste, among the best and brightest young men from all over Greece. He ended up giving her to Megacles, the son of the Alcmaeon of the previous story, who thereby greatly increased the family’s prestige (6.126– 30). The third section briefly reports the birth from Megacles and Agariste of the Athenian democratic reformer Cleisthenes and another son, named Hippocrates. The offspring of this Hippocrates included a younger Agariste, who married Xanthippus. When she was pregnant, she dreamed she had given birth to a lion; after a few days, she became the mother of Pericles (6.131). Scholars have observed that these narratives undermine the defense to which they are appended by highlighting the Alcmaeonids’ long-standing connections with Eastern monarchs and Greek tyrants.96 It is important, however, to establish to what extent they do so and in what way. The possibility of the Alcmaeonids’ closeness to the Persians is not even taken into consideration.97 Herodotus rather refutes the charge on the basis of the Alcmaeonids’ hostility toward the Pisistratids (6.121– 23), a point that the subsequent “renown” narratives never undermine. We find no reference, for example, to the fact that Megacles, who here marries Agariste and fathers the sons Cleisthenes and Hippocrates, gave a daughter in marriage to Pisistratus, as we learn elsewhere in the Histories.98 We have in these sections not an implicit retraction of the defense but a cheerful reflection on the charge, marked by a striking change in the mode, tone, and level of the discourse. In the first two narratives—on Alcmaeon and on Agariste’s wedding, 96. See especially Strasburger 1955, 15– 18; also Thomas 1989, 264– 72. Thomas attributes the jarring effect to the combination of an apologetic family tradition (adapted to the patriotic polis tradition) in the first part of the defense and popular traditions in the second part. 97. We should perhaps infer a pro-Persian policy on the part of Cleisthenes and the Alcmaeonids on the basis of the embassy Herodotus reports at 5.73.3. But Herodotus does not mention Cleisthenes or the Alcmaeonids in that passage, nor does he ever suggest that the aitie at the time of Marathon is somehow connected to the aitie previously incurred by the ambassadors. See Fornara and Samons 1991, 19– 20. 98. See 1.61.1. This silence is noted by Plutarch (De Malign. Herod. 27 ⫽ Mor. 863A– B). Herodotus’ claim that the Alcmaeonids were in exile “for the entire time of the tyranny” (6.123.1) is not inconsistent with his statement at 1.64.3 and with the apparent unawareness of other ancient sources concerning Cleisthenes’ archonship under Hippias in 525 b.c. (ML 6). See Davies 1971, 372; Fornara and Samons 1991, 17.

Thoma

263

respectively—concrete novelistic details proliferate, some very funny. Let loose in Croesus’ treasure-house, Alcmaeon is the ethical opposite of Solon on a similar occasion.99 But he is also his opposite from the point of view of his dramatic role, verbally incapacitated by the gold he has packed into his mouth, his logos replaced by body language. “Stuffed up” with gold in his robes, hair, and boots, he staggers out of the room “similar to anything but a human being” (6.125.4). We are similarly made to see the suitors (µνηστηρες) of Agariste “stuffed up” with lineage and pride.100 They stick to their best behavior in the gracious and tense atmosphere of Cleisthenes’ court while the tyrant assesses their performance in the games and at dinner. During the last and most lavish of an excruciating yearlong series of banquets, in the midst of civilized competitions in public speaking and song, and in front of all the Sicyonians who have been invited to the party, the favored contender, Hippoclides, throws it all away when he starts dancing bottom-up on the table, moving his legs about in the air as if they were arms (6.129.4). “Son of Tisander,” Cleisthenes bursts out, “you have just danced away your marriage.” Hippoclides replies, “Hippoclides does not care,” with a phrase that has become proverbial, as the narrator’s gloss explains (6.130.1). This Hippoclides, who gets tired of good manners, is not just a crude  finally showing his true colors. He is one who, by Homeric µνηστηρ breaking the rules of the polite competition, rebels against the political constraints of his tyrannical audience. His performance is just as whimsical and unconventional as the turnabout of Herodotus from the earnest polis patriotism of his Alcmaeonid defense to the irreverent and factually dubious novelistic narratives that follow. As Plutarch observes, in his criticism of a different passage, It seems to me that, like Hippoclides standing on his head on the table and waving his legs like arms, Herodotus would dance away the truth and say, “Herodotus doesn’t care.”101 An ancient reader here exploits the proverbial character of Hippoclides’ utterance reported by Herodotus to respond to Herodotus’ own 99. See 1.30– 33; Strasburger 1955, 18. 100. Chamberlain (1997, 66– 67) notices the verbal correspondence between παντα   ε ξωγκωτο at 6.125.5 and ε ξωγκωµενοι at 6.126.3. 101. Plut. De Malign. 33 ⫽ Mor. 867B, on Hdt. 7.233.3. In reference to the Alcmaeonid defense, Plutarch uses another ainos, comparing Herodotus to a predator who catches a crab and then promises to let it go (De Malign. Herod. 27 ⫽ Mor. 862F– 863A).

264

Telling Wonders

subversive narrative methods. The self-referential aspect of Hippoclides seems especially apt when applied to the very context where Hippoclides does his dance, that of Herodotus’ extravaganza on the topic of the Alcmaeonids.102 But the transition from argument to lighthearted narratives that work through connotation and symbolism, and from explicit praise to implicit ambivalence, is not unique to this passage. We have already noticed the shift in tone and substance from the encomium of Athens at 7.139 to the report of the antics of Artemisia and others in the narrative of Salamis. Like Artemisia, the narrator of the Histories is a free agent. Like Hippoclides, who frees himself from the constraints of the tyrant Cleisthenes, he displays his freedom as a performer vis-a-vis ` the political establishment of the leading city (the tyrant city) of his time. Herodotus dances away not “the truth,” as Plutarch says, but the question of what the truth is, in the old controversy of the shield at Marathon. The specific accusation against the Alcmaeonids stands rejected, and he no  , as Plutarch paraphrases). That allegalonger cares (ου ␸ροντ ι ς Ηροδοτω tion of the past is now a thoma good for thinking in metaphorical terms about the present. Just as Artemisia at Salamis deconstructs the antithesis between male and female, Greek and Barbarian, friend and enemy, in relation to the polis turannos, so the old rumor about Alcmaeonid medism at Marathon leads Herodotus to a survey of family history that confounds the distinctions between East and West, democracy and tyranny, citizen and foreigner, in reference to its leaders. Proceeding from the eponymous ancestor Alcmaeon, to his son, Megacles, (to whom a foreign tyrant gives  “according to the nomoi of the his daughter in marriage [verb ε γγυω] Athenians” [6.130.2]), to the democratic reformer who bears the name of his tyrant grandfather, the Alcmaeonids’ ambivalent line leads directly to the ambivalent lion Pericles, the one and only Alcmaeonid who was still relevant in Herodotus’ time.103 As a case that engages Herodotus in his role of arbitrator of differences, the discussion of the aitie against the Alcmaeonids conforms in its own way to the rule of explicit acquittal 102. Chamberlain (1997, 34– 81, especially 52– 65) is the first, as far as I know, who saw Hippoclides as a double of Herodotus. Cf. Dewald 1987, 151. 103. Dewald (1998, 691) remarks that Cleisthenes’ remarkable phrase at 6.130.2 “reminds H’s contemporary readers of Pericles’ citizenship law of c. 450 bce, demanding that an Athenian citizen have two Athenian parents. Here Pericles’ own great-grandmother is the foreigner involved.” Pericles claimed an exception from his own law for the sake of his son from Aspasia of Miletus (Plut. Per. 37.5). For the lion imagery, see “Wondering Why” earlier in the present chapter. On Pericles as a lion, see chap. 2, “The Monarchical Model in Athens.”

Thoma

265

and implicit warning or blame.104 This combination is here achieved through a mixture of historie with the mode of the ainos.105 From the point of view of Herodotus’ notion of wonder, however, the Alcmaeonid passage features the narrator more clearly than elsewhere demonstrating through his own discourse what he wants his listeners to do with a thoma—freely associate and reflect, to find a broader context or a different plane of experience where the absurd becomes intelligible and the abnormal meaningful.

104. See chap. 2, “Disputes, Arbitration, and the Subjectivity of Opinions.” 105. See Nagy 1990, 310– 13.

Conclusion

The discussion in this book has sought access to the Histories first and foremost through the discourse: it has attempted to understand what the text says by examining how it says it. I have analyzed three functions of referential metanarrative (comparison, interpretation, and, in a more limited way, celebration) to show how they contribute to transforming historical narrative and ethnographic description—the surface declarative modality of the text—into a lesson for the here and now of narration. Largely based on Herodotus’ experience of foreign societies through time and space, this lesson includes a warning to the Greeks to take stock of essential similarities beyond contingent differences and to recognize the likely human responses, common values, and constraints that emerge from the cultural norms and historical vicissitudes of “all men.” Herodotus’ warning is moralistic, because it is implicitly based on the idea that human beings have control over their behavior and consequently, to some extent, over their fortunes. The text also leaves room, however, for the adviser’s ultimate aporia with regard to the ways in which the recipients of the narrative can usefully apply the teachings he provides. It conveys no confidence that they will in fact be able, or still in time, to benefit from them. Thus, the imperative and interrogative features of the Histories radically modify the movement toward a happy resolution in which the narrative, when considered only from the point of view of its declarative aspects, may appear to be emplotted. Herodotus’ performance wants to be considered from the point of view of the audiences for which it was intended and in light of the circumstances that obtained at the time. This is a risky undertaking for us modern readers: precisely what times, what places, what circumstances, and what audiences are we talking about? Herodotus no doubt composed and performed the Histories piecemeal. They must represent a considerable portion of his life’s work. Some parts perhaps originated as independent expositions and have been only imperfectly adapted to the final form 266

Conclusion

267

of the whole. The whole itself appears at once both finished and provisional. Nevertheless, we can safely bracket off the third quarter of the fifth century b.c. as an extratextual context of Herodotus’ performances. We are relatively well informed about the most memorable events of that time and about some of the contemporary reactions and ideological predispositions of the public. Against this background, I have tried to show that the ideological underpinnings of the ethnographic descriptions represent a well-suited complement to the political thought that emerges from the history, including from those historical narratives that reflect domestic concerns and belong to the most recent identifiable layer. A more precise reconstruction of the circumstances of performance would require relating the histor Herodotus to the real author Herodotus at a particular time. If we do so on the basis of the evidence that is available to us, however, we may end up exercising our imagination to a greater degree than is appropriate to a scholarly task. The external tradition about the life of Herodotus is fragmentary, composite, sometimes derived from the interpretation of hints in the Histories, and for the most part late. It portrays an itinerant lecturer, an Ionized Dorian, possibly half-Carian, born from a distinguished family but perhaps not an aristocrat, politically active but not a military man.1 A figure on the margins in the most literal sense, he started out on the eastern border of the Greek world, a Persian subject; he moved away (somewhat like those Ionians of whom the Histories approve) and eventually ended up in a mixed Greek democratic community in the far west. The Thurii Herodotus joined was not an Athenian imperialistic venture but a unique Athenian initiative with Panhellenic aims; it was inspired, at least for some of those involved, by the atmosphere of conciliation that briefly followed the stipulation of the Thirty Years’ Peace.2 What we   1. On Herodotus’ life, see especially Suda, s.vv. Ηροδοτος and Πανυασσις, and the sources collected and discussed by Jacoby (1913, 205– 47) and Legrand (1932, 5– 15). See also Brown 1988, especially 4: “Unlike the three major historians with whom he is usually compared—Thucydides, Xenophon and Polybius—Herodotus was not a military man, he remained a civilian all his life. And that should not be forgotten if we are to understand the way he thought” (emphasis mine). I also agree with Brown (1988, 8) that the statement in the  ε πι␸ανων)  does not necessarily Suda that Herodotus came from a distinguished family (των imply a noble birth. The personal narrative at 1.143 seems ironically to imply that, unlike his predecessor Hecataeus, Herodotus was not in a position to show off his lineage. 2. The main primary sources for the foundation of Thurii are Diod. 12.9– 11 and Strabo 6.1.13– 15. The original aims and meaning of the enterprise are controversial. See Cloch´e 1945, 95– 103; Ehrenberg 1948; Wade-Gery 1958; Kagan 1969, 154– 78, 382– 84. Kagan (1969, especially 168) attributes the initiative to the desire of Pericles to deflate the

268

Telling Wonders

know about the foundation of Thurii reveals practical and political motives;3 it also suggests, however, that a utopian impulse may have played a role in this enterprise. Plato, after all, would later find inspiration for his ideal state precisely in this part of the world. The architect of Thurii was Hippodamus of Miletus, who reorganized Piraeus and codified the orthogonal subdivision of urban space on the model of the Greek cities of Asia. This eccentric polymath was also versed in political theory, concerned with the ideal size and proportions of the democratic city-state and the best form of government. His proposed constitution envisioned equal political rights for a population of ten thousand citizens divided into three classes: artisans, farmers, and an armed force for the purposes of defense. We can perhaps relate these features to Herodotus’ interest in such subdivisions (e.g., at 2.164– 68), his essentially democratic preferences, his concern vis-a-vis ` the excessive growth of cities, and his approval of a state’s readiness against external attacks. Hippodamus also apparently maintained that in a well-run state, judges should be allowed to render qualified verdicts. This measure, objects Aristotle, would turn a judge into an arbitrator and obliterate the clarity of the judicial process. But Herodotus’ own subtle attempts to settle discords may indicate that he shared Hippodamus’ concerns about the possible unfairness of absolute condemnations or acquittals. Similarly, Hippodamus’ idea that there should be a law for awarding honors to any citizen who should make an invention of benefit to the state may throw light on Herodotus’ exploration of specious or wise inventions by reformers and nations around the world.4 Whatever Hippodamus’ influence might have been on the constitution of Thurii, Protagoras of Abdera undertook the task of writing the new city’s laws. This is the thinker, as we have seen, with whom Herodotus’ text carries on an implicit dialogue on the subjects of relativism and the knowability of the gods in relation to the world’s different religious rhetoric of his domestic opponents by demonstrating “his moderation, his lack of imperial ambition, and his Panhellenic sentiments.” 3. The practical impulse for the first Athenian settlement at Sybaris would have been represented by the opportunity to provide for surplus citizens. The second dispatch of Athenian colonists and individuals from other Greek city-states to the new location of Thurii partly filled the need to reinforce the original Athenians of Sybaris. An allotment of land would have provided a strong economic incentive for the colonists. See Kagan 1969, 157, 166– 67. 4. See Arist. Pol. 2.5, 1267b– 1269a. On inventions (verb ε ξευρι σκειν), see especially chap. 2, n. 35 and corresponding text, in the present book.

Conclusion

269

traditions. Equally remarkable is the fact that one of the oikists was Lampon, a diviner (µαντις) and friend of Pericles.5 Plutarch anecdotally relates how Lampon once explained a unicorn as an ominous phenomenon, while Anaxagoras showed that the malformation of the animal’s cranium was due to natural causes. Modern historians maintain that Pericles doubtless accepted the scientific explanation as correct, while recognizing that the conservative religious perspective was comforting and necessary for the uneducated masses.6 Herodotus, for one, would not have been so ready to interpret the disagreement between the natural philosopher and the seer in this manner. The two explanations reflect the double level of causality that, as Herodotus teaches us, is applicable to the same phenomenon. In the new enterprise of Thurii, the planned cooperation of Protagoras and Lampon may have represented a deliberate match of the best sort of innovation with the healthiest respect for religious tradition.7 The invitation to Greeks of different cities to participate in the settlement, the decision to start over in a new site after eliminating the supremacist group of the original Sybarites,8 the constitution of democratic stamp—all these features are consistent with an experiment in diversity, equality, and harmony. The settlement seems to have employed a number of advisers renowned for their experience of the world at large. Herodotus, the sophos and the traveler, had much to tell about Greeks and non-Greeks, the dangers of leadership in all forms of government, the importance of negotiation and arbitration, the validity of different ways and points of view, the causality and the evils of war, the communality of experience and the role of the divine in the affairs of all men: for him, an unofficial role in the Thurii project was the perfect job. Here was a brand-new city that could learn from the wisdom of foreigners; it had the potential to fulfill the best values of the Greeks, avoid the mistakes of the past, be eudaimon without pleonexie, and grow in moderation without engineering its own decay. Here, the audiences from almost all over Greece, such as Herodotus had addressed in the past at the ephemeral gatherings of 5. See Graham 1964, 36– 37. 6. See Plut. Per. 6, on which see Ehrenberg 1948, 164– 65, quoted with approval by Kagan (1969, 168– 69). 7. Cf. the religious aspect of Hippodamus’ planned state, see Arist. Pol. 2.5.1267b. 8. For the troubles between the Sybarites and the Athenians who had taken part in the first settlement of Sybaris (probably in 446/445) and the move to the new site of Thurii, see Diod. 12.10– 11; Strabo 6.1.13.

270

Telling Wonders

Olympia or separately in their own city, were represented in a single place by a permanent Panhellenic community of fellow citizens.9 The new state must have soon turned out to be a disappointment to the original settlers.10 At about the same time as the conflict of Corcyra with Corinth that Thucydides regarded as one of the precipitating factors for the outbreak of the Peloponnesian War, Thurii experienced internal struggles of its own: perhaps the developments in mainland Greece increased the tension between the Athenian and the Peloponnesian elements of the colony’s diversified population.11 We may fantasize that Herodotus applauded both the decision of the Thurians to seek the arbitration of the oracle of Delphi and the Delphic response. Asked who was to be called the oikist of Thurii, the Pythia named none other than Apollo himself. This made the colony responsible for resolving its own internal troubles and gave it divine sanction for managing its external policy autonomously from the political pressures back home. The colony distanced itself from Athens and was content with a cautious and mostly inactive alliance.12 But Herodotus’ adoptive city now reflected in a microcosm contemporary intercity discords in the rest of Greece, just as his native Halicarnassus had been a sample of the earlier fights of the Greeks on behalf of constitutional freedom. The stasis in Thurii, as well as the stasis outside, 9. The population of Thurii was divided into ten tribes. One was for Athens; three were for Euboea, Ionia, and the islands, respectively; three were Peloponnesian tribes (Arcadians, Eleans, and Achaeans); and three tribes included other Dorians (from Boeotia, Amphictyonis, and Doris). See Diod. 12.10– 11. As Kagan observes (1969, 162– 63), Spartans and Corinthians must have been few and grouped with other Dorians—not surprisingly, since Sparta was underpopulated and Corinth had ample means for supporting her citizens. 10. Most scholars assume that Herodotus was one of them. See, e.g., Legrand 1932, 15. However, according to Parke (1946, 88– 89), Herodotus only went to Thurii just after the publication of the Histories in their current form (i.e., after 430), when the grant of citizenship and allotment of land made him able to support himself. Parke’s argument and the speculation that Herodotus gave up public recitation after his departure for Thurii are far from convincing, but the fact remains that we do not know when Herodotus joined the colony or, for that matter, whether or not he moved elsewhere at the end of his life. See Legrand 1932, 18– 19. 11. This is so at least according to Diod. 12.35, though Graham (1964, 198) remarks, “this may be an over-simplification, since there were settlers from many other parts of Greece.” The civil struggles in Thurii ca. 434 b.c. were preceded, early after the city’s foundation, by an unsuccessful war with the neighboring Spartan colony of Taras over the territory of Siris (see Strabo 6.1.15; cf. SIG 69). 12. On the subsequent history of Thurii and its alignment during the Peloponnesian War, see Graham 1964, 198– 99.

Conclusion

271

must have affected both Herodotus’ relationship with his audiences and the import of his message to them. Either as a citizen or as a performer, the Herodotus of the biographical tradition always seems to have attracted both praise and blame.13 At Thurii as at Halicarnassus and other cities of Greece in his earlier years, his position may not have been entirely stable and comfortable if the honors accorded to him were mixed 14 The charm of Herodotus’ narrative,  with hostile reactions and ␸θονος. the caution with which he expresses himself, his silences about current affairs, his diplomatic elusiveness in arbitrating differences, and his adoption of the mode of the ainos are joined to a sometimes breathtaking tactlessness and irony. The narrator of the Histories seems to both follow and disdain the Pindaric “norm of the polyp.”15 The allusive features of his discourse may have obscured for his audiences the true worth of his political message. At the same time, as even Plutarch’s later reaction demonstrates, they were not destined to hide his ambivalence (or what Plutarch unfairly calls viciousness) toward all sides. At the end of our study of the Histories, the best we can do is observe that “Herodotus of Halicarnassus” and “Herodotus of Thurii,” both attested in the programmatic first sentence, emerge as a strong unified 13. The biographical tradition is insistent in registering episodes of politically motivated public approval or disapproval to Herodotus. At Athens, he allegedly received an exorbitant honorarium granted to him on the proposal of one Anytus (according to Diyillus, FGrHist 73 F 3, cited at Plut. De Malign. Herod. 26 ⫽ Mor. 862A– B), and he was honored by the Council as a result of a reading from his work (Eus. Chron. Olymp. 83.4 ⫽ 83.3 in the Armenian version). According to Aristophanes the Boeotian (FGrHist 379 F 5, cited at Plut. De Malign. Herod. 31 ⫽ Mor. 864D), the Thebans refused payment to Herodotus, and their magistrates prevented him from conversing with the city’s young. See also n. 14 in the present chapter. 14. The Vita of Herodotus from the Suda relates that after helping to drive the tyrant out of Halicarnassus, Herodotus migrated to Thurii “after he saw himself being the object  of resentment/envy [␸θονουµενον] on the part of the citizens.” For the narrator of the  Histories as an object of ␸θονος, see 7.139.1, discussed especially in chap. 3, “Specific Glosses of Interpretation,” in the present book; his alter ego Artemisia is similarly the object  of ␸θονος (8.69.1). Xerxes’ odd generalization about strangers being less invidious to one another than are fellow citizens (7.237.2– 3) perhaps functions as an indirect reassurance of Herodotus’ goodwill to the Greeks. 15. See Gentili 1988, 132: “the difficult external circumstances in which Pindar’s professional duties have placed him are a visible source of embarrassment. It is as if the poet were unable to conceal completely his efforts to follow the precept of Amphiaraus and imitate the polyp that assumes the color of the rock to which it clings.” Gentili goes on to quote Pindar frag. 43 SM: “Let your mind, my son, behave like the skin / of the rockclinging beast of the sea / and consort with men of all nations; go along willingly with those around you, / change your thoughts to suit the seasons” (trans. Gentili and Cole).

272

Telling Wonders

subject to whom the logos as a whole belongs.16 The image of the road he travels as researcher and narrator shows him to be transient, unestablished, and uncommitted to one place. He is comfortable away from home and comfortable with the fragmented diversity of the world. Though powerless to coerce and unlikely to persuade, he is also more qualified than most of his fellow Greeks to undermine current assumptions. He is free like Artemisia, an original like Hippoclides, and subject, like Arion, only to the technical and ethical requirements of his task. He sets himself up as a foil to a variety of types that populate his logos and potentially his public: partisan citizens, conventional thinkers, the politically unfree, performers working for a patron, scientific or juridical histores with vested interests or perverted methods, poets who invent myths, tourists and mindless colonists who demean native traditions, and the overly rationalistic deniers of the work of the divine. At the center of Herodotus’ concern is the need to place the experience of the Greeks in a broader context. Only cultural comparison will allow them to define and assess correctly their language, their customs, their values, their public and private actions, and the cultural tendencies these reveal. The problematic embodied in the questions “What is Greek?” and “What is not Greek?” competes with the question about the legitimacy of formulating them in this way. The importance of these issues is related to the political significance, as it emerges from Herodotus’ text, of both the historical and the ethnological activity. The professional histor, his sources, his audience, and men in general are all retellers of the past as well as observers of customs. The thoughts they hold on these matters are bound to affect their actions toward others. Herodotus achieves a demythologized reconstruction of Greek resistance to the Persian invader and the later attempts of different groups of Greeks to interpret that event in their own ways. He portrays what the Greeks conceive Greekness to be as it actually manifested itself in precious moments at Marathon, Thermopylae, Salamis, and Plataea. His is,   16. The unanimous reading in the manuscript tradition is Ηροδοτου ‘Αλικαρνησσεος. But earlier sources and especially the quotation by Aristotle (Rhet. 3.9.2.1409a) give  Ηροδοτου Θουρι ου. Some later authors testify to his migration to Thurii by saying specifically that he eventually “was called Thurian” (see especially Strabo 14.2.16; Plut. De Malign. Herod. 35 ⫽ Mor. 868A). Plutarch attests that in his day, some manuscripts gave Θουρι ου (De Ex.13). Jacoby (1913, 205) and Legrand (1932, 12– 15) argue that Θουρι ου is the original reading, and Legrand has adopted it in his edition. It would be fitting, however, and no less a possibility, that both versions were authored by Herodotus and in circulation during his lifetime.

Conclusion

273

however, a qualified portrayal that reveals at least the fragility, if not the utter falsity, of the cultural superiority to which the Greeks lay a claim. From their success against the Persians, the Greeks have derived a selfconfidence and a sense of entitlement able to transform a defensive stance into aggression, first against the enemy, then against each other. This always occurs in the name of the same old antithesis between “us” and “them” that the conflict with the Persians seemed historically to embody. Who plays the civilizing role now and who are the Amazons, however, will depend on the point of view. To recognize one’s own subjectivity is equivalent to fulfilling the Delphic imperative “Know thyself” and is in turn the basis for acting according to “nothing in excess.” Not to recognize it falls under the heading of what we would call ethnocentrism and leads to oppressive, tyrannical behavior. For Herodotus of Halicarnassus and Thurii, who is centered in no place and belongs to nowhere, what is wrong with the Greeks of his time in their dealings with one another was already visible during their common resistance against the Persians and even earlier still. Their difficulties are partially due to a cognitive error he attempts to correct: their inadequate understanding of themselves and others. The subversiveness of his logos was perhaps destined to exceed Herodotus’ intention. It set out to explain causes, celebrate achievements, and present realistic paradigms drawn from long stretches of time and space; in the end, it produced a statement that the Greeks and Greekness itself were bound for an uncertain future. What remains in the Histories, so irresistibly seductive to this day, is the cumulative representation of the many different ways in which a civil society can be what it ought to be, coupled with an astonishing awareness of that goal’s always imperfect fulfillment.

Bibliography

Aly, W. 1921. Volksmarchen, ¨ Sage und Novelle bei Herodot und seinen Zeitgenossen. Gottingen. ¨ Reprint, 1969. ———. 1929. Formprobleme der fruher ¨ griechischen Prosa. Philol. Suppl. Leipzig. Amandry, P. 1950. La mantique apollinienne a` Delphi. Paris. Andrewes, A. 1938. “Eunomia.” CQ 32:89– 102. Armayor, O.K. 1987. “Hecataeus’ Humor and Irony in Herodotus’ Narrative of Egypt.” AW 16, nos 1– 2:11– 18. Asad, T. 1986. “The Concept of Cultural Translation in British Social Anthropology.” In Clifford and Marcus 1986, 141– 64. Asheri, D., ed. 1988. Erodoto: La Lidia e la Persia, Libro I delle storie, testo e commento. Milan. ———, ed. 1990a. Erodoto: La Persia, Libro III delle Storie, testo e commento. Milan. ———. 1990b. “Herodotus on Thracian Society and History.” In Nenci 1988, 131– 63. Austin, J.L. 1962. How To Do Things with Words. Cambridge, Mass. Austin, M.M., and P. Vidal-Naquet. 1977. Economic and Social History of Ancient Greece: An Introduction. Berkeley and Los Angeles. Austin, N. 1994. Helen of Troy and Her Shameless Phantom. Ithaca. Bach, K., and R. Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge. Badian, E. 1993. From Plataea to Potidaea: Studies in the History and Historiography of the Pentecontaetia. Baltimore. ———. 1994. “Herodotus on Alexander I of Macedon: A Study in Some Subtle Silences.” In Hornblower 1994b, 107– 30. Bal, M. 1985. Narratology: Introduction to the Theory of Narrative. Toronto, Buffalo, and London. Baldry, H.C. 1965. The Unity of Mankind in Greek Thought. Cambridge. Barley, N. 1992. The Innocent Anthropologist. New York. Barth, H. 1968. “Zur Bewertung und Auswahl des Stoffes durch Herodotus (Die   Begriffe θωµα, θωµαζω, θωµασιος und θωµαστος).” Klio 50:93– 110. Barthes, R. 1970. S/Z. Paris. ———. 1986. “The Discourse of History.” In The Rustle of Language, trans. R.

275

276

Telling Wonders

Howard, 127– 40. New York. Originally published as “Le discours de l’histoire,” Social Science Information 6 (1967): 65– 75. Basso, K. 1976. “‘Wise Words’ of the Western Apache: Metaphor and Semantic Theory.” In Meaning in Anthropology, ed. K. Basso and H. Selby, 93– 121. Albuquerque. Beck, I. 1971. Die Ringomposition bei Herodot. Hildesheim and New York. Belsey, C. 1980. Critical Practice. London. Reprint, 1991. Benardete, S. 1969. Herodotean Inquiries. The Hague. Benedict, R. 1989. The Chrysanthemum and the Sword: Patterns of Japanese Culture. 1946. Reprint, with a foreword by E.F. Vogel, Boston. Benveniste, E. 1971. Problems of General Linguistics. Trans. M.E. Meek. Miami. Originally published as Probl`emes de linguistique gen´erale (Paris, 1967). Bernabo, ` L. 1977. “Oracoli come messaggio: Erodoto testimone di una dimensione orale dei responsi oracolari?” BIFG 4:157– 74. Bett, R. 1989. “The Sophists and Relativism.” Phronesis 34/2. 139– 69. Bickerman, E.J. 1952. “Origenes Gentium.” CP 47:65– 81. Bischoff, H. 1932. Der Warner bei Herodot. Diss., Marburg. Bloomer, W.M. 1993. “The Superlative Nomoi of Herodotus’s Histories.” CA 12:30– 50. Boedeker, D. 1987a. “The Two Faces of Demaratus.” In Boedeker 1987b, 185– 201. ———. 1993. “Hero Cult and Politics in Herodotus.” In Dougherty and Kurke 1993, 164– 77. ———, ed. 1987b. Herodotus and the Invention of History. Arethusa 20, nos. 1–2. ———. Forthcoming. “Mythical Patterns and Epic Heritage.” In de Jong, Bakker, and van Wees forthcoming. Boedeker, D., and K. Raaflaub, eds. 1998. Democracy, Empire, and the Arts in Fifth-Century Athens. Cambridge, Mass. Booth, W.C. 1983. The Rhetoric of Fiction. 2d ed. Chicago. Bornitz, F. 1968. Herodot Studien: Beitra¨ ge zum Versta¨ ndnis der Einheit des Geschichtswerkes. Berlin. Bowen, A.J., ed. 1992. Plutarch: The Malice of Herodotus. Trans. A.J. Bowen. Warminster. Bowie, E.L. 1993. “Lies, Fiction, and Slander in Early Greek Poetry.” In Gill and Wiseman 1993, 1– 37. Bowra, C.M. 1963. “Arion and the Dolphin.” MH 20:121– 34. Briant, P. 1988. “H´erodote et la societ´e perse.” In Nenci 1988, 69– 104. Broadhead, H.D., ed. 1960. The Persae of Aeschylus. Cambridge.  Brown, F.S., and W.B. Tyrrell. 1985. “Εκτιλ ωσαντο: A Reading of Herodotus’ Amazons.” CJ 80:297– 302. Brown, T. 1988. “Early Life of Herodotus.” AW 17:3– 15. Burkert, W. 1985. “Herodot uber ¨ die Namen der Gotter: ¨ Polytheismus als historisches Problem.” MH 43:121– 32. ———. 1988. “Herodot als Historiker fremdem Religionen.” In Nenci 1988, 1– 32.

Bibliography

277

Burn, A.R. [1962] 1984. Persia and the Greeks: The Defense of the West, c. 546– 478 b.c. 2d ed., with a postscript by D.M. Lewis. London. Castriota, D. 1992. Myth, Ethos, and Actuality: Official Art in Fifth-Century b.c. Athens. Madison. Chamberlain, D. 1997. “Herodotean Voices: Reading Characters in the Histories.” Ph.D. diss., University of California, Berkeley. ———. 1999. “On Atomics Onomastic and Metarhythmic Translations in Herodotus.” Arethusa 32, no. 3:263– 312. ———. Forthcoming. “‘We the Others’: Interpretive Community and Plural Voice in Herodotus.” Chatman, S. 1978. Story and Discourse: Narrative Structure in Fiction and Film. Ithaca. Chiasson, C.C. 1986. “The Herodotean Solon.” GRBS 27:249– 62. Christ, M.R. 1994. “Herodotean Kings and Historical Inquiry.” CA 13, no. 2:167– 202. Cingano. 1985. “Clistene di Sicione: Erodoto e i poemi del ciclo tebano.” QUCC 49:31– 40. Clifford, J. 1986. “On Ethnographic Allegory.” In Clifford and Marcus, 1986, 98– 121. Clifford, J., and G.E. Marcus, eds. 1986. Writing Culture: The Poetics and Politics of Ethnography. Berkeley and Los Angeles. Cloch´e, P. 1945. “P´ericl`es et la politique ext´erieure d’Ath`enes entre la paix de 446– 445 et les pr´eludes de la guerre du P´eloponn`ese.” AC 14:93– 128. Cobet, J. 1971. Herodots Exkurse und die Frage der Einheit seines Werke. Historia Einzelschriften 17. Wiesbaden. ———. 1977. “Wann Wurde Herodots Darstellung der Persekriege Publiziert?” Hermes 105:2– 27. Cole, T. 1967. Democritus and the Sources of Greek Anthropology. Chapel Hill. Reprint Atlanta, 1990. Connor, W.R. 1971. The New Politicians of Fifth-Century Athens. Princeton. ———. 1984. Thucydides. Princeton. ———. 1985. “Narrative Discourse in Herodotus and Thucydides.” In The Greek Historians: Literature and History, Papers Presented to A.E. Raubitschek, ed. M.H. Jameson, 1– 17. Saratoga, Calif. ———. 1987. “Tribes, Festivals, and Processions: Civic Ceremonial and Political Manipulation in Archaic Greece.” JHS 107:40– 50. ———. 1993. “The Histor in History.” In Rosen and Farrell 1993, 127– 40. Cook, A. 1976. “Herodotus: The Act of Inquiry as a Liberation from Myth.” Helios 3:23– 66. Cook, A.B. 1907. “Nomen Omen.” CR 21:69. Cooper, G.L. 1974. “Intrusive Oblique Infinitives in Herodotus.” TAPA 104:23– 76.  Constructions  (ως) ———. 1975. “The Ironic Force of the Pure Optative in οτι of the Primary Sequence.” TAPA 105:29– 34. Corcella, A. 1984. Erodoto e l’analogia. Palermo.

278

Telling Wonders

———, ed. 1993. Erodoto le Storie: La Scizia e la Libia, Libro IV testo e commento. Milan. Crahay, R. 1956. La litt´erature oraculaire chez H´erodote. Paris. Darbo-Peschanski, C. 1987. Le discours du particulier: Essai sur l’enquˆete H´erodot´eenne. Paris. Davies, J.K. 1971. Athenian Propertied Families. Oxford. Davison, J.A. 1962. “Literature and Literacy in Ancient Greece.” Phoenix 16:141– 56, 219– 33. Defradas, J. 1954. Les th`emes de la propagande delphique. Paris. de Jong, I.J.F. 1987. Narrators and Focalizers: The Presentation of the Story in the Iliad. Amsterdam. ———. 1998. “Aspects Narratologiques des Histoires d’H´erodote.” Lalies: Actes des sessions de linguistique et de litt´erature 19. de Jong, I.J.F., E. Bakker, and H. van Wees, eds. Forthcoming. A Companion to Herodotus. Leiden. Demand, N. 1987. “Herodotus’ Encomium of Athens: Science or Rhetoric?” AJP 108:746– 58. Denniston, J.D. 1934. The Greek Particles. Reprint, Oxford, 1987. de Romilly, J. 1971. “La vengeance comme explication historique dans l’oeuvre d’H´erodote.” REG 84:314– 37. de Sanctis, G. 1933. “Intorno al razionalismo di Ecateo.” RFC, n.s., 11: 5– 15. Detienne, M. 1973. Les maˆıtres de la verit´e dans la Gr`ece archa¨ıque. Paris. ———. 1977. The Gardens of Adonis: Spices in Greek Mythology. Trans. Janet Lloyd. New York. Reprint, Princeton, 1994. Originally published as Les jardins d’Adonis (Paris, 1972). ———. 1989. “Culinary Practices and the Spirit of Sacrifice.” In Detienne and Vernant 1989, 1– 20. Detienne, M., and J.P. Vernant. 1989. The Cuisine of Sacrifice among the Greeks. Trans. Paula Wissung. Chicago. Originally published as La cuisine du sacrifice en pays grec (Paris, 1979). Dewald, C. 1981. “Women and Culture in Herodotus’ Histories.” Women’s Studies 8, nos. 1– 2:93– 126. ———. 1985. “Practical Knowledge and the Historian’s Role in Herodotus and Thucydides.” In The Greek Historians: Literature and History, Papers Presented to A.E. Raubitschek, ed. M.H. Jameson, 47– 63. Saratoga, Calif. ———. 1987. “Narrative Surface and Authorial Voice in Herodotus’ Histories.” In Boedeker 1987b, 147– 70. ———. 1990. Review of Hartog 1988. CP 85, 217– 24. ———. 1993. “Reading the World: The Interpretation of Objects in Herodotus’ Histories.” In Rosen and Farrell 1993, 55– 70. ———. 1997. “Wanton Kings, Pickled Heroes, and Gnomic Founding Fathers: Strategies of Meaning at the End of Herodotus’ Histories.” In Classical Closure: Reading the End in Greek and Latin Literature, ed. D.H. Roberts, F.M. Dunn, and D. Fowler, 62– 82. Princeton. ———. 1998. Herodotus: The Histories. Trans. Robin Waterfield. Oxford.

Bibliography

279

———. 1999. “The Figured Stage: Focalizing the Initial Narratives of Herodotus and Thucydides.” In Contextualizing Classics: Ideology, Performance, Dialogue; Festschrift Peradotto, ed. N. Felson, D. Konstan, and T. Faulkner, 229– 61. Lanham, Md. ———. Forthcoming a. “Emoi ou genelogesanti emeouton: Herodotus and the Authorial Persona.” In de Jong, Bakker, and van Wees forthcoming. ———. Forthcoming b. “Form and Content: The Question of Tyranny in Herodotus.” In Popular Tyranny, ed. K. Morgan. Austin. Diels, H. 1910. “Die Anfa¨ nge der Philologie bei den Griechen.” Neue Jahrbucher ¨ fur ¨ das Klassische Altertum, Geschichte und Deutsche Literatur 25, no. 13:1– 25. Reprinted in H. Diels, Kleine Schriften zur Geschichte der Antiken Philosophie, 68– 92 (Hildesheim, 1969). Diller, H. 1961. “Die Hellenen-Barbaren-Antithese im Zeitalter der Persekriege.” In Grecs et Barbares, 39– 68. Fondation Hardt, Entretiens 8. Geneva. Dillery, J. 1992. “Darius and the Tomb of Nitocris (Hdt. 1.187).” CP 87:30– 38. ———. 1996. “Reconfiguring the Past: Thyrea, Thermopylae, and Narrative Patterns in Herodotus.” AJP 117:317– 54. Donadoni, S. 1947. “Erodoto, Plutarco e l’Egitto.” Belfagor 2:203– 8. Dorati, M. 2000. Le Storie di Erodoto: etnografia e recconto. Pisa and Rome. Dougherty, C., and L. Kurke, eds. 1993. Cultural Poetics in Archaic Greece: Cult, Performance, Politics. Cambridge. Drews, R. 1970. “Herodotus’ other logoi.” AJP 91:181– 91. ———. 1973. The Greek Accounts of Eastern History. Cambridge, Mass. Drexler, H. 1972. Herodot Studien. Hildesheim. DuBois, P. 1982. Centaurs and Amazons: Women and the Pre-History of the Great Chain of Being. Ann Arbor. Durand, J.-L. 1989. “Greek Animals: Towards a Topology of Edible Bodies.” In Detienne and Vernant 1989, 87– 118. Easterling, P.E., and J.V. Muir, eds. 1985. Greek Religion and Society. Cambridge. Eco, U. 1962. Opera aperta: Forma e indeterminazione nelle poetiche contemporanee. Milan. Ehrenberg, V. 1948. “The Foundation of Thurii.” AJP 59:149– 70. Elayi, J. 1979. “Deux oracles de Delphes: Les r´eponses de la Pythie a` Clisth`ene de Sicyon et aux Ath´eniens avant Solamine.” REG 92:224– 30. Else, G.F. 1958. “‘Imitation’ in the Fifth Century.” CP 53:73– 90. Erbse, H. 1956. “Der erste Satz im Werke Herodots.” In Festschrift B. Snell, 209– 22. Munich. ———. 1961. “Tradition und Form in Werke Herodots.” Gymnasium 68: 339– 57. Euben, J.P., ed. 1986. Greek Tragedy and Political Theory. Berkeley. Evans, J.A.S. 1961. “The Dream of Xerxes and the ‘Nomoi’ of the Persians.” CJ 57:109– 11. ———. 1965. “Despotes Nomos.” Athenaeum 43:145– 53. ———. 1968. “Father of History or Father of Lies: The Reputation of Herodotus.” CJ 64:11– 17.

280

Telling Wonders

———. 1976. “Herodotus and the Ionian Revolt.” Historia 25, no. 1:31– 37. ———. 1979. “Herodotus and Athens: The Evidence of the Encomium.” AC 48:112– 18. ———. 1980. “Oral Tradition in Herodotus.” Canadian Journal of Oral History 4, no. 2:8– 16. ———. 1981. “Notes on the Debate of the Persian Grandees in Herodotus, 3.80– 82.” QUCC 45:79– 54. ———. 1982. Herodotus. Boston. ———. 1988. “The Medism of Pausanias: Two Versions.” Antichthon 22: 1– 11. ———. 1991. Herodotus, Explorer of the Past. Princeton. Fehling, D. 1989. Herodotus and His “Sources”: Citation, Invention, and Narrative Art. Trans. J.G. Howie. Leeds. Originally published as Die Quellenangaben bei Herodot (Berlin and New York, 1971). Ferrill, A. 1978. “Herodotus on Tyranny.” Historia 27:385– 98. Fetterman, D.M. 1989. Ethnography Step by Step. Newbury Park, Calif. Figuera, T.J. 1985. “Herodotus and the Early Hostilities between Aegina and Athens.” AJP 106:49– 74. ———. 1986. “Population Patterns in Late Archaic and Classical Sparta.” TAPA 116:165– 212. Finley, M.I. 1968. “Sparta.” In Probl`emes de la guerre en Gr`ece ancienne, ed. J.P. Vernant, 143– 60. Paris. ———. 1980. Ancient Slavery and Modern Ideology. London. Fisher, N.R.E. 1988. “Drink, Hybris and the Promotion of Harmony in Sparta.” In Powell 1988b. Flory, S. 1978. “Arion’s Leap: Brave Gestures in Herodotus.” AJP 99:441– 21. ———. 1980. “Who Read Herodotus’ Histories?” AJP 101:12– 28. ———. 1987. The Archaic Smile of Herodotus. Detroit. Flower, H. 1991. “Herodotus and Delphic Traditions about Croesus.” In Georgica: Greek Studies in Honor of George Cawkwell, ed. M.A. Flower and M. Toher, 57– 77. London. Ford, A. 1990. “Unity in Greek Criticism and Poetry.” Review of Unity in Greek Poetics, by M. Heath. Arion 3, no. 1:125– 54. Fornara, C.W. 1966. “Some Aspects of the Career of Pausanias of Sparta.” Historia 15, no. 3:257– 71. ———. 1971a. Herodotus: An Interpretive Essay. Oxford. ———. 1971b. “Evidence for the Date of Herodotus’ Publication.” JHS 91: 25– 34. ———. 1981. “Herodotus’ Knowledge of the Archidamian War.” Hermes 109: 149– 56. ———. 1983. The Nature of History in Ancient Greece and Rome. Berkeley. ———. 1990. “Human History and the Constraint of Fate in Herodotus.” In Conflict, Antithesis and the Ancient Historian, ed. J.W. Allison, 25– 45. Columbus, Ohio. Fornara, C.W., and L.J. Samons II. 1991. Athens from Cleisthenes to Pericles. Berkeley.

Bibliography

281

Forrest. W.G. 1966. The Emergence of Greek Democracy: 800– 400 b.c. New York. ———. 1984. “Herodotus and Athens.” Phoenix 38:1– 11. Fowler, R.L. 1996. “Herodotos and His Contemporaries.” JHS 116:62– 87. Fra¨ nkel, H. 1924. “Eine Stileigenheit der fruhgriechischen ¨ Literatur.” NAWG 63– 126. Reprinted in Wege und Formen der fruhgriechischen ¨ Denkens, 2d ed., 40– 96 (Munich, 1962). French, A. 1972. “Topical Influences in Herodotus’ Narrative.” Mnemosyne 25:9– 27. Frisch, P. 1968. Die Traume ¨ bei Herodot. Beitrage ¨ zur klassische Philologie 27. Meisenheim am Glam. Froidefond, C. 1971. Le mirage e´ gyptien dans la lit´erature grecque d’Hom`ere a` Aristote. Paris. Gammie, J.G. 1986. “Herodotus on Kings and Tyrants: Objective Historiography or Conventional Portraiture?” JNES 45:171– 95. Gantz, T. 1993. Early Greek Myth: A Guide to Literary and Artistic Sources. Baltimore and London. Garlan, Y. 1988. Slavery in Ancient Greece. Trans. J. Lloyd. Ithaca. Originally published as Les esclaves en Gr`ece ancienne (Paris, 1982). Garland, R. 1985. The Greek Way of Death. Ithaca. Geertz, C. 1973. The Interpretation of Cultures. New York. ———. 1983. Local Knowledge. New York. ———. 1984. “Anti-anti Relativism.” American Anthropologist 86:263– 78. ———. 1986. “The Uses of Diversity.” In The Tanner Lectures in Human Values 7:253– 75. Cambridge. ———. 1988. Works and Lives: The Anthropologist as Author. Stanford. Genette, G. 1980. Narrative Discourse: An Essay in Method. Trans. J.E. Lewin. Foreword by J. Culler. Ithaca, NY. Originally published as Discours du r´ecit (Paris, 1972). Gentili, B. 1988. Poetry and Its Public in Ancient Greece. Trans. A.T. Cole. Baltimore. Originally published as Poesia e pubblico nella Grecia antica (Rome, 1985). Georges, P. 1994. Barbarian Asia and the Greek Experience. Baltimore. Gigante, M. 1956. Nomos Basileus. Naples. Gill, C., and T.P. Wiseman, eds. 1993. Lies and Fiction in the Ancient World. Austin. Gillis, D. 1979. Collaboration with the Persians. Historia Einzelnschriften 34. Stuttgart. ´ Giraudeau, M. 1984. Les notions juridiques et sociales chez H´erodote: Etudes sur le vocabulaire. Paris. Goff, B., ed. 1995. History, Tragedy, Theory: Dialogues on Athenian Drama. Austin. Gomme, A.W. 1956. A Historical Commentary on Thucydides, Vol. II. Oxford. Gould, J. 1985. “On Making Sense of Greek Religion.” In Easterling and Muir 1985, 1– 33. ———. 1989. Herodotus. London.

282

Telling Wonders

———. 1994. “Herodotus and Religion.” In Hornblower 1994b, 91– 106. Graham, A.J. 1964. Colony and Mother City in Ancient Greece. Manchester. Gray, V. 1996. “Herodotus and Images of Tyranny: The Tryants of Corinth.” AJP 117:361– 89. ———. 1997. “Reading the Rise of Pisistratus: Herodotus 1.56– 69.” Histos 1 (August). Available at ⬍http://www.dur.ac.uk/histos⬎ (May 2000). Greenblatt, Stephen. 1991. Marvelous Possessions: The Wonder of the New World. Chicago. Grene, D., trans. 1987. Herodotus, the History. Chicago. Griffith, G.T. 1966. “Isegorie in the Assembly in Athens.” In Ancient Society and Institutions: Studies Presented to Victor Ehrenberg on his Seventy-Fourth Birthday, ed. E. Badian, 115– 38. Oxford. Griffiths, A. 1988. “Was Cleomenes Mad?” In Powell 1988b, 51– 78. Guthrie, W.H.C. 1950. The Greeks and Their Gods. Boston. ———. 1965. A History of Greek Philosophy. Vol. 2. Cambridge. ———. 1969. A History of Greek Philosophy. Vol. 3. Cambridge. ———. 1971. The Sophists. Cambridge. Hall, E. 1989. Inventing the Barbarian: Greek Self-Definition through Tragedy. Oxford. Hall, J.M. 1997. Ethnic Identity in Greek Antiquity. Cambridge. Hanson, F.A. 1979. “Does God Have a Body? Truth, Reality, and Cultural Relativism.” Man, n.s., 14:515– 29. Hanson, V.D. 1989. The Western Way of War: Infantry Battle in Classical Greece. Oxford. Harvey, D. 1994. “Laconica: Aristophanes and the Spartans.” In Powell and Hodkinson 1994, 35– 58. Halperin, D.M. 1990. One Hundred Years of Homosexuality and Other Essays on Greek Love. New York. Harmatta, J. 1988. “Herodotus, Historian of the Cimmerians and the Scythians.” In Nenci 1988, 115– 30. Harrison, T. 1998. “Herodotus’ Conception of Foreign Languages.” Histos 2 (March). Available at ⬍http://www.dur.ac.uk/histos⬎ (May 1998). ———. 2000. Divinity and History: The Religion of Herodotus. Oxford. Hart, J. 1982. Herodotus and Greek History. London. Hartog, F. 1988. The Mirror of Herodotus: The Representation of the Other in the Writing of History. Trans. J. Lloyd. Berkeley. Originally published as Le miroir d’Herodote (Paris, 1980). Hatch, E. 1983. Culture and Morality: The Relativity of Values in Anthropology. New York. Havelock, E.A. 1982. The Literate Revolution in Greece and Its Cultural Consequences. Princeton. Heidel, W.A. 1935. “Hecataeus and the Egyptian Priests in Herodotus, Book II.” American Academy of Arts and Sciences 18, no. 2:53– 134. Heinimann, F. 1945. Nomos und Physis. Basel. Reprint, 1965. Hellmann, F. 1934. Herodots Kroisos-Logos. Berlin. Helms, L.V. 1882. Pioneering in the Far East and Journeys to California in 1843 and to the White Sea in 1848. London.

Bibliography

283

Hohti, P. 1974. “Freedom of Speech in Speech Sections in the Histories of Herodotus.” Arctos 8:19– 27. ¨ ———. 1975. “Uber die Notwendigkeit bei Herodot.” Arctos 9:31– 37. ———. 1977. “συµβαλλεσ␽αι: A Note on Conjectures in Herodotus.” Arctos 11:5– 14. Holscher, ¨ T. 1998. “Images and Political Identity: The Case of Athens.” In Boedeker and Raaflaub 1998, 153– 83. Hornblower, S. 1987. Thucydides. Baltimore. ———. 1991. A Commentary on Thucydides. Vol. 1, Books I– III. Oxford. ———. 1992. “Thucydides’ Use of Herodotus.” In Philolakon: Lakonian Studies in Honor of Hector Catling, ed. J.M. Sanders, 141– 54. Athens. ———. 1994a. “Narratology and Thucydides.” In Hornblower 1994b, 130– 63. ———, ed. 1994b. Greek Historiography. Oxford. How, W.W., and J. Wells. 1928. A Commentary on Herodotus. 2 vols. Oxford. Reprint, 1964. Howald, E. 1923. “Ionische Geschichtsschreibung.” Hermes 58:113. ———. 1944. Vom Geist antiker Geschichtsschreibung. Munich. Humphreys, S. 1987. “Law, Custom, and Culture in Herodotus.” In Boedeker 1987b, 211– 20. Hunter, V. 1982. Past and Process in Herodotus and Thucydides. Princeton. Immerwahr, H.R. 1954. “Historical Action in Herodotus.” TAPA 85:16– 45. ———. 1956. “Aspects of Historical Causation in Herodotus.” TAPA 87: 247– 80. ———. 1960. “Ergon: History as a Monument in Herodotus and Thucydides.” AJP 81:261– 90. ———. 1966. Form and Thought in Herodotus. Cleveland. Jacobson, R. 1960. “Closing Statement: Linguistics and Poetics.” In Style in Language, ed. T.A. Sebeok, 350– 76. New York. Jacoby, F. 1913. “Herodotus.” RE Suppl. 2: 205– 520. ———. 1944. “GENESIA: A Forgotten Festival of the Dead.” CQ 38:65. Johnson, W.A. 1994. “Oral Performance and the Composition of Herodotus’ Histories.” GRBS 35:229– 54.  Jones, C.P. 1996. “Εθνος and γενος in Herodotus.” CQ 46:315– 20. Jordan, B. 1986. “Religion in Thucydides.” TAPA 116:119– 47. Jouanna, J. 1981. “Les causes de la defaite des barbares chez Eschyle, H´erodote et Hippocrate.” Ktema 7:3– 15. Kagan, D. 1969. The Outbreak of the Peloponnesian War. Ithaca. Kallet, L. 1998. “Accounting for Culture in Fifth-Century Athens.” In Boedeker and Raaflaub 1998, 43– 58. Kennedy, G.A., trans. 1991. Aristotle, On Rhetoric: A Theory of Civic Discourse. Oxford. Kerferd, G.B. 1981. The Sophistic Movement. Cambridge. Kindstrand, J.F. 1981. Anacharsis: The Legend and the Apophthegmata. Uppsala. Kirchberg, L. 1965. Die Funtion der Orakel im Werke Herodots. Gottingen. ¨ Klees, H. 1964. Die Eigenart des griechischen Glaubens an Orakel und Seher. Stuttgart.

284

Telling Wonders

   : Untersuchungen zur Geschichte einer ευρετ ης Kleingunther, ¨ A. 1933. Πρωτος Fragestellung. Leipzig. Reprint, 1975. Kleinknecht, H. 1940. “Herodot und Athen. 7.139, 8.140– 144.” Hermes 75:241– 64. Knox, B. 1952. “Lion in the House.” CP 47:17– 25. Reprinted in B. Knox, Word and Action: Essays on the Ancient Theater, 27– 38 (Baltimore, 1979). ———. 1961. “The Ajax of Sophocles.” HSCP 65:1– 36. Reprinted in B. Knox, Word and Action: Essays on the Ancient Theater, 125– 60 (Baltimore, 1979). ———. 1971. Oedipus at Thebes: Sophocles’ Tragic Hero and His Time. New York. Originally published in New Haven, 1957. Konstan, D. 1983. “The Stories in Herodotus’ Histories: Book I.” Helios 10, no. 1:1– 22. ———. 1987. “Persians, Greeks, and Empire.” In Boedeker 1987b, 59– 70. Kothe, H. 1969. “Der Skythenbegriff bei Herodot.” Klio 51:15– 88. Krischer, T. 1965. “Herodots Prooimion.” Hermes 93:159– 67. Kurke, L. 1995. “Herodotus and the Language of Metals.” Helios 22, no. 1: 36– 64. Labov, W. 1972. Language in the Inner City: Studies in the Black English Vernacular. Philadelphia. Labov, W., and J. Waletzky. 1966. “Narrative Analysis: Oral Versions of Personal Experience.” In Essays on Verbal and Visual Arts: Proceedings of the 1966 Annual Spring Meeting of the American Ethnological Society, ed. J. Helm, 12– 44. Seattle. Lachenaud, G. 1978. Mythologies, religion, et philosophie de l’histoire dans H´erodote. Lille and Paris. Lang, M.L. 1984. Herodotean Narrative and Discourse. Cambridge, Mass. Lardinois, A.P. 1995. “Wisdom in Context: The Use of Gnomic Statements in Archaic Greek Poetry.” Ph.D. diss., Princeton University. Lasserre, F. 1976. “H´erodote et Protagoras: Le d´ebat sur les constitutions.” MH 33, no. 2:65– 84. Lateiner, D. 1977. “No Laughing Matter: A Literary Tactic in Herodotus.” TAPA 107:173– 82. ———. 1980. “A Note on dikas didonai in Herodotus.” CQ 30:30– 32. ———. 1984. “Herodotean Historiographical Patterning: The Constitutional Debate.” QS 20:257– 84. ———. 1985. “Polarita` : Il principio della differenza complementare.” QS 22: 79– 103. ———. 1986. “The Empirical Element in the Methods of Early Greek Medical Writers and Herodotus: A Shared Epistemological Response.” Antichthon 20:1– 26. ———. 1989. The Historical Method of Herodotus. Toronto. ———. 1990. “Deceptions and Delusions in Herodotus. CA 9, no. 2: 229– 46. Lattimore, R. 1939. “The Wise Advisor in Herodotus.” CP 34:24– 39. ———. 1958. “The Composition of the History of Herodotus.” CP 53:9– 21. Legrand, E. 1932. H´erodote: Introduction. Paris. ———, ed. 1946– 60. H´erodote. Vols. 2– 11. Paris.

Bibliography

285

Lesky, A. 1966. A History of Greek Literature. Trans. J. Willis and C. de Heer. New York. Originally publsihed as Geschichte der griechischen literatur (Bern, 1957). L´evˆeque, P., and P. Vidal-Naquet. 1996. Cleisthenes the Athenian: An Essay on the Representation of Space and Time in Greek Political Thought. Trans. D.A. Curtis. Atlantic Highlands, N.J. Originally published as Clisth`ene l’ath´enien (Besan¸con, 1964). Levy, E. 1981. “Les origines du mirage Scythe.” Ktema 6:57– 68. Lewis, D.H. 1985. “Persians in Herodotus.” In The Greek Historians: Literature and History, Papers Presented to A.E. Raubitschek, ed. M.H. Jameson, 101– 17. Saratoga, Calif.  Linforth, I.M. 1918. “Οι αθανατ ι ζοντες (Hdt. 4.93– 96).” CP 13:23– 33. ———. 1924. “Herodotus’s Avowal of Silence in his Account of Egypt.” UCPCPh 9:269– 92. ———. 1926. “Greek Gods and Foreign Gods in Herodotus.” UCPCPh 9:1– 25. ———. 1928. “Named and Unnamed Gods in Herodotus.” UCPCPh 9:201– 43. Lissarague, F. 1990. The Aesthetics of the Greek Banquet. Trans. A. SzegedyMaszak. Princeton. Originally published as Un Flot d’Images: une esth´etique du banquet grec (Paris, 1987). Lloyd, A.B. 1975. Herodotus, Book II: Introduction. Leiden. ———. 1976. Herodotus, Book II: Commentary, 1– 98. Leiden. ———. 1988a. Herodotus, Book II: Commentary, 99– 182. Leiden. ———. 1988b. “Herodotus on Egyptians and Libyans.” In Nenci 1988, 215– 44. Lloyd, G.E.R. 1966. Polarity and Analogy: Two Types of Argumentation in Early Greek Thought. Cambridge. Lloyd-Jones, H. 1971. The Justice of Zeus. Berkeley. Lombardo, M. 1988. “Erodoto, Storico dei Lidi.” In Nenci 1988, 171– 203. Long, T. 1984. Barbarians in Greek Comedy. Carbondale and Edwardsville. ———. 1987. Repetition and Variation in the Short Stories of Herodotus. Frankfurt am Main. Loraux, N. 1977. “La ‘belle mort’ spartiate.” Ktema 2:105– 20. ———. 1986. The Invention of Athens: The Funeral Oration in the Classical City. Trans. A. Sheridan. Cambridge, Mass. Originally published as L’invention d’Ath`enes: Histoire de l’oraison fun`ebre dans la “cit´e classique.” (Paris, 1981). Lovejoy, A.O., and G. Boas. 1935. Primitivism and Related Ideas in Antiquity. Baltimore. Reprint, New York, 1965. Macan, R.W. 1895. Herodotus: The Fourth, Fifth, and Sixth Books. 2 vols. London. Reprint, New York, 1973. ———. 1908. Herodotus: The Seventh, Eighth, and Ninth Books. 2 vols. London. Marg, W., ed. 1965. Herodot: Eine Auswahl aus der neueren Forschung. Darmstadt. Marincola, J. 1987. “Herodotean Narrative and the Narrator’s Presence.” In Boedeker 1987b, 121– 38. ———. 1997a. Authority and Tradition in Ancient Historiography. Cambridge. ———. 1997b. “Odysseus and the Historians.” Histos 1 (October): 1– 26. Available at ⬍http://www.dur.ac.uk/histos⬎ (May 1999).

286

Telling Wonders

Martin, R.P. 1989. The Language of Heroes: Speech and Performance in the Iliad. Princeton. Masaracchia, A. 1975. Studi Erodotei. Rome. ———, ed. 1977. Erodoto: La Battaglia di Salamina, Libro VIII delle Storie, testo e commento. Milan. ———, ed. 1978. Erodoto: La Sconfitta dei Persiani, Libro IX delle Storie, testo e commento. Milan. Matthews, V. 1974. Panyassis of Halicarnassus. Leiden. Mazzarino, S. 1966. Il Pensiero Storico Classico. Vol. 1. Bari. McGlew, J.F. 1993. Tyranny and Political Culture in Ancient Greece. Ithaca. Meier, C. 1990. The Greek Discovery of Politics. Trans. D. McLutock. Cambridge, Mass. Originally published as Die Entstehung des Politischen bei den Griechen (Frankfurt, 1980). Mitchell, L.G. 1997. Greeks Bearing Gifts: The Public Use of Private Relationships. Cambridge. Moles, J. 1993. “Truth and Untruth in Herodotus and Thucydides.” In Gill and Wiseman 1993, 88– 121. ———. 1996. “Herodotus Warns the Athenians.” Papers of the Leeds International Seminar 9:259– 84. Momigliano, A.D. 1954. “On the Causes of War in Ancient Historiography.” In Acta Congressus Madvigiani: Proceedings of the Second International Congress of Classical Studies, 1:199– 211. Reprinted in A.D. Momigliano, Studies in Historiography (New York, 1966), 112– 26. ———. 1958. “The Place of Herodotus in the History of Historiography.” History 43:1– 13. Reprinted in A.D. Momigliano, Studies in Historiography (New York, 1966), 127– 41. ———. 1975a. Alien Wisdom: The Limits of Hellenization. Cambridge. ———. 1975b. “The Fault of the Greeks.” Daedalus 104, no. 2:1– 15. ———. 1978. “The Historians of the Classical World and Their Audiences: Some Suggestions.” ASNP, 3rd ser., 8:59– 75. Mora, F. 1985. Religione a Religioni nelle Storie di Erodoto. Milan. Morris, I. “Beyond Democracy and Empire: Athenian Art in Context.” In Boedeker and Raaflaub 1998, 59– 86. Muller, ¨ D. 1980. Satzbau, Satzgliederung und Satzverbindung in der Prosa Herodots. Meisenheim am Glam. Muller, ¨ K.E. 1972. Geschichte der antiken Ethnographie und ethnologischen Theoriebildung, Vol. 1. Wiesbaden. Munson, R.V. 1983. “Transitions in Herodotus.” Ph.D. diss., University of Pennsylvania. ———. 1986. “The Celebratory Purpose of Herodotus: The Story of Arion in Histories 1.23– 24.” Ramus 15, no. 2:93– 104. ———. 1988. “Artemisia in Herodotus.” CA 7, no. 1:91– 106. ———. 1991. “The Madness of Cambyses (Herodotus 3.16– 38).” Arethusa 24:43– 65. ———. 1993a. “Herodotus’ Use of Prospective Sentences and the Story of Rhampsinitus and the Thief in the Histories.” AJP 114:27– 44.

Bibliography

287

———. 1993b. “Three Aspects of Spartan Kingship in Herodotus.” In Rosen and Farrell 1993, 39– 54. ———. Forthcoming. “Αναγκη in Herodotus.” JHS. Murray D. 1987. “Herodotus and Oral History.” In Achaemenid History, vol. 1, Sources, Structures, and Syntheses, ed. H. Sancisi-Weerdenberg, 93– 115. Leiden. Myres. J.L. 1953. Herodotus, Father of History. Oxford. Reprint, Chicago, 1971. Nagy, G. 1979. The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry. Baltimore. ———. 1990. Pindar’s Homer: The Lyric Possession of an Epic Past. Baltimore. Nenci, G. 1957. “La concezione del miracoloso nei poemi omerici.” Atti dell’Accademia di Scienze 2, Classe di scienze morali, storiche e filologiche 92:275– 311. ———. 1962. “La concezione del miracoloso in Esiodo.” Critica Storica 31: 251– 57. ———, ed. 1988. H´erodote et les peuples non grecs. Fondation Hardt, Entretiens 35. Geneva. ———. 1994. Erodoto, le Storie: La rivolta della Ionia, Libro V, testo e commento. Milan Nilsson, H.P. 1955. Geschichte der griechischen Religion, Vol. 1. Munich. North, H. 1966. Sophrosune. Ithaca. Nussbaum, M. 1986. The Fragility of Goodness. Cambridge. Ober, J. 1989. Mass and Elite in Democratic Athens. Princeton. ———. 1993. “The Athenian Revolution of 508/7 B.C.E.: Violence, Authority, and the Origins of Democracy.” In Dougherty and Kurke 1993, 215– 32. Ogden, D. 1993. “Cleisthenes of Sicyon, ΛΕΥΣΤΗΡ.” CQ 43:353– 63. Ostwald, M. 1965. “Pindar, ΝΟΜΟΣ, and Heracles (Pindar frg. 169 [Snell3] and P.Oxy. No. 1450, frg. 1).” HSCP 69:109– 38. ———. 1969. Nomos and the Beginning of the Athenian Democracy. Oxford. ———. 1986. From Popular Sovereignty to the Sovereignty of Law. Berkeley. ———. 1988. Αναγκη in Thucydides. Atlanta. ———. 1991. “Herodotus and Athens.” ICS 16:137– 48. ———. 1995. “Freedom and the Greeks.” In The Origins of Modern Freedom in the West, ed. R.W. Davis, 35– 63. Stanford. Packman, Z.M. 1991. “The Incredible and the Incredulous: The Vocabulary of Disbelief in Herodotus, Thucydides, and Xenophon.” Hermes 119:399– 414. Pagel, K.A. 1927. “Die Bedeutung des aitiologischen Moments fur ¨ Herodots Geschichtschreibung.” Ph.D. diss., Berlin. Parke, H.W. 1946. “Citation and Recitation: A Convention in Early Greek Historians.” Hermathena 67:80– 92. Parsons, P.J. 1992. “3965. Simonides, Elegies.” Oxyrhynchus Papyri 59:4– 50. Payen, Pascale. 1997. Les ˆıles nomades: Conqu´erir et r´esister dans l’Enquˆete d’H´erodote. Paris. Pearce, T.E.V. 1981. “‘Epic Regression’ in Herodotus.” Eranos 79:87– 90. Pearson, L. 1939. Early Ionian Historians. Oxford. ———. 1941. “Credulity and Scepticism in Herodotus.” TAPA 72:335– 55.

288

Telling Wonders

Pelling, C. 1997. “East Is East and West Is West—or Are They? National Stereotypes in Herodotus.” Histos 1 (March). Available at ⬍http://www.dur.ac.uk/ histos⬎ (September 1998). Pembroke, S.G. 1967. “Women in Charge: The Function of Alternatives in Early Greek Tradition and the Ancient Idea of Matriarchy.” JWCI 30:1– 35. Perlman, S. 1976. “Panhellenism, the Polis, and Imperialism.” Historia 25:1– 30. Podlecki, A.J. 1977. “Herodotus in Athens?” In Greece and the Eastern Mediterranean: Studies Presented to F. Schacherneyer on the Occasion of His Eightieth Birthday, ed. K.H. Kinzl, 246– 65. Berlin. Pohlenz, M. 1937. Herodot, der erste Geschichtsschreiber des Abendlandes. Leipzig. Reprint, Darmstadt, 1973. Powell, A. 1988a. “Mendacity and Sparta’s Use of the Visual.” In Powell 1988b, 173– 92. ———, ed. 1988b. Classical Sparta: Techniques Behind Her Success. Oklahoma Series in Classical Culture. London. Powell, A., and S. Hodkinson, eds. 1994. The Shadow of Sparta. London. Powell, J.E. 1938. A Lexicon to Herodotus. Cambridge. Reprint, Hildesheim, 1950. ———. 1939. The Histories of Herodotus. Cambridge. Pratt, M.L. 1977. Toward a Speech-Act Theory of Literary Discourse. Bloomington. ———. 1986. “Fieldwork in Common Places.” In Clifford and Marcus 1986, 27– 50. Prince, G. 1973. “Introduction a` l’´etude du narrataire.” Po´etique 14:178– 96. ———. 1977. “Remarques sur les signes m´etanarratifs.” Degr´es 11, no. 12: 1– 10. ———. 1980. “Notes on the Text as Reader.” In The Reader in the Text: Essays on Audience and Interpretation, ed. S.R. Suleiman and I. Closman, 225– 40. Princeton. ———. 1982. Narratology: The Form and Function of Narrative. Berlin, New York, and Amsterdam. ———. 1987. A Dictionary of Narratology. Lincoln. Pritchett, W.N. 1979. The Greek State at War. Vol. 2, Religion. Berkeley. ———. 1993. The Liar School of Herodotos. Amsterdam. Raaflaub, K.A. 1979. “Polis Tyrannos: Zur Entstehung einer politischen Metapher.” Arktouros: Hellenic Studies Presented to B.M.W. Knox, ed. G.W. Bowersock et al., 239– 41. Berlin and New York. ———. 1987. “Herodotus, Political Thought, and the Meaning of History.” In Boedeker 1987b, 221– 48. ———. Forthcoming. “Herodotus and Current Intellectual Trends.” In de Jong, Bakker and van Wees forthcoming. Raubitschek, A. 1939. “Εργα µεγαλα τε κα ι* θωµαστα.” RAE 41:217– 22. Rawlinson, G., trans. 1880. History of Herodotus. 4th ed. 4 vols. New York. Redfield, J. 1985. “Herodotus the Tourist.” CP 80:97– 118. Richardson, S. 1990. The Homeric Narrator. Nashville. Robinson, T.M. 1979. Contrasting Arguments: An Edition of the Dissoi Logoi. New York.

Bibliography

289

Romm, J.S. 1989. “Herodotus and Mythic Geography: The Case of the Hyperboreans.” TAPA 119:97– 117. ———. 1992. The Edges of the Earth in Ancient Thought, Geography, Exploration, and Fiction. Princeton. ———. 1998. Herodotus. New Haven. Rood, T. 1998. Thucydides: Narrative and Explanation. Oxford. Rosaldo, R. 1986. “From the Door of His Tent: The Fieldworker and the Inquisitor.” In Clifford and Marcus 1986, 77– 97. ———. 1989. Culture and Truth: The Remaking of Social Analysis. Boston. Rose, P.W. 1995. “Historicizing Sophocles’ Ajax.” In Goff 1995, 59– 90. Rosellini, M., and S. Sa¨ıd. 1978. “Usages de femmes et autres nomoi chez les ‘Sauvages’ d’H´erodote.” ASNP, 3d ser., 8:849– 1005. Rosen, R.M., and J. Farrell, eds. 1993. Nomodeiktes: Greek Studies in Honor of Martin Ostwald. Ann Arbor. Rosler, ¨ W. 1991. “Die ‘Selbhistorisierung’ des Autors: Zur Stellung Herodots zwischen Mundlichkeit ¨ und Schriftlichkeit.” Philologus 135:215– 20. Sa¨ıd, E.W. 1978. Orientalism. London. Sa¨ıd, S. 1980. “Guerre, Intelligence et Courage dans les Histoires d’H´erodote.” Ancient Society: 11– 12, 83– 117. ———. 1998. “Tragedy and Politics.” In Boedeker and Raaflaub 1998, 275– 95. Salmon, A. 1956. “L’Experience de Psamm´etique (II.2).” EC 24:321– 29. Samons, L. 1998. “Kimon, Kallias, and Peace with Persia.” Historia 47:129– 40. Schmid, W., and O. Sta¨ hlin. 1934. Geschichte der Griechischen Literatur. Munich. 1/2. 550– 673. Schmitt, R. 1976. “The Medo-Persian Names of Herodotus in the Light of the New Evidence from Persepolis.” Acta Antiqua Academiae Scentiarum Hungaricae 24:25– 35.   Schroeder, O. 1917. “Νοµος ο παντων βασιλευς.” Philologus 74:195– 204. Searle, J.R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge. ———. 1976. “A Classification of Illocutionary Acts.” Language in Society 5: 1– 23. ———. 1979. Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge. Segal, C. 1971. The Theme of the Mutilation of the Corpse in the Iliad. Mnemosyne Suppl. 17. Leiden. ———. 1982. Dionysian Poetics and Euripides’ Bacchae. Princeton. Seitel, P. 1969. “Proverbs: A Social Use of Metaphors.” In The Wisdom of Many: Essays on the Proverb, ed. W. Weider and A. Dundes, 122– 39. New York. Shapiro, S. 1994. “Learning through Suffering: Human Wisdom in Herodotus.” CJ 89, no. 4:349– 55. ———. 2000. “Proverbial Wisdom in Herodotus.” TAPA 130:89– 118.   η µει ς ιδµεν.” Eranos 71:45. Shimron, B. 1973. “Πρωτος των ———. 1979. “Ein Wortspiel mit homoioi bei Herodot.” RhM, n.s., 122: 131– 33. ———. 1989. Politics and Beliefs in Herodotus. Historia Einzelschriften 58. Stuttgart.

290

Telling Wonders

Sinos, R.H. 1993. “Divine Selection: Epiphany and Politics in Archaic Greece.” In Dougherty and Kurke 1993, 15– 45. Smith, S.M. 1992. “Herodotus’ Use of Animals: A Literary, Ethnographic, and Zoological Study.” Ph.D. diss., Ohio State University. Smyth, H.W. 1956. Greek Grammar. Revised by G.M. Messing. Cambridge, Mass. Sourvinou-Inwood, C. 1991. Reading Greek Culture: Texts and Images, Rituals and Myths. Oxford. Stadter, P. 1992. “Herodotus and the Athenian arche.” ASNP, 3d ser., 22: 781– 809. ———. 1997. “Herodotus and the North Carolina Oral Tradition.” Histos 1 (January). Available at ⬍http://www.dur.ac.uk/histos⬎ (February 1999). Stahl, H.P. 1975. “Learning through Suffering? Croesus’ Conversations in the History of Herodotus.” YCS 24:1– 36. Stein, H. 1877– 1883. Herodot. 5 vol. Berlin.   Stier, H. 1928. “Νοµος βασιλευς.” Philologus 83:225– 58. Strasburger, H. 1955. “Herodot und das perikleische Athen.” Historia 4:1– 25. Strassler, R.B., ed. 1996. The Landmark Thucydides. New York. Svenbro, J. 1976. La parole et le marbre: Aux origines de le po´etique grecque. Lund. ———. 1993. Phrasikleia. An Anthropology of Reading in Ancient Greece. Trans. Janet Lloyd. Ithaca. Originally published as Phrasikleia: Anthropologie de le lecture en gr`ece ancienne. (Paris, 1988). Thomas, R. 1989. Oral Tradition and Written Record in Classical Athens, Cambridge. ———. 1997. “Ethnography, Proof, and Argument in Herodotus’ Histories.” PCPS 43:128– 48. ———. 2000. Herodotus in Context: Ethnography, Science and the Art of Persuasion. Cambridge. Thomson, G., ed. 1938. Aeschylus, Oresteia. 2 vols. Cambridge. Thomson, N. 1996. Herodotus and the Origins of the Political Community: Arion’s Leap. New Haven. Tourraix, A. 1976. “La femme et le pouvoir chez H´erodote.” DHA 2:366– 86. Tyrrell, W.B. 1984. Amazons: A Study in Athenian Mythmaking. Baltimore. Tyrrell, W.B., and F. Brown. 1991. Athenian Myths and Institutions: Words in Action. Oxford. Untersteiner, M. 1967. I Sofisti. 2 vols. Milan. van Dijk, T. 1976. “Pragmatics and Poetics.” In Pragmatics of Language and Literature, ed. T. Van Dijk, 23– 57. Vandiver, E. 1991. Heroes in Herodotus: The Interaction of Myth and History. Frankfurt. Van Groningen, B.A. 1958. La composition litt´eraire archa¨ıque grecque: proc´ed´es et realisations. Amsterdam. van Leyden, V. 1949– 50. “Spatium Historicum: The Historical Past as Viewed by Hecataeus, Herodotus, and Thucydides.” DUJ 11:89– 104. Van Otterlo, W.A.A. 1944. Untersuchungen uber ¨ Begriff, Anwendung und Entstehung der griechischen Ringkomposition. Amsterdam.

Bibliography

291

Van Paassen, C. 1957. The Classical Tradition of Geography. Groningen. Vernant, J. 1988. “City-State Warfare.” In Myth and Society in Ancient Greece, trans. J. Lloyd, 29– 53. New York. ———. 1989a. “At Man’s Table: Hesiod’s Foundation Myth of Sacrifice.” In Detienne and Vernant 1989, 21– 86. ———. 1989b. “Food in the Countries of the Sun.” In Detienne and Vernant 1989, 164– 69. Veyne, P. 1971. Comment on e´ crit l’histoire. Paris. Vidal-Naquet, P. 1981. “Slavery and the Role of Women in Tradition, Myth and Utopia.” Myth, Religion, and Society, ed. R.C. Gordon, 187– 200. Cambridge. Von Bothmer, D. 1957. Amazons in Greek Art. Oxford. von Fritz, K. 1936. “Herodotus and the Growth of Greek Historiography.” TAPA 67:315– 40. Wade-Gery, H.T. 1958. “Thucydides the Son of Melesias.” In Essays in Greek History, 239– 70. Oxford. Walbank, F.W. 1951. “The Problem of Greek Nationality.” Phoenix 5:41– 60. Waters, K.H. 1971. Herodotus on Tyrants and Despots: A Study in Objectivity. Weisbaden. ———. 1985. Herodotus the Historian: His Problems, Method, and Originality. Norman, Okla. West, M.L. 1992. Iambi et Elegi Graeci. Vol. 2. Oxford. ———. 1993. “Simonides Redivivus.” ZPE 98:1– 14. West, S. 1985. “Herodotus’s Epigraphical Interests.” CQ 35:278– 305. ———. 1988. “The Scythian Ultimatum (Herodotus IV 131, 132).” JHS 58: 207– 11. Williams, B. 1993. Shame and Necessity. Berkeley. Wolff, E. 1934. “Das geschichtliche Verstehen in Tacitus Germania.” Hermes 69:121– 66. Wood, H. 1972. The Histories of Herodotus: An Analysis of the Formal Structure. The Hague. Zeitlin, F. 1984. “The Dynamics of Misogyny: Myth and Mythmaking in the Oresteia.” In Women in the Ancient World: The Arethusa Papers, ed. J. Peradotto and J.P. Sullivan, 159– 94. Albany. ———. 1990. “Thebes: Theater of Self and Society in Athenian Drama.” In Nothing to Do with Dionysus? Athenian Drama in Its Social Context, ed. J.J. Winkler and F. Zeitlin, 130– 67. Princeton.

General Index

Names of modern scholars are included only when they appear in the text. Addressee, 28n. 21, 29n. 22, 37. See also Audiences  αδικος, etc., 31n. 25, 71, 85, 147, 182, 186, 187, 188, 189– 90, 195, 196, 211– 12 Advertisements of narratability. See Glosses of celebration; Narratability Adyrmachidae, 82 Aeacidae, 191n. 146 Aegina, 42, 189– 91, 193 Aeginetans as sources, 222n. 220 Aeschylus, 239n. 25, 258 Aesop, 5, 7, 227n. 230 Agathyrsi, 79– 80 Ainos, 5– 8, 13, 50, 180n. 122, 227, 251, 253, 254, 263n. 101, 265, 271 α ι τ ι η, 31, 182, 188, 192, 221, 222, 223, 260, 262n. 98 αιτιος, 31n. 25, 182, 188, 189n. 142, 192, 207n. 182, 211– 12, 222, 224 η.

See Glosses of historie; Hearsay ακο Alcmaeonids, 55n. 37, 222n. 220, 237n. 17, 260– 64 Alexander of Macedon, 29, 39 Alexander of Troy, 142– 44, 185

etc., 173– 74, 178, 197– 98, αληθ ης, 258– 59 “All men,” 100, 169– 70, 171, 226, 228, 229– 30  αλλος, 31, 50

Also (ka ι
), 50, 56, 118, 137 Alterity. See Other and otherness Alternative versions, 35n. 42, 41, 222– 23, 224, 225– 26 Alyattes, tomb of, 104– 5, 106, 235, 237 Amazons, 109, 123– 32, 144, 202n. 169, 212, 220, 233n. 5, 255– 56 Anacharsis, 115– 16, 117, 118– 19, 120 αναγκ-, 213, 246, 257 of the narrator, 34n. 38, 94, 166, 173, 257– 59 Analepsis. See Narratives, analeptic Analogy, 46, 47– 50, 95 explicit, 50– 66, 73, 82, 90 horizontal, 46, 50, 73, 132 implicit, 66– 73, 123, 100– 132, 191n. 146, 219, 238, 255 vertical, 46, 50, 56, 61– 62, 73, 86, 88, 90, 122, 132, 169, 251– 59 See also Comparison; Similarity Anaximander, 85n. 125 ανδραγαθ ι η, 114, 150, 154 ανδρη ι η, 67, 102– 3, 137, 154n. 54 Androphagoi, 80, 108, 136– 37, 170 Animals, 46, 80, 88n. 131, 90nn. 135, 136, 95– 96, 97, 122n. 232, 236, 243– 51 worship of, 93– 95 See also Asses; Camels; Hares; Horses; Lions; Mules 293

294

General Index



ανθρωπ ηιον, etc., 99n. 169, 101, 107 Apis bull, 51n. 25 αποδεκ-, 7, 32, 119n. 221, 182  αρα, 92n. 151, 159, 187 Arabia, 75n. 88, 85, 233n. 4, 235– 36 Arbitration, 7– 8, 166, 217, 222– 31, 260 η,

50, 55, 203, 204, 216– 17, 226 αρχ Athenian, 58, 203 Arete (moral excellence or courage), 43, 44n. 68, 51, 114n. 205, 208, 211, 213. See also

ανδραγαθ ι η; ανδρη ηη Argippaeans, 79, 108, 137 Argives, 225– 29 as sources, 225, 226– 27 Argos, 61, 64, 112, 120n. 225, 225, 229 Arimaspi, 79, 83n. 117 Arion, 252– 55, 256, 259 Aristagoras, 69, 209– 10 Aristides, 43, 261n. 94 Aristotle, 178– 79, 234 Artabanus, 48n. 9, 101n. 171, 122, 146, 213, 245, 257n. 86 Artaxerxes, 203, 205, 226 Artemisia, 41n. 62, 112, 126n. 245, 255– 59, 271 Asses, 55, 64n. 65, 243, 248– 50 Assyria, 86 Assyrians, 8, 10–13, 91n. 143, 138n. 10 Astrabacus, 64n. 65 Astyages, 23– 24 Ate, 184, 200 Athenian demos, 53, 56– 58, 208– 10 Athenians, 38, 42, 57– 59, 86, 92n. 149, 105n. 179, 116– 17, 125, 132, 136n. 3, 189– 94, 197, 212n. 194, 213– 17, 218, 219, 220, 221 as sources, 174, 222n. 220, 223– 24 Athens, 3, 4, 6n. 16, 48n. 9, 52– 59, 63, 68, 90, 109– 10, 124, 172– 75, 178, 181, 189– 94, 198n. 158, 199, 202– 5, 206– 11, 255– 56, 260– 65

Athletics, 49, 61 Audiences encoded in metanarrative, 37– 38, 139, 233 historical, 4, 38, 40, 219, 266, 269– 71 implied, 6, 10n. 25, 14, 16, 19 internal, 4n. 10, 252n. 69, 248, 251, 255, 259, 263 knowledge, expectations, and attitudes of, 38, 74, 98, 101, 102, 137, 147, 161, 174, 198, 251 Autopsy, 35, 86, 143, 164, 241 Babies, 63n. 62 Babylon, 9, 12, 14, 51– 52, 86, 104, 156n. 63, 241 Babylonians, 77, 78, 92n. 151, 98, 167n. 91 “Bad” as label applied to actions and/ or customs, 42, 137, 140, 156, 163, 209 Bald Men. See Argippaeans Balinese, 157– 58 Banquets, 69, 70, 155, 263 Barbarians (non-Greeks) attitudes of, 145– 54, 161– 63 Greek attitudes toward, 8, 12, 69, 79– 80, 81, 87, 93n. 152, 103, 107– 10, 114, 121, 124, 141– 42, 143, 147, 154– 55, 209– 10 as positive paradigms, 8, 12– 13, 106, 136– 37, 141, 142– 44, 212 in relation to Greeks, 8, 18, 31, 69n. 80, 76– 77, 91, 92, 97, 101– 32, 133, 135– 36, 155– 56, 211, 239– 40 Barbarity, 58, 122, 135, 141, 155, 253 Barbaros, 101, 135 Beliefs cultural, 34n. 40, 148, 149, 158– 59, 161

General Index Beliefs (continued) of the narrator (see Glosses, selfreferential, of opinion;

νοµηζειν) religious, 164– 66 Benedict, R., 13, 14 Blame, 16, 42, 218, 224. See also Glosses, referential, evaluations of worth Boas, G., 79 Boats, 90 Assyrian, 1, 8, 10– 14, 15, 26, 233, 241 Boundaries, 85– 86, 103, 204n. 176 violations of, 11n. 26, 48, 49, 50, 131, 135, 200, 201, 250n. 63 See also Rivers Bravery, 43 See also ανδραγαθ ι η; ανδρη ι η; Arete Buto, sanctuary of Apollo and Artemis at, 237– 39 Callatians, 167n. 91, 170, 236n. 16 Callias of Athens, 226, 229n. 237, 260– 61 peace of, 131n. 258, 229n. 237, 261n. 94 Callias of Elis, 72n. 83 Callicles, 171 Cambyses, 51n. 25, 57n. 43, 63n. 61, 65, 166, 168– 72, 229n. 237, 245n. 47 Camels, 244– 47, 249 Candaules, 28n. 20, 40, 42n. 64 Cannibalism, 77, 137, 161– 63, 170, 172 Causality, 30, 172, 183, 188, 190, 192, 243– 44, 245, 246, 249, 251, 253, 269 Celebration, 13, 29n. 22, 43– 44, 145n. 32, 232– 65 Celebratory code. See Codes, celebratory Celebratory glosses. See Glosses, referential, of celebration

295

Character identification of, 39 narrator as, 35 self-referential, 255, 264 Chauvinism. See Ethnocentrism Chemmis, 238– 39, 254 Chians, 195 Choaspes River, 9, 11 Classification, 46, 47– 48 Cleisthenes of Athens, 52– 57, 68, 262 Cleisthenes of Sicyon, 52, 54– 56, 68, 262, 263 Cleomenes, 65, 68, 169n. 95, 177, 209, 225 Climate, 75n. 88, 87– 88 Codes, 10– 11nn. 25, 26, 28, 41, 48, 102, 219 of animals, 252 celebratory, 31, 32, 44, 135, 235, 252, 254, 259 conventional, of wonders, 235 of customs, 40 ethical, 11nn. 27, 28, 30, 31, 70, 87, 208, 215 historical, 31 of historie, 33, 143n. 24, 159 juridical, 31, 70 of kingship (monarchical), 11, 28, 49n. 21, 64, 65, 104, 122, 154n. 54, 171n. 100, 241 mythical, 60 of narration, 33 political, 11n. 27, 44n. 20, 146, 174n. 110, 126 of refutation, 254, 259 of sacrifice, 161 scientific, 87, 244 Scythian, 114– 15 semic (indices), 10n. 25, 12n. 30, 32n. 28, 46, 49, 54, 61– 62, 63, 68– 69, 70n. 81, 122, 155, 254, 257, 264 Spartan, 112n. 198, 114– 15 symbolic, 10n. 25, 11nn. 27, 28, 49, 61– 62, 131, 218, 244, 246, 251 theological, 187, 244

296

General Index

Cognitive statements, 120, 140, 146– 48, 150– 51, 160, 162n. 72 definition of, 34n. 40 Coincidences, 50– 51, 193, 194 Common sense, 12 Comparison, 15– 16, 45– 133, 228, 232 diachronic, 16, 46– 47, 47– 73, 132 synchronic, 16, 46– 47, 73– 133 See also Analogy; Difference; Glosses, referential, of comparison Conclusions, 17, 23, 24, 41 mixed, 27 programmatic, 26, 30, 44, 164, 248 retrospective, 25, 28– 29, 42, 69, 74 summary, 25, 27– 28, 70, 248, 249 types and rhetorical value of, 24– 32 Conjecture, 36, 143 See also Glosses, self-referential, of opinion; συµβα λλεσθαι Connotation. See Codes, semic Constitutional Debate, 50n. 22, 101, 207, 210 Contemporary rhetoric, 174, 204n. 175, 215, 216, 217– 18, 229, 231 Contexts, 45n. 1, 48, 51, 90, 178, 205, 216– 17, 221, 234, 243, 267 Continents, 84– 85 Convictions and indictments, 222– 23, 224, 230 Corcyreans, 221, 224– 25 as sources, 224 Corinthians, 97, 174n. 107, 203n. 170, 223– 24, 258n. 89 as sources, 223 Croesus, 6n. 16, 11, 13, 25, 28, 30, 39, 50n. 24, 80n. 104, 102– 4, 106, 119n. 221, 135, 162n. 72, 182– 85, 188n. 138, 211– 12, 237n. 17, 245, 249, 262, 263 Culture, 40, 136, 168 material, 88– 89, 110, 137

in relation to climate, 87– 88 in relation to history, 106 See also Nomos Cultures hard, 77– 78, 79, 250 primitive, hard and soft. See Primitives soft, 77– 78, 80, 104– 6, 250 Customs. See Codes of customs; Nomoi Cypselus, 63n. 62, 246 Cyrene, 187, 188n. 136 Cyrus, 9– 11, 25, 33n. 32, 39, 41n. 62, 50n. 24, 51n. 27, 63, 102, 103, 130, 188n. 138, 245, 249, 250n. 63 Darius, 11, 63, 69, 88n. 131, 112, 171n. 100, 187, 188n. 138, 192, 202, 205, 233n. 5, 240, 248, 249

24 δε, Deception in Herodotus, 49, 57n. 44, 63, 71, 72, 135n. 2, 147, 148, 210 of Herodotus, 7, 19n. 49 δει , 34 Deioces, 48n. 8, 63– 64 De¨ıphonus, 70 Delos, 201, 204n. 175, 238– 39 earthquake of, 201– 5 Delphi, 6, 7, 53, 104, 124n. 239, 218, 247n. 55. See also Oracles Demaratus, 41n. 62, 64, 67, 111, 112n. 198, 208, 213, 246n. 50, 250n. 61 Demeter, 190, 194n. 151, 239n. 25, 243– 44 Democracy, 52– 53, 56– 59, 207– 11 Demonstratives, 25– 26, 28 Description, 21, 39, 74 nonevaluative, 156– 63 Detienne, M., 173 διαβαι νειν, 131, 214, 250n. 63 Difference, 46– 47, 51, 74– 76, 79, 89,

General Index 92n. 145, 95, 102, 126, 132, 133, 142n. 21, 163, 166, 167, 230, 235 See also Comparison; Other and otherness; Uniqueness Digressions, 242 See also Narratives, inserted δι κη, etc., 70– 73, 80, 130, 136– 37, 143– 44, 189– 90, 199– 200, 211 Dionysus, 55, 92, 119– 21, 146n. 33 Discourse, 14, 17n. 44, 19, 22, 24, 43, 51, 73, 93– 94, 95n. 156, 98, 156, 223– 25, 232, 244– 45, 248– 49, 251, 265, 266 definition of, 20n. 4 Disputes, 217– 31, 260 metanarrative, among sources, 35, 219, 221– 23, 225– 28 narrative, among historical agents, 35, 188, 219– 21, 222– 23, 224– 25 See also Stasis Dissoi Logoi, 169– 70, 228n. 236 Divine action, 71– 73, 121, 181, 195, 230– 31, 243, 253. See also Retribution Divine communication, 24, 71– 73, 181, 194– 206, 231 See also Dreams; Omens; Oracles Divine principle, 165, 169, 185, 243– 44, 247 Divine representations, 166, 238 Divine sphere, 95, 244, 247 δι ζοµαι, ε πιδι ζοµαι, 33, 37n. 55, 242

42n. 63, 54 δοκεω, Donkeys. See Asses Dorians, 55, 56, 203n. 172 Dorieus, 65n. 66, 72n. 83

etc., 28, 63, 207n. 183 δουλοω, Dreams, 23, 41, 49, 196, 197, 246 Dress, 69, 121, 122, 128 ethnographic description of, 81, 92n. 151, 121

δυναστευειν, 53, 209

297

Ecbatana, 64n. 63, 90 Egypt, 51, 75, 84, 86, 92, 103, 104, 142, 143, 168, 185– 86, 236n. 16, 237– 39, 241 Egyptians, 7, 13, 63, 92– 97, 98– 100 as critics, 145n. 31 as ethnographic subjects, 75– 76, 77, 78, 81n. 109, 87– 88, 118, 136n. 3, 141– 44, 167n. 91, 196, 223n. 225 as sources, 143, 239 ε ι κα ζω, 30, 60– 61, 62, 83n. 114, 184 Ephialtes, 17n. 47, 222– 23 Ephorus, 108, 141n. 17 Epitaphios. See Funeral oration ε πιθυµ- (desire), 91, 94, 98, 122, 164– 65, 171n. 100, 184 Eros, 65n. 67, 69 Ethea, 99, 119 in relation to nomoi, 137n. 7 Ethiopians, 75n. 88, 77, 79, 98n. 165, 108n. 186, 145n. 31, 167n. 91 Ethnocentrism, 118, 145– 46, 151, 164, 166, 171, 231 Ethnographic glosses. See Glosses, referential, ethnographic Ethnographic present. See Present tense, ethnographic Ethnography ancient Greek, 6, 107– 8, 141n. 17, 234, 235, 240, 241, 242 in Herodotus, 1– 2, 6– 8, 13– 14, 18, 74– 132, 134, 135– 36, 143, 153– 55, 168, 244, 254n. 75 modern, 12n. 31, 13– 14, 35, 42n. 65, 107n. 182, 141n. 18, 145– 46, 146n. 36, 156– 59, 160, 163n. 76 totalizing statements in, 118n. 220, 137, 138 See also Description ευδαιµον ι η (happiness), 101, 177, 184, 185, 186, 190, 269 instability of, 181– 83, 206, 230 Euenius, 70– 73 Euethie (simplemindedness), 142, 210

298

General Index

Euripides, 120 Evaluation (praise or blame), 30, 43, 119, 134– 35, 158– 59 in the ethnography, 135– 72 in the history, 172, 175– 78, 206– 11, 211– 17, 226, 227, 264 Evaluations. See Glosses, referential, evaluations of worth Evaluations of accuracy. See Glosses, self-referential, evaluations of accuracy Events after 479, 190n. 143, 193– 94, 198n. 158, 201– 5, 213, 219, 221, 229, 267– 71 Herodotus’ explicit references to, 45, 190, 191– 93, 204n. 176, 214, 215 Herodotus’ reticence about, 4, 6 See also Peloponnesian War Evidence, 180– 81, 182, 183, 186, 188, 190, 193, 196, 197, 198– 201, 206, 207, 209, 216, 256 See also Glosses, self-referential, of evidence  ι σκειν, 54, 55, 91n. ε ξευρι σκειν, ευρ 143, 105, 120, 137, 138, 268n. 4 Expansionism. See Imperialism and expansionism Explanation, 23, 43, 72– 73, 232 See also Glosses, referential, explanatory First person. See Person, first; Person, first plural Flashback, 39, 189. See also Narratives, analeptic Focalization, 23n. 12, 34n. 39, 144– 45, 148n. 39, 161 Focalizer, embedded, 23 narrator as primary, 23n. 12 Fornara, C., 3– 5, 6 Frame, metanarrative, 22n. 11, 28n. 20 Freedom, 63, 111– 12, 201, 211, 212,

213, 231, 237, 257– 58, 264– 65 Froidefond, C., 241 Funeral oration, 4, 125n. 241, 203n. 169, 220n. 216 Funeral rituals, 95n. 157, 101, 160– 63, 167– 72 γα ρ, 22, 27, 30, 244 Geertz, C., 157, 171 Gelon, 83n. 114, 218– 19 Geloni, 81, 92n. 146 Generalizations (gnomic statements), 178– 83 from a dream, 196 made by characters, 101n. 171, 122n. 232, 180, 184, 211– 16 of the narrator, 7, 41, 94, 100, 165, 168, 180– 81, 182, 185, 186– 87, 186n. 134, 194– 95, 197– 201, 202, 206– 10, 211n. 193, 212, 217, 226– 31 primary referent of, 227 Geography, 73, 74, 82– 87, 237 Getae, 101n. 172, 137, 164 Glaucus, 188– 89, 190 Glory. See Kleos Glosses, 14, 17, 23, 28 self-referential, 32– 38, 41, 42, 44, 55, 82, 182, 228 of corroboration, 35– 36, 38, 39n. 58, 101, 143, 147 evaluations of accuracy, 35– 36, 226 of evidence, 35– 36, 38, 39, 41, 118, 173, 187, 192– 93, 194– 95, 215 of historie (autopsy, hearsay, etc.), 31– 32, 35, 36nn. 49, 51, 37, 43, 70, 119, 143, 241 of identification, 39, 43, 58, 60, 69, 177– 78, 246, 255 of knowledge, 35– 36, 39, 41, 149 of narration, 30, 32– 34, 37, 38n. 56, 42, 93– 94, 226

General Index of noncorroboration, 35– 36, 226, 239, 240, 259 of not knowing (ignorance), 35– 36, 37, 85n. 123, 221, 226, 227, 261 of opinion, 36– 37, 41, 42, 43, 44, 83n. 114, 138, 149, 166n. 87, 173, 244, 254, 261 of refutation, 36, 41n. 62, 142, 143, 223, 260, 261 of rejection, 36, 141, 142, 143, 260, 261, 264 of source, 34– 35, 39, 83n. 115, 116, 141, 142, 143, 148, 241, 252 of testimony, 38– 39, 57n. 42, 64, 239n. 25, 253, 263 referential, 36, 38– 44, 95n. 156, 228 of anticipation of doom, 42n. 64, 120 of celebration, 36, 43– 44, 74, 75, 133, 235, 237– 38, 239, 241– 42, 247n. 55, 248, 253, 254, 256– 57 of comparison (analogy, similarity, difference or uniqueness), 40– 41, 45, 50, 54, 56, 59, 60, 75– 76, 77n. 95, 79, 80, 81, 82, 83n. 118, 89– 100, 100– 101, 118, 133, 137, 166n. 87, 217, 241, 260 ethnographic, 40, 69n. 80, 90, 102, 146– 47, 153n. 51 evaluations of worth, 36, 42– 43, 69, 76n. 93, 134, 136– 40, 150, 178, 206– 7, 208– 10, 226, 228 explanatory, 27, 36n. 47, 39– 41, 64, 69, 120, 244, 248, 252n. 68, 261 historical, 40, 105 historico-ethnographic, 103, 155 of interpretation, 36, 39n. 58, 41– 42, 53, 118, 132, 136n. 3, 137, 151, 165, 168, 173– 231

299

metalinguistic, 40n. 61, 94– 95 See also Conclusions; Introductions

γνωµη, 50n. 28, 59n. 51, 164, 173– 74, 176, 177– 78, 186, 193, 226 Gnomic statements. See Generalizations Gods. See Beliefs, religious; Divine action; Divine communication; Divine principle; Nomoi, religious Gold, 103, 104, 105, 235, 236 Golden Age, 79 “Good,” 42, 43, 93, 148, 155, 163, 209 γρα ␸ω, 33n. 30, 222 Greatness, 38n. 57, 44, 211, 240n. 33 Greeks defining characteristics of, 18, 105, 120 as sources, 223, 224– 28 Gyges, 28n. 20, 40, 42n. 64, 183 Hares, 22– 23, 34, 250 Hartog, 111, 156, 158, 234 Health, 88n. 131 Hearsay, 35, 156n. 63, 164, 239, 242 Hecataeus, 85n. 12, 222n. 220, 239, 240, 267n. 1 Hegemonie (leadership), 214– 17, 218, 225– 26 Hegesistratus, 60, 66– 68, 70 Helen, 142– 44, 185– 86, 193 Hellespont, 11, 135, 193, 200, 204 Hellman, F., 78 Helms, L. V., 157, 160, 161 Helots, 68n. 75, 70n. 81, 113 Heracles, 65, 66, 84n. 121, 94, 124n. 237, 128, 142, 143, 233n. 5, 239– 40 Hermotimus, 47 Herodotus as real author, 31, 257n. 88, 267– 73 See also Histor; Narrator

300

General Index

Heroes and heroic figures, 52, 55, 60, 65– 66, 67, 124, 176, 254 Heroic past. See Mythical past Hesiod, 5, 165 Hipparchus, 196, 246 Hippoclides, 263– 64 Hippocratics, 37n. 55, 84n. 120, 87, 90, 141n. 17, 159 Histor, 8n. 21, 24, 31– 32, 35, 49, 82n. 113, 141n. 20, 206, 209, 217, 221, 225, 257, 260, 267 audience as, 37 Historie, 6– 8, 31– 32, 96, 98, 132, 141, 146– 48, 166, 181, 182, 201, 221– 22, 224, 225, 227, 230, 254, 265 See also Codes of historie; Glosses, self-referential, of historie Histories Book 2 of, 5, 35n. 45, 96 dating, of, 3n. 6, 5, 267 end of, 15n. 40, 57– 58, 256 genesis of, 2n. 1, 5 message of, 4, 5– 6, 8, 16, 18, 24, 70, 93n. 153, 116, 132, 136, 146, 171, 178, 179– 81, 210, 212, 223, 230– 31, 253 moralism of, 18, 44, 88n. 131, 163, 184– 85, 266 obscurity of, 6, 7, 18 open-endedness of, 15n. 40 purpose of, 3, 4– 8, 12– 13 structure of, 15, 181 unity of, 2n. 1, 15– 16 Homer and Homeric poetry, 4, 14n. 37, 15, 24, 39n. 60, 43– 44, 90, 119, 143, 147n. 38, 165, 166, 176, 178, 186, 247 Iliad, 23n. 12, 107, 211, 218n. 211 Odyssey, 33n. 34, 182, 247 Homoioi, 111 Hoplites, 113, 114n. 205 Horses, 17, 22– 23, 34, 244n. 43, 245, 248– 50 Hubris, 6, 69, 155, 189, 190, 199–

200, 207, 210, 213, 215, 220, 248, 249, 250n. 63 Hyperboreans, 79, 108nn. 186, 187 Hypothetical constructions, 41, 169, 173, 175 Illocutionary acts, 16, 17, 179n. 119 Imperialism and expansionism, 13, 49, 50, 85, 188, 212 Athenian, 3, 57, 124, 156, 174n. 110, 204n. 175, 209– 10, 213– 14, 231 Persian, 11, 102, 122n. 232, 151– 53, 155, 187– 88 Index. See Codes, semic India, 235– 36 Indians, 91n. 143 See also Callatians; Padaeans Inquiry, 19, 188, 189, 230 See also Historie Interpretatio Graeca, 91 Interpretation, 13, 27, 28, 29n. 22, 30, 36, 43, 51, 54, 73, 82, 94, 132, 134– 35, 232, 233, 236, 245, 247, 248 in the ethnography, 168– 71 in the history, 172– 231 See also Glosses, referential, of interpretation Introductions, 17, 23, 24, 32, 41, 237 mixed, 27, 30 programmatic, 26, 29n. 22, 30– 32, 36n. 51, 44, 94, 181, 183, 235, 242 prospective, 26, 29, 30, 42, 160 summary, 26n. 16, 27, 29– 30, 95n. 156, 184, 192, 244n. 44 types and rhetorical value of, 24– 32 Inventions, 92n. 149, 136  ι σκειν See also ε ξευρι σκειν, ευρ Ionia, 50, 188n. 138, 233n. 6 Ionians, 52, 53– 54, 56, 57n. 42, 83n. 114, 88n. 131, 92n. 151, 105n. 179, 106, 111– 12, 136n. 3, 138n. 11, 146n. 33, 195,

General Index 204– 5, 209– 10, 212, 216, 221, 267 as sources, 221, 237 Issedones, 91, 108n. 187, 137, 167n. 91 Ister, 75n. 88, 84 Jacoby, F., 5 Jokes, 69n. 79, 112n. 198 κακα (evils), 195, 202– 5, 206, 211– 12, 214– 15, 225, 226, 227, 228– 29 Kings, 102, 176, 184, 185, 241, 262 as founders, 105, 188 as inquirers, 49, 141, 166, 168, 170, 252n. 69 in relation to citizens, 13, 62, 66, 121– 22, 178 three-king period, 203– 5 Kingship, 9– 13, 49– 50, 60– 66, 73, 132, 136, 153– 54, 196, 207, 237, 245 dual, 62, 64– 65 See also Codes of kingship; Monarchical model; Tyranny Kleos, 6, 65, 176– 77, 211 κως, 41, 86– 87, 195 κου, 41n. 62 Labyrinth, 241– 42 Lacedaemonians. See Spartans Laughter and derision, 55, 70n. 81, 169 of the narrator, 85 Leonidas, 65– 66, 69n. 80, 172, 176– 78, 246– 47 Leotychides, 4n. 10, 51n. 27, 65nn. 66, 68, 188– 89 Libya, 84, 86n. 130, 105n. 179, 156n. 63 Libyans, 79, 80, 92n. 151, 147 Linus, 98– 99, 233n. 4, 238– 39 Lions, 58, 171n. 100, 196, 244– 47, 249, 260

301

Literary theory, 2, 8n. 21, 10n. 25, 12n. 30, 15n. 40, 16– 17, 28n. 21, 42n. 64, 248n. 57 See also Narratology; Speech-act theory Lloyd, A. B., 241 Logios, 35, 117, 143n. 24 Logoi reports of sources, 33, 34, 37, 41, 56n. 42, 141– 42, 166, 175, 222– 24, 225– 30, 237, 259 semiautonomous narratives, 2n. 1, 13, 15nn. 39, 40, 20n. 4, 33, 122, 123, 126, 130, 141, 186n. 133, 187, 193, 234n. 9, 256 Logos, of Herodotus, 33– 34, 44, 94, 135, 173, 182– 83, 188, 191, 196, 201, 206, 218, 242, 252, 253, 257, 258 Lovejoy, A. O., 79 Luxury, 69, 79– 80, 250 Lydia, 235, 236– 37, 246 Lydian appendix, 103– 6 Lydians, 25, 28, 40, 80, 123, 245, 246, 249 as ethnographic subjects, 76n. 93, 77, 78, 92n. 151, 97, 102– 7, 136, 150, 153, 167n. 91 as historical agents, 136 Magi, 77n. 95, 166n. 89, 167 Maps, 84– 85, 209 Marathon, 203, 246n. 50, 247n. 55, 258, 260, 264 Mardonius, 135, 146, 147, 153, 233n. 6 Marketplace, 140, 147, 148 Marriage, 68– 69, 127, 139 irregular, 49, 65, 121 Marriage customs. See Nomoi, sexual Massagetae, 11 as ethnographic subjects, 77, 81, 97– 98, 161– 63, 164, 167n. 91 Mataios, 141, 146

302

General Index

Medes, 9, 39, 92n. 147, 109, 136, 152– 53 Media, 63 Medism, 218, 220– 21, 222– 25 Megistias, 66 Melampus, 60– 62, 64– 66, 70, 71 Menalaus, 144, 186 Metanarrative, 14, 16– 17, 20– 44, 156, 254 definition of, 22n. 11 level, as opposed to narrative, 120, 218, 219 levels, 259 referential, 38– 44, 134 self-referential, 20, 30, 32– 38, 82– 83, 181, 186, 187, 191– 92, 194, 195, 197– 98 Metaphors, 50, 56– 57, 61– 63, 66, 73, 83n. 114, 91, 140, 153, 200, 219, 247, 249, 250– 51, 258 Miltiades, 57, 210 Mimesis within the narrative, 52, 54, 55– 56, 60– 61, 68, 89 narrative as, 17n. 44, 21, 261 Monarchical code. See Codes of kingship Monarchical model, 49, 52– 70, 105, 121– 23, 135, 168, 188, 200, 219 Monarchy. See Kingship Motives of characters, 41n. 62, 42, 172, 173, 175  µουνος, 47n. 6, 59 Mules, 23, 63, 64, 242– 44, 247– 50 Muller, ¨ K., 78 muthos, 84n. 121 Mutilations, 49, 69, 104, 110, 121, 141, 153n. 51, 154, 169n. 94 of self, 67, 154, 167 Mycale, 51nn. 25, 27, 70, 194– 95, 197, 204 Mythical past, 6, 61, 80, 142– 44, 185– 86, 218, 219, 220

Myths, 94, 124– 25, 128, 132, 166n. 89, 238, 239n. 24, 253 of resistance, 120 Nagy, G., 5– 8, 14, 174, 227 Narratability, 44, 54, 59, 97, 98, 211, 232, 252, 254 advertisements of, 31, 43– 44, 47, 67, 232, 242 See also Glosses, referential, of celebration Narrated, the, 38 world of, 32, 43– 44, 218, 221, 228, 232, 247, 260 Narratee, 37n. 54 See also Audiences Narration, 26, 33– 34 as a journey, 33, 242 real world of, 45, 228 time of, 21, 39, 48, 57n. 42, 102, 123, 146, 204, 205– 6, 216– 17, 217, 218, 221 verbs denoting, 33nn. 30– 33 See also Codes of narration; Glosses, self-referential, of narration Narrative, 16– 17, 20– 24, 101, 248 boundaries of, 45 definition of, 21 directions for reading, 27, 35, 38, 51, 58, 120, 232 main, 38, 70 “pure,” nonnarrated, or minimally narrated, 16n. 43, 17, 21, 23, 41, 159, 244 recipient of (see Audiences) summary, 27 Narrative levels, 20, 23, 32, 35, 38, 120, 145, 217– 18, 255 Narratives, 15, 33, 39, 40, 180, 225, 261, 263 anachronic, 21n. 6, 252n. 68 analeptic, 21n. 6, 54, 59, 62n. 59, 67 connections among, 21, 46, 51, 54, 252

General Index goal of, 4 historico-ethnographic, 105, 118, 123 inserted, 39, 59 personal (of historie), 35 proleptic, 21n. 6, 59, 62n. 59, 67 See also Logoi Narratology, 19n. 49, 20, 20n. 4, 21nn. 7, 9, 23n. 12, 27n. 17, 37n. 55, 41n. 62, 45n. 1, 241 Narrator, 8n. 21, 14, 16– 17, 23n. 12, 30, 32, 33– 34, 37, 38, 47, 73, 98, 102, 119n. 221, 132, 134, 135, 159– 60, 166, 169, 180– 81, 198, 241– 42, 246, 255– 59, 264– 65 as adviser, 6, 16, 17, 179– 81 ambivalence of, 93 apologies of, 83n. 115, 94, 166n. 89 corrections of, 69, 81, 143, 161 invisible, 23, 218, 221, 225 omniscient, 42n. 63 as prophet, 50, 205– 6 in relation to the speakers, 187, 207 as warner, 6, 16, 179, 266 See also Beliefs; Character; Focalizer; Histor; Laughter and derision; Obligation of the narrator or his logos; Polemics; Relativism; Subjectivity Negation, 41, 74, 93, 129, 146– 49, 150, 151, 155, 159, 161, 162, 188n. 139, 245, 248 in a gloss of narration, 32, 38n. 56, 93– 94, 198 Nemesis, 30, 184 Nile, 75– 76, 83– 84, 237, 239, 241 Nitocris, 9, 12, 51– 52, 241 Nomads, 80, 81, 107, 114, 250 νοµι ζειν, 37, 164 Nomoi (customs), 7, 54n. 35, 55, 88, 99, 142, 149, 229, 232, 244 borrowing of, 80n. 105, 92, 117, 152– 53 dietary, 77, 161, 163

303

Greek, 76, 91, 96– 97, 145– 46, 147n. 37 medical, 139– 40 religious, 77, 92– 95, 120, 137, 139, 149, 168, 169 secular, 119 sexual, 77, 79, 81, 82, 92n. 148, 98, 138– 39, 155n. 59 value of, 168– 72 See also Funeral rituals; Sacrifice Nomos, 85, 88– 100, 111, 120, 168– 71, 208, 236, 252, 253, 259 in relation to phusis, 75, 99– 100, 168– 69 Non-Greeks. See Barbarians Obligation of the narrator or his logos, 34, 94, 174, 242, 258– 59. See also αναγκOcean, 84, 237, 239 Oikoumene, 78, 85– 86, 123 Omens, 22– 23, 41, 175, 195, 201– 6, 245, 247, 249, 251  οψις. See Autopsy; Glosses, selfreferential, of historie Oracles, 41, 46n. 5, 49, 70– 73, 165, 196, 197– 201, 247n. 55 Delphic, 55, 58, 59, 62, 80n. 104, 175– 76, 183– 84, 188, 189, 195n. 153, 197, 198n. 158, 199, 246, 249 Orality in relation to the Histories, 14– 15, 33 Orientalism, 136, 153 ! οστις, 37 Otanes, 49– 50, 64n. 63, 65n. 67, 130n. 235 Other and otherness, 18, 91, 96, 104, 107, 110, 118, 123, 130, 133, 156, 234n. 9, 256 Padaeans, 162– 63, 167n. 91 Panhellenism, 174, 178, 215, 267n. 2 Paradigms, 168, 180, 186, 188, 195, 200, 204, 206, 222 barbarians as (see Barbarians)

304

General Index

Paradoxes, 13, 42, 53n. 34, 65n. 69, 122, 166, 210, 236, 256 Past, 39, 216, 218 tense, 21, 35, 178, 242n. 39 Patterns, 48, 50n. 22, 63, 68, 86, 112, 133, 143, 235. See also Analogy Pausanias, 68– 70, 90, 115, 121– 22, 135, 177, 215, 246n. 50 Peloponnesian War, 3, 116, 146, 193, 201, 203– 5, 213, 220, 221, 270 Performance, 5, 14– 17, 28, 32, 260, 266 Pericles, 58, 146, 202n. 169, 203n. 170, 219, 246n. 48, 262, 264, 267n. 2 Persian army, 11, 82n. 111, 131n. 258, 152, 244– 45 Persians, 1, 13, 25, 28, 29, 66, 69, 80n. 104, 101, 102, 123, 137, 153, 154, 175, 177, 187, 188, 197, 200– 201, 209, 217, 225, 229, 243– 44, 247– 50, 260, 262 as critics, 146– 47, 148 in an ethnographic context, 77, 78, 80n. 104, 85, 90, 97, 140, 149– 56, 167 as sources, 35 Person first, 17, 26, 34, 36n. 51, 149, 241– 42 first plural, 34n. 35, 37n. 54, 241– 42 second, 37, 119, 139 third imperative, 119 Phatic function, 28 Pheretime, 187– 88 Phoenicians, 35, 92n. 151, 98, 147 Phthon- (envy), 79, 174, 271 divine, 184– 85, 187 Phusis, 76, 83, 88n. 131, 95, 99– 100, 142, 143, 169, 236, 247n. 54, 251n. 67

summary of use of in Herodotus, 99n. 169 Pindar, 5– 6, 7, 39n. 58, 44, 72, 170, 271 Pisistratids, 67, 196, 260 Pisistratus, 47, 57, 63, 64, 207, 250n. 61 Plataea, 51n. 25, 59, 60, 61, 66– 70, 174n. 107, 177, 194– 95, 220 Plato, 171, 252 Plutarch, 7, 13, 143, 155, 175, 263, 264, 271 Polarities, 75, 76, 124, 132 Polemics of ethnographic subjects, 145– 46, 149, 151 of the narrator, 35n. 45, 75n. 87, 84, 141, 146– 47, 237, 239, 241 Praise of referent, 218, 222, 264 (see also Glosses, referential, evaluations of worth) of subject (see Celebration; Glosses, referential, of celebration) Present of narration. See Narration, time of Present tense ethnographic, 21, 159 gnomic, 21n. 9, 178, 183 “Primitive opponent,” 49, 78, 250 Primitives, 76– 77, 95, 236n. 16 hard, 79– 80, 107– 8 soft, 79– 80, 107n. 185 Primitivity, 12, 126n. 245

Prodigy, 120. See also τερας Program/Programmatic statements. See Conclusions, programmatic; Glosses, self-referential, of narration; Introductions, programmatic Prolepsis. See Narratives, proleptic Prophecy. See Divine communication; Dreams; Narrator as prophet; Omens

General Index Prospective introductions. See Introductions, prospective Protagoras, 165, 166, 170n. 98, 207, 268– 69  πρωτος, 36n. 49 Proust, M., 15n. 40 Psammetichus, 84n. 121, 141 Puns, 69, 119. See also Jokes Pyramids, 241 Quarrels. See Disputes Queens, 9, 12, 51– 52 Questions, 29, 41, 142n. 21 Raaflaub, K., 6 Redfield, J., 77, 80 Relativism, 169– 71 cognitive, 163– 66 cultural, 167– 72 of the narrator, 18, 91, 94n. 154, 137, 138, 161, 230, 268 Requirements. See Obligation of the narrator or his logos Retribution divine, 6, 181, 183– 94, 206, 212, 223 human, 186– 88, 189– 91, 192– 93 See also τιµωρι η; τι σις Retrospective sentences. See Conclusions, retrospective Rhampsinitus, 63n. 62 Rhodopis, 26, 142n. 21 Rivers, 9– 12, 50, 75, 83– 85, 86, 88n. 131, 103, 131 Romm, J., 79 Rosaldo, R., 145 Rosellini, M., 77 Sacrifice, 77, 91, 135n. 1, 149, 154 human, 142, 144, 154, 157– 58, 159– 63, 165 Sa¨ıd, S., 77 Sais, 35n. 45 Salamis, 43, 51nn. 25, 27, 57n. 44,

305

173, 174, 199– 201, 223– 24, 258, 264 Salmoxis, 137 Sameness, 82, 96, 100, 107, 117, 123– 32, 133, 239, 256 Samian War, 202n. 169, 220 Saying, verbs of, 52nn. 30, 32 Scyles, 118, 119– 23 Scyllias, 254– 55 Scythia, 75, 86, 90, 137, 233n. 4, 239– 40, 242– 43 Scythians, 7, 13, 79, 123– 32, 137, 138, 154n. 58, 188n. 138, 248– 50 as critics, 120– 21, 126n. 33, 233n. 5 as ethnographic subjects, 75, 77, 79, 80, 81, 83n. 114, 91n. 143, 107– 23, 159– 60, 167n. 91, 212 Second person. See Person, second Seers, 59– 62, 65– 66, 66– 73 Self-referential characters. See Character, self-referential Self-referential metanarrative. See Metanarrative, self-referential Ships, 107, 126n. 245, 204, 256. See also Boats Similarity, 76, 82– 87, 88– 93, 97, 98, 100, 133, 235, 238, 239 See also Analogy; Glosses, referential, of comparison Slavery and enslavement, 102, 103, 111– 12, 153, 187, 200– 201,

207. See also δουλοω Solon, 4n. 10, 47, 92n. 149, 101n. 171, 117, 162n. 72, 184– 85, 187, 263 Sophie (wisdom, intelligence, cleverness), 10, 12, 32n. 28, 33n. 32, 51, 63, 71, 105, 115– 16, 117, 118, 138, 139, 185, 210 Sophists, 99, 165, 166, 169– 71, 198– 99, 207 Sophrosune, 12, 115– 16, 117, 118, 131

306

General Index

Sources as characters, 35 ethnographic subjects as, 148 historical agents as, 223– 24, 228 See also Glosses, self-referential, of source See also under specific ethnographic groups Sparta, 3, 48n. 9, 59– 70, 109– 10, 112– 16, 176– 78, 198n. 158, 209 Spartans, 38, 53, 121, 125, 135, 146, 189, 192– 93, 202n. 169, 208, 212– 13, 218, 221n. 218, 225, 227, 233n. 5, 245– 46 as ethnographic subjects, 76, 96– 97 as sources, 114 Spartiates, 113 Speakers, 48, 49– 50, 180, 183– 85, 187, 207 as narrators, 4, 188 Speaking in relation to the Histories. See Orality in relation to the Histories Speech act, 5, 14, 19, 32, 196 Speech-act theory, 16nn. 41, 42, 179, 222 Speeches, 10, 17n. 44, 177, 180, 182n. 124, 224 Sperthias and Boulis, 43, 147n. 37, 175, 192– 93, 246 Stasis, 53, 130, 214– 17, 258, 270 Sthenelaidas, 115n. 211, 174n. 109, 215, 220 Story, 20– 21, 23, 27, 32, 41, 44, 51, 60, 241 Subcodes. See Codes Subjectivity, 166, 170, 171, 181, 225– 30 of the audience, 34 of characters, 35, 207, 145– 46, 213 of ethnographic subjects, 145– 46 of the narrator, 94, 138, 247 of sources, 35, 221– 22, 225, 228 συµβα λλειν, 45n. 1, 83– 84, 85

συµβα λλεσθαι, 83n. 116, 126 Summarization, 23, 24, 25, 27, 30, 31, 32 element of, 25, 26, 28– 29, 41, 235 Summary conclusions. See Conclusions, summary Summary introductions. See Introductions, summary Summary narrative. See Narrative, summary Superlatives, 44n. 70, 47, 67 Symbolic actions, 111, 135 Symbolism, 52, 96, 212, 217, 221, 249– 50, 252, 255, 264 Symbols, 11n. 27, 122, 132, 155, 168, 204n. 176, 245, 256 See also Codes, symbolic Tellus of Athens, 13

τερας, 22, 23, 27, 95, 195– 96, 202, 243n. 41, 247 Text definition of, 20n. 4 types of, 19 Thebans, 189, 193n. 149, 229 Themistocles, 43, 48n. 8, 51n. 27, 57, 67, 180, 187, 207n. 183, 210, 221n. 218 Theognis, 5 Thermopylae, 112n. 198, 145n. 32, 172, 175– 78, 213, 222– 23, 233n. 5, 246– 47, 258 Third person imperative. See Person, third imperative Thoma, 44n. 70, 47n. 6, 177 See also Wonder; Wonders Thracians, 154n. 58 as ethnographic subjects, 74, 77, 79, 81n. 109, 96, 138n. 11, 160– 61, 162, 163, 164, 167n. 91 Thucydides, 17n. 45, 18n. 49, 23n. 12, 44, 130n. 255, 136, 146, 201, 208n. 184, 213, 215, 216, 220, 221, 261, 267n. 1

General Index Thurii, 267– 71 τιµωρι η, 184, 186, 187– 88, 189, 195. See also Retribution; τι σις τις, 37, 173, 174, 184 Tisamenus, 59– 62, 64– 66, 68, 72 τι σις, 183, 188n. 139, 189, 191. See also Retribution; τιµωρι η Tragedy, 48n. 9, 229 Translation, of culture, 82, 91, 114 linguistic, 40n. 61 See also Glosses, referential, metalingustic Troy, 80, 142– 44, 185, 190, 193, 195, 202n. 169 Truth, 35n. 46
See also αληθη ς Tyranny, 6, 13n. 32, 49, 56, 58, 62n. 58, 63, 64, 65n. 67, 69, 188n. 136, 196, 206, 209, 245 See also Kingship Tyrant city, 58, 174, 257, 264 Tyrants, 53, 54– 55, 56, 257, 262, 264 Uniqueness, 48, 59, 74, 75– 76, 77n. 95 War, 108n. 187, 110, 114, 125, 128, 129, 130– 31, 146, 186, 187, 202– 5, 208– 9, 211– 17, 218, 220, 229, 231, 246, 257

307

Women, 65n. 67, 102, 110, 112, 125, 129– 30 in ethnographic descriptions, 74, 78, 79, 81, 104, 129, 132, 139, 150, 160– 61 See also Nomoi, sexual Writing in relation to the Histories, 14, 33 Wonder metanarrative, 19, 26, 31, 44, 98, 177, 232– 65 narrative, 233n. 5, 243n. 41, 246, 247 Wonders ethnographic, 9, 13– 14, 75, 83n. 118, 95, 98– 99, 133, 233, 234– 42 historical, 71, 177– 78, 233, 243– 65 Xanthippus, 58, 262 Xenie, 142– 44 Xenophanes, 166 Xerxes, 11, 22– 24, 39, 41n. 62, 48n. 9, 51, 57, 68, 102, 111, 114, 122, 130, 135, 146, 152n. 47, 171n. 100, 172, 173, 174, 185, 188n. 138, 200, 202, 212n. 194, 213, 221, 224, 226, 233n. 5, 244, 249, 257, 271n. 14

Index of Passages

Aelian Varia Historia 4.1: 161n. 71 Aeschylus Agamemnon 717– 36: 245n. 47 766: 200n. 163; 1231– 32: 125n. 242 Choephorae 161– 63: 110n. 191 Eumenides 625– 28: 125n. 242 681– 706: 109, 125 700– 703: 109– 10 Persians: 133 Prometheus Bound 709– 14: 109n. 188 Seven against Thebes 727– 33: 229n. 237 Supplices 234– 37: 125n. 241 277– 90: 125n. 241 Aesop 266 Perry: 227n. 230 Anacreon frag 11b Page (⫽PMG 356): 114n. 208 Anaxagoras DK 59 B21a: 84n. 120 Anaxandrides frag. 39 Edmunds: 93n. 152 Antiphon DK 87 B44, frag. 2: 99n. 168, 231n. 241 Apollodorus 1.9.1– 12: 60n. 53 2.5.9: 124n. 237, 128n. 252 Aristophanes Acharnians 702– 3: 109n. 188 Ecclesiazusae 613– 35: 79n. 103, 139

Knights: 210 1111– 12: 58n. 49 Lysistrata 672– 80: 124n. 239, 125n. 241, 255n. 81 Thesmophoriazusae 1070– 1175: 109n. 188 Aristophanes the Boeotian FGrHist 379 F 5: 271n. 13 Aristotle Athenaion Politeia 22.1: 53n. 34 Metaphysics 1.2.8– 9: 234n. 8 Politics 2.1.4– 18: 79n. 103 2.5: 268n. 4, 269n. 7 2.7.6: 117n. 216 Rhetoric 1.7.34: 219n. 212 2.21.2: 178– 79 3.9.2: 272n. 16 3.10.7: 219n. 212 Bacchylides frag. 4.50– 51 SM: 60n. 53 Charon of Lampsacus FGrHist 262 F 1: 251n. 66 Critias DK 88 B32– 37: 114n. 208 frag. 6 West: 114n. 208 Demosthenes Epitaphios 60.4– 8: 125n. 241, 132n. 259 Diodorus 2.45.1– 3: 128n. 251 3.53.1– 2: 128n. 251 4.68.4– 5: 60n. 53 12.4: 131n. 258 309

310

Index of Passages

Diodorus (continued) 12.9– 11: 267n. 2 12.10– 12: 269n. 8, 270n. 9 12.35: 270n. 11 17.77.1– 3: 123n. 236 Diogenes Laertius 1.33: 124n. 240 Dissoi Logoi DK 90 2.14: 163n. 74 2.15: 155n. 59 2.18: 170, 228n. 236 2.26: 170n. 98, 228n. 236 Diyllus FGrHist 73 F 3: 15n. 38, 271n. 13 Ephorus FGrHist 70 F 42: 80n. 106, 108, 141n. 17 Euripides Andromache 448: 228n. 234 Bacchae: 120 Heraclides 215– 17: 124n. 237 Hercules Furens 408– 18: 124n. 237 Hippolytus 10: 124n. 237 305– 9: 124n. 237, 128n. 252 Ion 1144– 45: 124n. 237 Supplices 410– 25: 210n. 191 432– 41: 210n. 191 Eusebius Chronica Olympiaca 83.4 (83.3): 15n. 38, 271n. 13 Hecataeus FGrHist 1 F 185: 80n. 105 302: 237n. 19 305: 239n. 127 Hellanicus FGrHist 4 F 106: 124n. 237 323a F 16, F 17: 125n. 243 Heraclitus DK 22 B114: 169n. 97 Herodotus Book 1 First sentence: 13, 14nn. 35, 37, 30– 32, 43, 44, 47, 177 1.1: 31n. 25

1.4.4: 85 5.3– 4: 181– 83 5.3: 31n. 25, 33nn. 29, 31, 48, 61n. 55, 212 5.4: 179 6: 39n. 60 6.2: 103 7.1: 26n. 16, 28n. 20 8– 13: 63, 65n. 67 8.2: 42n. 64 10.3: 40, 76n. 93 12.2: 39n. 58 13.2: 184 14.1: 28n. 20, 215 14.3: 237n. 17 19.1– 21.1: 253n. 73 20: 36n. 49 23– 24: 21n. 6, 252– 55 23: 39n. 60 23.1: 34n .39 24.5: 10n. 10 24.8: 39n. 58 25.2: 47n. 6 26.3: 188n. 138 27.4: 28 29– 33: 117n. 217 29.1: 103, 119n. 221 30– 33: 263n. 99 30.1: 119n. 221 30.3– 5: 13 30.4– 5: 4n. 10 32: 184– 85 32.1: 182n. 124 32.2– 9: 162n. 72 34.1: 30, 184– 88 35.2: 92n. 145 46– 49: 197 46: 212n. 194 46.1: 188n. 138 50.1– 3: 237n. 17 51.1– 3: 237n. 17 51.5: 237n. 17 53.3: 197 55.2: 63n. 62, 80n. 104, 249n. 60 57.3: 81n. 107

Index of Passages 59– 63: 63 59.1: 207n. 183 59.4: 63, 250n. 61 59.5– 6: 210 59.6: 207n. 183 60.3: 12, 47, 55, 210 61.1: 65n. 67 64.2: 66 64.3: 262n. 98 66– 68: 212n. 197 66.4: 38 70.2– 3: 222n. 120 71: 155 71.1: 188n. 138 71.2– 4: 250 71.2: 12 71.4: 80n. 104 72: 103 73– 75.1: 188n. 138 73: 110n. 193 73.1: 11n. 26 74.1: 40n. 61 74.6: 92n. 145, 97 75: 11, 103 75.6: 41n. 62, 46n. 22 79.3: 103 80: 245, 249, 251n. 66 82– 83: 213n. 199 84.3: 246 85.1: 29n. 23 86– 91: 104 86.2: 41n. 62, 154n. 56 86.5: 50n. 24 87.3: 183n. 126 87.3– 4: 211– 12 89.1: 28n. 19 89.2: 99n. 169, 155, 249 91: 184, 212 91.4: 183n. 126 91.5– 6: 63n. 62, 249n. 60 92– 94: 25, 103– 6, 149n. 41 92.1: 27, 50, 237n. 17 92.4: 26 93.1: 235 94.1: 92n. 145, 235n. 11 94.1– 7: 28

311

94.1– 2: 92n. 151 94.2: 150n. 44 94.7: 25, 237n. 18 95.1: 23, 25, 33nn. 30, 32, 33 95.2: 63 96.1: 48n. 8, 63 96.2: 65n. 67 98.5: 90 103.1: 51n. 26 103.3– 106.1: 212n. 196 105.4: 37n. 54 106.2: 16n. 40 107– 8.3: 23 117.2: 33n. 32 125– 26: 155n. 60 131– 39: 149– 56 131.1: 75n. 87, 164, 166n. 89 132.2: 91 136.1: 91 136.2: 250n. 63 137.1: 42, 113, 140 138: 140 139: 61n. 55 140: 154n. 53, 167 140.2: 77n. 95 140.3: 33nn. 29, 30, 164n. 79 141: 251n. 67 142: 88n. 131 143.2: 88n. 131, 138n. 11 143.3: 57n. 42 147.2: 47n. 6 152: 213n. 199 153: 97 153.1: 227n. 232 155.4: 102n. 174 164: 204n. 176 167.1: 243n. 42 168: 204n. 176 169.1: 88n. 131, 208n. 185, 212 169.2: 50 170: 204n. 176, 212 171.4: 92n. 151 172.1: 77n. 95, 92n. 147, 148n. 39 173.4: 77n. 95 177: 211

312

Index of Passages

Herodotus Book 1 (continued) 178– 87: 9 179.1: 34 182.1– 2: 98n. 164 182.3: 98n. 165 183.1– 3: 156n. 63 183.3: 11, 42 184: 16n. 40, 33n. 29 184– 85.1: 51– 52 185.3: 232n. 2, 241 187: 11 188– 89: 9 189: 153n. 51 189.1: 249n. 59 189.2: 250n. 63 190– 92.1: 10 191.3: 51n. 27 192: 10, 17 193: 10 193.1– 2: 86 193.1: 16 193.3: 140n. 16 193.4: 38n. 56 193.5: 91n. 143 194: 1, 9– 14, 233, 250n. 61 194.1, 9, 13– 14, 26 194.2: 90 194.3: 241 195.1: 89 196: 139, 140 196.1– 4: 104 196.1: 10, 26, 37, 98n. 165 196.5: 104 197: 10 198: 98n. 164, 167n. 91 199: 138– 39 199.1: 162, 171 199.4: 37n. 55 199.5: 98n. 165 200: 10 201: 81 202.2: 91n. 143 203.2: 77, 136n. 5 207.2: 183 207.5: 102n. 174

207.6– 7: 155n. 59 208: 11 211: 155n. 59 212– 14: 102n. 174 215.1: 81 215.1– 2: 250n. 62 215.4: 250n. 62 216.1: 81, 98n. 164 216.2– 3: 161– 63 216.2: 167n. 91 Book 2: 13– 14, 57n. 45 2.2: 141 2.3.2: 94, 164– 65, 230, 259n. 91 2.4.1: 143n. 24 2.5: 84n. 121 3.1: 82n. 113 3.2: 33n. 30, 34n. 38, 37 4.1: 92n. 149 4.2: 92n. 149 5.1: 60– 61 5.2: 37n. 55 6– 11: 241n. 35 7.1: 90n. 139 10.1– 3: 75n. 89, 83n. 114, 84 11.3: 68n. 62 11.4: 41n. 62 12.2: 75n. 89 13.2– 14.1: 145n. 31 14.2: 75n. 89 15– 16: 84n. 121 15– 18: 241n. 35 15.2: 41n. 62 16: 86 16.1: 85n. 123 17.1: 86 17.2: 85n. 123 18.3: 85n. 123 19.2– 3: 75n. 89 20– 27: 83n. 118 20.1: 33n. 32, 84n. 121, 237 21: 84n. 121, 237, 259n. 93 22.1: 33n. 32 22.2: 41n. 62 22.3: 88n. 131 22.4: 36n. 47

Index of Passages 23: 36n. 47, 54n. 35, 84nn. 121, 122, 237 24.1: 75n. 89 25: 92n. 148 25.1: 83n. 118, 85n. 123 27: 85n. 123 28.2: 35n. 45 29.2: 90 29.3: 84n. 119 29.5: 37n. 55 30.1: 37n. 55 31: 37n. 54, 85n. 123 33.2: 84n. 120 34: 84n. 120 35: 96, 233 35.1: 33n. 30, 75, 103, 235n. 10 35.2– 36.4: 76, 87– 88 35.2: 148n. 39 36.2: 95 37.3: 147n. 37 37.4– 5: 147n. 37 40.1: 81n. 109 40.2: 85n. 123 41.3: 118n. 220 42.1: 81n. 109 42.3– 4: 94, 166n. 89 42.4: 81n. 107 43.2: 92n. 149 43.4– 44: 82n. 113 44.1: 35n. 45 45: 84n. 121 45.1– 3: 142 45.1: 210 45.2: 41n. 62, 99n. 169, 147n. 37 45.3: 34n. 36, 41n. 62, 166n. 89 46.1– 4: 93– 95 46.1: 147n. 37 47.2: 94n. 154 49: 92n. 149 49.2: 92n. 149, 120 50.1– 51.1: 92n. 151 50.1: 92n. 149 50.3: 92n. 151 52.2: 92n. 149 53: 165

313

53.5: 85n. 123 54– 57: 92n. 149 54: 147n. 37 56: 147n. 37 57.2: 41n. 62 58: 92n. 149 64.1: 196 65– 76: 95 65.2: 34n. 38, 94n. 154, 259n. 91 66.2: 95, 236n. 16 66.3: 95, 247n. 56 67.2: 89n. 136 68.2– 3: 89n. 135 69.3: 83n. 114 70.1: 33n. 30 73.2: 89n. 136 76.1: 89n. 135, 118n. 220 76.3: 89n. 135 77.1: 117n. 217, 142, 143n. 24 77.3: 140n. 16, 179 79.1– 2: 98– 100, 238 79.1: 147n. 3 80: 96n. 161 81.2: 92n. 150 82: 92n. 149 82.2: 153n. 48 91.1: 118n. 220 91.2: 96 91.4: 92n. 148 92.1: 54n. 35 92.2– 4: 89 92.2: 75n. 90 96.1: 89 97.2: 37n. 55 99.1: 33n. 31, 36n. 51 100.2: 51 101.1: 44n. 70 102.5: 102n. 174 104.1: 92n. 146 104.2: 75n. 88 104.4: 52n. 30 105: 81n. 107 108: 92n. 146, 250n. 62 109: 92n. 149 109.3: 92n. 151

314

Index of Passages

Herodotus Book 2 (continued) 110.1– 3: 51n. 26 112– 20: 142– 44, 185– 86 115.3: 33n. 32 117: 33n. 31, 164n. 79 118.1: 143 120.5: 185, 186, 193, 195, 200 121: 63n. 62 121δ: 250n. 61 123.1: 33n. 30, 34n. 38, 37, 228n. 233 123.1– 2: 92n. 149 123.3: 33n. 30, 39n. 58, 239n. 25 124.1: 51n. 26 125.7: 41n. 62 127.1: 51n. 26 129: 186n. 133 129.3– 130.1: 95n. 157 131.3: 39n. 58 133: 186n. 133 135: 92n. 148 135.3: 37n. 54 135.6: 26, 33n. 30 143– 46: 66n. 89 143: 35n. 45 145.3: 34n. 39 146.1: 32n. 28 147.2: 88n. 131, 136n. 3 148– 49: 241– 42 148.1: 44n. 70 150.2– 4: 98n. 166 154.4: 37n. 54 155– 56: 238– 39 156.2: 254n. 75 157: 211 160: 92n. 149 160.2– 4: 146n. 33 161.3, 42n. 64 162.1– 2: 63 164– 68: 268 167.1– 2: 97 170: 90n. 139 171.2– 3: 92n. 149 172.2: 63

175: 241n. 34 177.2: 92n. 149 Book 3 3.3: 169n. 94 6.2: 29n. 22, 37n. 55 8.1: 40n. 61 10.3: 243n. 41 11.3: 236n. 16 12: 83n. 118 12.1: 37n. 55, 236n. 16 16: 57n. 43, 154n. 54, 169n. 94 20.2: 77n. 95 22.3– 4: 145n. 31 23.2: 233n. 5 24: 167n. 91 24.2: 98n. 165 27– 29: 57n. 43, 166n. 87 27.3: 154n. 54 29: 154n. 54 30: 154n. 54 30.1: 169n. 94 31: 155n. 59 31.1: 154n. 54 31.4: 54n. 35 32: 154n. 54, 245n. 47, 250n. 67 32.4: 52n. 30 33: 169n. 94, 186n. 134 34.1– 3: 154n. 58 34.1: 169n. 94 35: 153n. 51, 154n. 54 35.5: 169n. 94 36: 153n. 51 37: 57n. 43 37.1: 169n. 94 37.2– 3: 89 37.2: 38n. 56 38: 118, 168– 72, 236 38.1: 229– 30 38.1– 2: 55 38.2: 36n. 51 38.4: 7, 39n. 58, 167n. 91, 230 39– 60: 21n. 6 40.2: 185n. 132 43.1: 185n. 132 45.3: 36n. 47

Index of Passages 46: 115n. 211 47.2: 232n. 2 50.1: 65n. 67 52.4– 5: 174n. 108 53.4: 65n. 67 55.1: 47n. 6, 112n. 198 55.2: 35n. 45 56: 224n. 226 57.4: 197 59– 63: 63 60.1: 33n. 30 64.3: 51n. 25, 57n. 43, 166n. 87, 186n. 134 65.3: 99n. 169 66.2: 57n. 43 68.1: 29n. 23 69.5: 154n. 54 70: 63n. 61 75.1– 2: 259n. 92 79.1: 154n. 54 79.3: 155n. 59 80– 83: 207 80: 50 80.1: 38n. 56, 101, 233n. 6 80.2: 200n. 164 80.4: 200 80.5: 55n. 37, 65n. 67 80.6: 207n. 182 81.1: 200n. 164 81.2: 210 82– 83.1: 63 83.1: 207n. 182 85– 87: 63 86: 153n. 48 95.2: 33n. 30 98.2: 85n. 123 98.3: 81n. 107 98.4: 91n. 143 99: 162– 63, 167n. 90 100: 89n. 136, 167 101.1: 77, 101 102.1: 92n. 145 102.2– 105: 235n. 12 102.2: 89n. 136 103: 38n. 56 104.2– 3: 75n. 88

315

106.1: 86– 87, 88n. 131, 195, 236n. 15 106.2: 89n. 135 107.1: 75n. 88 108: 95n. 158, 195, 236n. 15, 243 108.2: 41n. 62, 185 108.4: 245n. 46 111.1: 34n. 39, 36n. 51, 236n. 13 111.2: 233n. 4 112: 140n. 16, 233n. 4, 236n. 13 113.1: 75n. 88, 140, 232n. 2, 236n. 13 114: 235n. 12 116.1: 235n. 12 116.2: 83n. 117, 99n. 169 116.3: 37n. 54 118.1: 200n. 164 118.2: 154n. 54 123.2– 3: 92n. 150 125.3: 154n. 54 126.2: 200n. 164 127: 153n. 51 127.3: 200n. 164 134.5: 153n. 48 139.2: 63n. 61 143.2: 136.3 151.2: 249n. 60 152.1: 51n. 27 153: 249n. 60 154.2: 154n. 55 Book 4 1.1: 188n. 136 1.1– 2: 212n. 196 1.3: 110, 113 3.3: 113n. 200 4.2: 110, 113 4.3– 4: 110– 11 5– 10: 113n. 202 5– 12: 116n. 214 8– 10: 240n. 30 8.2: 84n. 122 9.1: 33n. 33 9.2: 233n. 5 10.3– 11.1: 240n. 30

316

Index of Passages

Herodotus Book 4 (continued) 11: 113 11.4: 39n. 58 12.1: 39n. 58 17: 82n. 111 18.3: 80 20.1: 113 22.1: 81 23: 80 23.2– 5: 137n. 7 23.2: 81n. 107 23.3: 89n. 135 23.5: 108n. 186 25.1: 36n. 50, 83n. 117 26: 167n. 91 26.2: 91, 108n. 186 28.1: 37n. 55 28.2: 75n. 88 28.3: 243n. 41 29: 88n. 131 30: 242– 43 30.1: 33n. 33, 242 31.2: 83n. 114, 115 33.5: 98n. 166, 108n. 187 36.1– 2: 84n. 122 36.2– 45: 85– 86, 237n. 20 36.2: 209n. 188 39: 99n. 169 42.1: 237n. 20, 259n. 93 43.7: 33n. 30 45.1: 85n. 123 46.1– 3: 117 46.1– 2: 138n. 9 46.2: 54n. 35, 212 46.3: 41n. 62 47– 59: 240n. 29 47.1: 86, 110 50.1: 75n. 89, 84n. 119 53.1: 75n. 89 53.3: 232n. 2 53.5: 84n. 119 54: 113 56: 85n. 123 59.1: 110, 138n. 9, 140n. 16 61: 110n. 193

61.1: 89, 89n. 135 61.1– 2: 138 62: 135n. 1 62.2: 115 63: 147n. 37 64– 65: 110n. 193 64.1: 114, 129 64.3: 159 65.2: 119, 148n. 39, 167n. 91 66: 114, 129 71– 72: 163n. 74 72.1: 113n. 201 73: 163n. 75 74: 90n. 139, 138n. 9, 159 75.1: 91n. 143 75.3: 138n. 9 76: 118– 19 76.1: 118, 147n. 37 77: 115– 16, 118 78– 80: 118, 119– 23 78.1: 51n. 28, 119 79.1: 42n. 64, 120 79.3: 54n. 35, 146n. 33 79.5: 162n. 72 81.3: 90, 115 81.4: 38n. 56 82: 33n. 31, 233n. 4, 239– 40 84: 153n. 51 93: 101n. 172, 137 94: 137n. 8 96.2: 33n. 31, 164 99.4– 5: 86 99.5: 38n. 56, 83n. 114 103.2: 34n. 40 104: 148n. 39 105: 34n. 39 105.2: 83n. 117 106: 80, 108n. 187, 137n. 6, 146n. 36 107: 80 108.2: 81n. 107 109.1: 81 110– 17: 123– 32, 255– 56 110.2: 113nn. 203, 205 111.1: 233n. 5 117: 81n. 107

Index of Passages 118.3– 5: 188n. 136 119.2– 4: 188n. 136, 212n. 196 120: 113n. 204 120.1: 116n. 215 122: 116n. 215 122.2: 61n. 55 125.1– 5: 249n. 59 127.1: 116n. 215 127.1– 3: 114n. 206 127.3: 212 127.4: 111 129: 248– 51 131– 32: 115 132.2: 250 134.1: 110n. 193, 250 134.3– 135.3: 250n. 64 137– 39: 88n. 131 142: 88n. 131, 112, 146n. 33 147.3: 65n. 66 156.1: 86n. 129 162.3: 188n. 136 163– 64: 188n. 136 165– 66: 187 166.1: 52 166.2: 39n. 58 168.1: 81n. 108 169.2: 81, 188n. 139 170: 92n. 148 172.1: 74 172.2: 98n. 164 173.4: 82n. 111 176: 156n. 63 177: 89n. 135 178: 80n. 105, 82n. 111, 156n. 63 179: 80n. 105 180.1: 82n. 111 180.4: 36n. 50, 92n. 149 180.5: 77n. 95 182: 86n. 129 183.1: 86n. 129 183.3: 89n. 135 183.4: 156n. 63 184.1: 75n. 88 184.3: 156n. 63 186.1: 92n. 145

187.1: 80 187.3: 140n. 16 189: 80n. 105 189.1: 92n. 151 190: 80n. 105, 97, 167n. 91 191.1: 80 191.2– 4: 80 191.4: 83n. 117 192.1: 89n. 136 192.2: 89n. 136 192.3: 89n. 136 195: 98n. 166 195.2: 228n. 233 195.4: 36n. 51 197.2: 81n. 109 198: 140n. 16, 167n. 91 199.1: 232n. 2 202– 3: 187 205: 187 Book 5 3.1: 138n. 11 3.2– 4.1: 81n. 109 4:167 4.1: 82n. 111 5: 160– 61, 162n. 72 6: 74, 97 8: 167 9: 92n. 147 9.3: 186n. 134 10: 88n. 131 12– 13: 153n. 48 13.1: 233n. 5 21.1: 25 22: 36n. 48, 64 22.2: 29 25: 154n. 54 28.1: 204n. 17 30.1: 204n. 173 32: 39n. 60, 69 40.2: 65n. 67 42.2: 65n. 66 44– 45: 222n. 220 45.1: 34n. 39 45.2: 34n. 39, 37, 72n. 83 47: 62n. 58 49.1: 209n. 188

317

318

Index of Passages

Herodotus Book 5 (continued) 49.2– 3: 209n. 189 49.3: 209n. 190 49.3– 6: 209 49.4: 209n. 190 49.5: 103 50: 213n. 199 56.1: 196, 246 58: 92n. 151 58.3: 39n. 58 59: 90n. 139 65.5– 66.1: 52 66– 70.1: 52– 57 67.1: 54, 60 68.1: 250n. 61 69.1: 56, 60 71: 62n. 58 72.1– 2: 53 72.4: 25n. 15 73.3: 262n. 97 75.1: 213n. 199 76: 213n. 199 77: 214n. 202 77.1: 177, 213n. 199 77.3: 38 77.4: 200n. 164 78: 53, 206– 9, 213n. 201 79– 81: 191n. 146 79.1: 189, 199n. 161 81.2: 190 82– 88: 189, 191n. 146 83.2: 190 85– 86: 191n. 145 87.1– 2: 191n. 145 88.1: 92n. 151 89: 191n. 146 89.2– 90.1: 189 89.2: 190, 199n. 161 89.3: 191n. 145 90.2: 66n. 73 91– 93: 213n. 199 92β3: 58, 246 92δ1: 42n. 64 92γ2– 4: 63n. 62 92.η1– 3: 65n. 67

92: 4n. 10 93.2: 66n. 73 97: 214 97.1: 53n. 32, 209n. 187, 210 97.2: 36n. 51, 207– 10 97.3: 204, 210, 211 101.2: 103 118.2: 99n. 169 Book 6 11– 12: 88n. 131 14.1: 34n. 39, 36n. 50, 221 14.3: 39n. 58 19.2: 195n. 153, 199n. 161 21.2: 227n. 232 27: 195 30.1: 154n. 54 32: 50, 154n. 54 35: 39n. 60 36: 62n. 58 39.2: 53n. 32 43.3: 38n. 56, 101, 203n. 170, 207n. 182, 233n. 6, 259 49– 50: 189 57.2: 66n. 73 58.2: 76n. 94, 96n. 161 59: 96n. 161 60: 96n. 161 61.1: 41n. 62 62: 65n. 67 65: 65n. 66 65.2: 65n. 67 66: 197 66.2: 53n. 32 68– 69: 250n. 61 68.2: 64n. 65 69.4: 64n. 65 70: 65n. 68, 67 70.2: 68n. 75 72: 65n. 68 73.1– 2: 189 74.2: 208 75.2– 3: 68n. 75 77: 112, 208 83: 112 86α–δ: 4n. 10, 188– 90 86– 90: 189

Index of Passages 87: 190, 200n. 164 91.1: 190– 91, 206n. 180 92.1: 189 94.1: 67 98: 201– 5, 214 98.1: 41n. 62 98.2: 211, 217, 227n. 231, 229 102: 67 103: 62n. 58 106: 213n. 199 107.1: 67 108: 221n. 218 109.3: 213n. 201 115– 16: 260 116: 194n. 151 117.2: 247n. 55 119.4: 81n. 107 121– 31: 222n. 220, 259– 65 121: 229n. 239 121.1: 260 122: 62n. 58 125– 26: 62n. 58 125.4: 103, 237n. 17 127.3: 200n. 164 131: 39n. 60 131.1– 2: 58 131.1: 207n. 182 131.2: 246n. 48 132– 36: 57n. 43 132: 57, 210 135.3: 42n. 64 136.1: 210 137: 222n. 220 137.4: 219n. 214 139.1: 243n. 42 Book 7 5– 9: 188n. 138, 203n. 170 5.3: 153n. 48 6.2– 5: 67 6.3– 4: 66 6.3– 5: 197 7.3: 67 8α2: 153n. 48 10α– γ: 48n. 8 10δ1: 213n. 198, 257n. 86 10ε: 122, 185

319

10η1: 146n. 34 10θ3: 167n. 90 11.1: 102n. 174 11.2: 212n. 194 11.2– 4: 188n. 138 11.3: 130n. 255 12– 18: 197 16α1: 99n. 169 16α2: 200n. 164 18.2: 122n. 232, 182n. 124, 185 18.3: 213n. 198 20.2: 44n. 69 21.1: 11n. 29, 29n. 22 26– 31: 11n. 29 35: 11, 153 35.2: 135 35.3: 154n. 54 38– 39: 153n. 51 39.3: 154n. 54 40.2– 4: 250n. 63 40.2: 45– 53, 185n. 132 43.1: 11n. 29 44: 152n. 47 49.5: 245n. 45 50.4: 245n. 45 54.3: 41n. 62 57: 51n. 28, 247 57.1: 22– 23, 95n. 156, 250 57.2: 249n. 60 57.2– 58.1: 23 58.3: 11n. 29 61– 95: 82n. 111 62.1: 82n. 111 63.4: 82n. 111 70.1: 81n. 107 74.1: 92n. 145 83.2: 245n. 45 85.3: 81n. 107 99.1: 34n. 38, 102n. 174 101– 4: 208 102.1: 88n. 131 103.3: 145n. 32 103.4: 99n. 169, 257n. 86 104: 114n. 205 104.3: 213 104.4: 111

320

Index of Passages

Herodotus Book 7 (continued) 106– 7: 51 107.1: 47n. 6 107.2: 154n. 55 108.1: 256n. 87 108.2– 109.1: 11n. 29 109.2: 11n. 29 110: 256n. 87 111: 98n. 165 114: 154n. 56 114.2: 51n. 28 118– 20: 245n. 45 119: 155n. 59 125: 27n. 18, 244– 47, 249 127.2: 11n. 29 131: 192 132.2: 218n. 210, 256n. 87 133: 191– 92 133.2: 256n. 85 134– 36: 175 134.2: 99n. 169 135.1: 43, 155n. 59, 246 135.3: 208 136.1: 256n. 87 136.2: 112n. 198, 192 136.3: 115n. 211 137– 38.1: 192– 94, 206, 229n. 238 139.1– 5: 173– 75, 194, 258 139.1: 32n. 28, 33n. 30, 34n. 38, 38, 173– 75, 171n. 14, 198 139.2– 4: 41n. 62 139.3: 256n. 87 139.5: 204, 215, 261 139.6: 197 140– 41: 197 142– 43: 197, 203 142.3– 143.1: 66 143.1: 48n. 8 144: 209n. 186 144.1: 51n. 28 144.2: 42, 191, 204 148– 52: 225– 30 148.3: 197 151: 226, 229n. 239, 261n. 94

152: 226– 30 152.3: 34n. 38 153: 211n. 193, 247n. 54 157– 62: 218– 19 161.3: 132n. 259 162.2: 83n. 114 166: 51n. 25, 194n. 151 167.1: 36n. 51 167.2: 39n. 58 168: 220– 21, 224– 25 171.2: 243n. 42 172.1: 256n. 87 178.2: 39n. 58 180: 154n. 56 181.2– 3: 154n. 55 187: 245n. 45 187.1: 249n. 58 190: 255n. 77 191.2: 41n. 62 194.2: 153n. 51, 154n. 54 204– 5.1: 178 204: 176n. 114, 246 205.3: 229n. 237 208.1– 3: 145n. 32 208.1: 176n. 114 208.3: 233n. 5, 247 209.3: 114n. 205 210.2: 102n. 174 211.3: 246n. 51 213– 14: 222– 23 213.3: 15n. 40 214.3: 17n. 47, 33n. 30, 222, 261 215: 176n. 114 219.2: 175, 208n. 185 220: 175– 78 220.4: 146 222: 229n. 237 223– 25: 176n. 114 223.2– 3: 246n. 51 223.2: 176n. 114 225.2: 246 226.1– 2: 115n. 211 226.2: 33n. 30 233: 193n. 149, 206, 229n. 237 234.2: 112n. 198

Index of Passages 235: 203n. 170 236.1: 174n. 108 237.2– 3: 271n. 15 238.2: 153n. 51, 154n. 54 239.1: 33n. 30 239.2: 36n. 51, 41n. 62 Book 8 2.2: 217 3: 214– 17 3.1: 130n. 256, 227, 227n. 231 3.2: 69, 200n. 164 4– 5: 57n. 43 8: 254– 55 8.3: 32n. 28 15.1: 194n. 150 22.2: 256n. 87 27: 66 37– 38: 247n. 55 38: 99n. 169 41: 116n. 215 42: 51n. 27 56– 64: 221n. 218 57.2: 208n. 185 59– 61: 223 59– 62: 224n. 227 62.2– 63: 57n. 43 66.1: 51n. 27 67– 69: 256n. 87 68α1: 102n. 174 68γ: 112 69.1: 271n. 15 75– 76: 57n. 43 75: 221n. 218 77: 197– 201 78: 220n. 215 79.1: 39n. 60, 43, 261n. 94 83.2: 99n. 169, 146n. 191 84.1– 2: 224n. 227 85: 224n. 227 85.2: 33n. 30 87– 88: 256, 258 87.3: 41n. 62 88.3: 102n. 174 90.3: 154n. 54 92: 224n. 227 94: 223– 24

321

98.1: 54n. 35 98.2: 90– 91 99: 256– 57 105.1: 47 105.2: 76n. 93, 155n. 59 108– 9: 204, 207n. 183 109– 10.1: 57n. 43 109.3: 180, 185, 187 109.5: 57, 67 110.1: 210 111– 12: 57n. 43, 204, 218n. 210, 221n. 217 112: 154n. 54 118: 153n. 51 122.3: 130n. 255 124.2: 43n. 67 129.4: 165n. 84 140– 49.11: 135n. 2, 221n. 218 140α2: 256n. 87 142.3: 207n. 183 143– 44: 208n. 184 221: 66 238.2: 69n. 80 Book 9 2.3: 53n. 32 6– 8: 213n. 199 7: 208n. 184 11: 135n. 2, 208n. 184 11.2: 115n. 211, 218n. 210 17.1: 256n. 87 26– 27: 219– 20 27.2: 200n. 164 27.4: 125n. 241 33– 35: 59– 61 34.1: 60, 83n. 114 35: 59 35.1: 47, 61 35.2: 205n. 178 37– 38.1: 66– 68 46.2: 246 48.3: 177 54.1: 135n. 2 62: 220n. 215 64.1: 69, 177 65.2: 166n. 87, 243– 44 68: 41

322

Index of Passages

Herodotus Book 9 (continued) 73: 205n. 178 73.2: 200n. 164 75: 205n. 178 76: 69 77: 206n. 180 78– 79: 69 78.1: 42 78.1– 3: 177 79.2: 69 82: 69, 155n. 59 90– 91: 115n. 211 92.2: 70 93– 94: 70– 73 95: 70 98.4: 51n. 27 100– 101: 194– 95 100.2: 51n. 25, 155n. 59 105: 205n. 178 106.2– 4: 204 107: 101– 2 108– 13: 65n. 67 109.2: 42n. 64 109.3: 152n. 47 114.2: 58 116.3: 85 117: 58 119– 20: 58n. 46 122.1– 2: 88n. 131 122.2: 153n. 48 Hesiod MW frags 37: 60n. 53 129: 60n. 53 130: 60n. 53 131: 60n. 53 150: 116n. 213 Theogony 27: 254n. 76 Works and Days, 242– 44: 72 378: 142n. 22 Hippocratic Corpus Airs, Waters, Places 13, 15, 16: 87 17– 22: 109n. 188, 141n. 17 17: 129n. 253 On Joints 53: 125n. 243 Homer

Iliad 1.8: 29n. 22 1.146: 119n. 223 2.235: 102n. 173 2.308– 32: 46 2.552: 218n. 211 5.62– 63: 204n. 173 5.703: 29n. 22 6.152– 206: 124n. 237 7.125: 218n. 211 8.163: 102n. 173 11.86: 176n. 114 11.113– 21: 246n. 49 11.170– 78: 246n. 49 11.473– 86: 246n. 51 17.274– 87: 176n. 114 18.503– 4: 72 18.573– 81: 72 22.124– 25: 102n. 173 24.53: 185n. 130 Odyssey 1.325– 52: 4n. 10 6.119– 20: 74n. 84 8.266– 369: 4n. 10 8.471– 541: 4n. 10 9.105– 15: 74n. 84 11.281– 97: 60n. 53 11.333– 476: 4n. 10 12.127– 33: 71 12.439: 176n. 114 15.225– 46: 60n. 53 19.203: 254n. 76 Homeric Hymns Ap. 3.18: 238n. 22 Ap. 3.25– 138: 238n. 21 Ap. 3.48: 239n. 24 Ap. 3.55: 239n. 24 Ap. 3.60: 239n. 24 Ap. 3.72: 239n. 24 Ap. 3.113: 238n. 22 Ap. 388– 544: 253 Dem. 3.3: 239n. 25 Dion. 1.20: 164n. 79 Dion. 7: 253 Isocrates Panegyricus 4.68– 70: 125n. 241, 132n. 259

Index of Passages Lucian Herodotus 1– 2: 15n. 38 Lysias Epitaphios 2.4– 26: 125n. 241 2.4: 132n. 259 2.6 : 123n. 236 2.45: 174n. 107

Marcellinus Life of Thucydides 54: 15n. 38 Myron of Priene FGrHist 106 F 2: 113n. 199

Panyassis frag. 26K: 142n. 22 Pausanias 1.15: 124n. 239, 202n. 169 1.17: 124n. 239 6.2.5: 61n. 57 6.4.5: 62n. 58 6.17.5– 6: 62n. 58 Pherecydes FGrHist 3 F 15: 124n. 237 17: 142n. 22 F 33: 60n. 53 F 114: 60n. 53 F 115: 60n. 53 F 151: 124n. 237, 128n. 252 F 152: 124n. 237, 128n. 252 174: 116n. 213 Phocylides Frag. 4 Diehl: 240n. 32 Pindar Isthm. 6.23: 5n. 91 Nem. 3.36– 39: 124n. 237 Ol. 1.55– 57: 200 6.35– 72: 61n. 57 6: 62n. 58 13.10: 200n. 163 13.63– 92: 124n. 237 Paean 4.24– 26: 60n. 53 SM frags. 33d: 239n. 26 169.1– 4: 170 172: 124n. 237

323

175: 128n. 252 176: 124n. 237 Plato Apology 35b2– 3: 102n. 173 De Exilio 13: 271n. 16 Gorgias 482E– 84D: 171 483E: 171n. 100 491E: 171n. 100 492B: 171n. 100 Laches 190D– E: 114n. 205 191A– C: 114n. 205 Republic 392d– 94c: 17n. 44 454d: 252n. 70, 255n. 77 463c– 65b: 79n. 103 Theaetetus 155d: 234n. 8 Plutarch Lycurgus 12: 114n. 207 22: 108n. 187, 114n. 208 28: 113n. 199 On the Maliciousness of Herodotus, Mor.854E– 74C: 172 857A– B: 143n. 26 857A– F: 7 857C: 155n. 59 857F– 858A: 184 862A– B: 15n. 38, 271n. 13 862E: 261n. 95 862F– 63A: 263n. 101 863A– B: 262n. 98 863A: 224n. 226 863E: 228 864A– B: 175n. 112 864C: 255n. 77 864D: 271n. 13 867B: 263n. 101 868A: 172n. 16 870B– 71B: 224n. 226 871C: 255n. 77 871D: 7 Pericles 3: 58n. 48 6.2: 269n. 6 7.1– 2: 58n. 48 12: 58n. 48 16: 58n. 48 28: 203n. 170 28.7: 202n. 169

324

Index of Passages

Plutarch Pericles (continued) 31.4: 124n. 239 37.5: 264n. 103 Theseus 26– 28: 124n. 237 26: 128n. 252 Polybius 6: 123n. 235 Protagoras B4: 163n. 76 DK 80 A2: 166n. 87 Simonides frag. 11 W2, 42: 61n. 57 Solon frag. 5 Diehl3, 9: 200n. 163 Sophocles Ajax: 65n. 70 189n. 140 Oedipus Tyrannus: 198n. 158 872: 200n. 164 Trachiniae 1071– 72: 102n. 173 Strabo 6.1.13– 15: 267n. 2 6.1.13: 269n. 8 6.1.15: 270n. 11 7.3.9: 80n. 106, 108, 141n. 17 11.5.1: 128 11.5.4: 123n. 236 11.6.2: 80n. 106 11.6.3: 234n. 9 11.11.3: 161n. 71 15.3.17: 154n. 53 Suidas  s.vv. Ηροδοτος and Πανυασσις 267n. 1 Tacitus Germania 22: 154n. 56 Theognis 153– 54: 200n. 163 425– 28: 101n. 172 713: 254n. 76 825– 30: 114n. 208

Thucydides 1.1– 19: 44n. 69 1.22.1: 174n. 108 1.22.4: 99n. 169 1.23.1– 3: 46 1.23.6: 212n. 194, 213n. 200 1.24– 55: 221 1.69.5: 174n. 107, 203n. 170, 258n. 89 1.73.1: 220 1.73.2: 220 1.73.4– 74.4: 74n. 107 1.75: 208n. 184, 212n. 194 1.75.1: 216n. 206 1.76.4: 219n. 214 1.77.6: 216n. 206 1.79.2: 116n. 212 1.80.1: 213n. 200 1.82.1: 229 1.84.2: 116n. 212 1.84.3: 116n. 212 1.86.1: 174n. 109, 215, 220 1.86.3: 146n. 35 1.94: 68– 69 1.98.3: 205n. 178 1.102.4: 229n. 237 1.108.4: 190n. 143 1.118.2: 213 1.118.3: 198n. 158 1.122.3: 58n. 49 1.124.3: 58n. 49 1.131– 34: 68– 69 1.140.2: 146n. 35 1.143.5: 116n. 215 1.146: 146n. 35 2.7.1: 213n. 200, 229n. 237 2.8.2– 3: 198n. 158 2.8.3: 201 2.8.5: 174n. 108 2.17.2: 198n. 158 2.27.1: 190n. 142 2.28.3: 194n. 150 2.42: 189n. 140 2.52– 68: 221n. 218

Index of Passages 2.52– 54: 198n. 158 2.63.2: 58n. 48 2.64.5: 174n. 108 2.65.9: 58n. 48 2.67: 193, 229n. 238 2.70.3: 219n. 214 2.71– 78: 221n. 218 3.20– 24: 221n. 118 3.37.2: 58n. 48 3.50: 219n. 214 3.62.3– 4: 53n. 32 3.82.3: 99n. 169 3.96.1: 198n. 158 4.27.3: 102n. 173 4.36.3: 175n. 112 4.40.1: 175n. 112 4.40.2: 115n. 211 4.50: 229n. 238 5.14.4: 229n. 237 5.26.3: 198n. 158 5.28.1: 229n. 237 6.8– 26: 210 6.85.1: 58n. 49 7.28.3: 194n. 150 7.50: 194n. 150 7.51.1: 194n. 150 7.87: 44n. 69

325

Tibullus 4.1.144– 45: 162n. 72 Tyrtaeus frags. 12.13– 20, 11.31– 34 W: 114n. 205 Tzetzes Chiliades 7.629– 36: 83n. 117 Xenophanes DK 21 A14: 149n. 42 DK 21 A34: 166n. 88 Xenophon Anabasis 5.4.34: 151n. 45 Lakedaimonion Politeia 1.3: 104n. 177 5: 114n. 208 9.4: 114 10.7: 112n. 197 11.3: 102n. 173 13.1: 112n. 197 Oeconomicus 7.23: 130 [Xenophon] Athenaion Politeia 1.6.– 9: 208n. 184 2.2.1: 191n. 146 2.7: 2.14: 191n. 146, 156n. 61 2.20: 208n. 184

Chapter 2

Comparison

Comparison is an interpretive operation that “puts together” two facts for the purpose of explaining one on the basis of its similarity to or difference from the other.1 That which sparks the comparison is an element of the story. The second term can be also drawn from within the narrative, or it may come from outside of it, such as a past or present fact that belongs to the “real world” of the narration or to another “text” familiar to the narrator and his audience.2 One of the peculiarities of the Histories is that the boundaries of the narrative are especially fuzzy. By virtue of the contract that Herodotus establishes with his audience, everything is at least potentially part of the story he has to tell. In the actual telling, a fact that is brought in incidentally for the sake of comparison or some other reason may become the object of a narrative within the logos in a way that is hard to predict. Nevertheless, Herodotus’ references to events after 479, for example, coupled with his evident reluctance to include such references, demonstrate the existence of boundaries as well as their provisional nature.3 We expect a continuity between the logos and the “real world” of the narration in Herodotus that we do not expect from Thucydides, but at the same time, we acknowledge an inside and an outside and, between the two, a necessary break. Comparison may be implicit or explicit. It is explicit when the narrator directs the recipient of the narrative to consider a fact of the narrative in reference to some other fact by means of a gloss of comparison indicating similarity, analogy, or difference. It is implicit when the recipient of the narrative perceives on his or her own that a fact of the narrative wants to 1. On συµβαλλειν (put together, compare), see “The Texture of the Earth” later in the present chapter. 2. On the relation between a story and its extratextual context or subtext, see, e.g., Bal 1985, 81. 3. Cf. my introduction.

45

46

Telling Wonders

be considered in light of another and that a conceptual relation of analogy or (theoretically) difference links the two. In and by itself, difference establishes no relation at all. Even employing apples and oranges to denote items that cannot be “put together” at all is somewhat misleading. When two items are mutually related through difference, this can only be because they are similar in other respects. Resemblance lurks in the background every time comparison is an issue. If resemblance overcomes fundamental differences and makes them appear circumstantial, it constitutes analogy.4 Comparison and analogy may be activated “horizontally” to bind overlapping, concentric, or parallel classes of similar objects. But they also work “vertically,” through indices and symbols across different levels of reality, as in inductive prophecy. In the Iliad, the nine sparrows devoured by a red serpent somehow resemble and therefore represent the nine years of the Trojan War.5 In Thucydides, extraordinary natural cataclysms, though not ominous, are nevertheless analogous to and symbolic of the upheavals that the Peloponnesian War has produced (Thuc. 1.23.1– 3). In Herodotus, the world of animals mirrors the human world, while concrete actions and objects are indices for something different or more intangible. Whereas horizontal analogy is based on the notion that phenomena recur with variations, vertical analogy brings out the similarity of situations on different planes, so that one becomes a sign for the other. Whether explicit or implicit (and often simultaneously activated along the horizontal and the vertical planes), comparison and analogy are fundamental strategies by which the text of the Histories organizes its material. Because the logos contains so many story elements that escape the network of causal connections of the plot, classification and the comparative approach that classification entails provide a powerful glue; this in turn also acts on causally connected facts. In the historical narrative, it is most frequently facts belonging to different points of the chronological continuum that are compared to one another (diachronic comparison). The comparative field created by descriptions in the present tense extends not in time but in space (synchronic comparison). 4. Analogy is defined by Lloyd (1966, 175) as “any mode of reasoning in which an object or complex of objects is likened or assimilated to another.” 5. Il. 2.308– 32, especially 326– 29, cited by Corcella (1984, 33– 34). See also Lloyd 1966, 180– 85. The language of many oracles reported in the Histories also relies on symbolic analogy.

Comparison

47

Diachronic and synchronic comparison are largely distinct operations, partly because the material that is being compared in history and ethnography, respectively, belongs to different classes (“apples” and “oranges” in the proverbial sense). It just so happens, moreover, that in the two types of discourse, comparison mobilizes the narrator’s presence to a widely different degree. To put it simply, qualitative diachronic comparison largely imposes itself on the consciousness of the listener through the implicit similarity of the events narrated; synchronic comparison, by contrast, gains momentum by frequent explicit and far-ranging glosses that advertise the notion of qualitative similarity or difference by saying “X is like Y” or “X is different from Y.” What is the meaning of this discrepancy? To what extent are diachronic comparison and synchronic comparison mutually related and complementary strategies? I will begin dealing with these questions by considering comparison in the history. Comparison in Time Analogy as an Interpretive Tool Before I examine the forms and the contexts of glosses, I will briefly recall the role of implicit comparison. Like all the εργα µεγαλα τε κα ι θωµαστα [great and wonderful achievements] promised in the programmatic first sentence, each past event of Herodotus’ narrative is strictly speaking a particular and unique occurrence. In general, however, the exceptional historical event that deserves a place in the Histories is different mainly for its magnitude rather than for its quality.6 Superlative evaluations, for example, often serve to advertise narratability (the axion logou) by stating that certain story elements possess the greatest amount of a certain quality among others of the same class.7 Since they underline what is exceptional in a quantitative sense, these markers also indirectly identify classes of qualitatively similar phenomena. Thus Pisistratus’ “most simple-minded device” (1.60.3) and Hermotimus’ “greatest revenge” (8.105.1) make reference to other acts of deception and retribution across the broad range of 6. Thoma, “wonder,” the strongest marker for a qualitatively unique phenomenon, is  discussed in chap. 4. In past-tense narrative, µουνος more frequently singles out people who behaved differently from the others in the same situation (see, e.g., 1.147.2, 3.55.1, 7.107.1) rather than “the only time X ever happened” or “the only one who ever did X” in an absolute sense (see, e.g., 1.25.2, 9.35.1). 7. See Bloomer 1993.

48

Telling Wonders

the work and outside of it. Through classification, analogy wins the contest with difference. The proem of the Histories implies that several occurrences are similar enough to be grouped under the generalized headings “small city becomes great” and “large city becomes small” (1.5.4). Similarities among different actions, their motives, and their outcomes may emerge from the recurrence of words and concepts within different contexts.8 Occasionally, speakers compare and contrast. When they discuss circumstances of their present in light of events of their past, they encourage the audience to perform the same sort of analogical operation and apply to their own present the same or other parts of Herodotus’ logos.9 Helped by these clues and by the exceptional diachronic range of the Histories, we register the uniqueness of the events narrated, but at the same time, we also overcome it; we regroup facts in different ways on the basis of their mutual resemblance. In the theoretically endless variety of particulars, “X is like Y” over and over again; together these similar facts insistently recall elements of what lies outside of the logos and belongs to the real world of narrator and audience. Herodotus’ original listeners had far more practice than the modern reader for receiving the work in this manner.10 Assiduously studied, analogy in the Histories has provided an indispensable tool for interpreting the work.11 The repetition of similar events results in a number of patterns or—at the level of discourse—cultural codes that recur throughout the work.12 Some of the most conspicuous of these patterns have been identified, described, and named. They now represent canonical terms for speaking about the Histories: the crossing of geographical boundaries for the purpose of conquest;13 the “rise and 8. Cf., e.g., the first mentions of Deioces (1.96.1) and Themistocles (7.143.1): both are individuals rising in power. See Wood 1972, 17. 9. On the comparison between Sparta and Athens partially focalized through Croesus at 1.56– 69, see Gray 1997. Artabanus compares Xerxes’ prospective expedition against Greece to the earlier expeditions at 7.10α– γ. 10. See my introduction, n. 10 and corresponding text. For a parallel between tragedy and the work of Herodotus from the point of view of the audience’s receptiveness to analogy, see Raaflaub 1987, 231– 32. On the didactic and political dimension of tragedy, see, e.g., all the essays in Goff 1995; Sa¨ıd 1998. 11. See Immerwahr 1966; Wood 1972 (especially 17– 20); Lateiner 1984 and 1989; Corcella 1984. Cf. Waters 1971, but see Lateiner 1989, 165– 67. 12. To the extent that a code is the language the text uses to speak about something, patterns correspond to codes, though the notion of codes is broader. See my introduction, n. 25. 13. See my introduction, n. 26 and corresponding text.

Comparison

49

fall of the ruler”;14 the expedition of a superpower against a tough and poor nation, the so-called primitive opponent;15 the wise adviser or “tragic warner” mostly unheeded by the recipient of the advice, who rushes to his ruin;16 the pattern of imperialism;17 the exile who seeks refuge at the king’s court.18 An important recent addition is the kinginquirer, a figure of metahistorical significance who by analogy or opposition illuminates the purposes and methods of the histor of the Histories and his counterpart outside the text, Herodotus himself.19 Among the various concentric or overlapping patterns in the logos, the monarchical model is especially pervasive, since it tends to subsume many others to itself. This model is constituted by all the specific actions and features that serve as indices of actual or potential autocracy—of an individual’s attempt or ability to rise above a community, his own or someone else’s, and impose his will on it while living himself by different rules. Such actions and features are attributed especially to those who hold royal power or aspire to do so in a literal sense (though rarely, if ever, does a single character exhibit the full repertoire).20 They may be historically consequential (e.g., political manipulations; subversions of the social order; punishments; conquests; victories and defeats in war), or they may be mostly symbolic or connotative of the abnormal position of an individual within the state (e.g., crossings, mutilations, exceptional marriages, gift giving, athletic victories, trickster actions).21 An exceptional feature confirms the centrality of monarchy/tyranny as a predominantly negative paradigm in the Histories: one of the characters describes the phenomenon, and the components of his theoretical descriptions are 14. See Immerwahr 1966, especially 149– 98. 15. See Hellman 1934, 77– 98; developed by Cobet 1971, especially 172– 76. See also Flory 1987, 81– 118. 16. See Bischoff 1932; Lattimore 1939; Dewald 1985. Consider also the pattern of unheeded or misunderstood dreams and oracles (see Corcella 1984, 160). 17. See Evans 1991, 86– 87. 18. See n. 74 and corresponding text later in this chapter. 19. See Christ 1994. 20. See Immerwahr 1966; Lateiner 1984; Lateiner 1989, 163– 86; Corcella 1984, 163– 77; Hartog 1988, 331– 34; Gammie 1986; Dewald forthcoming b. For the pattern of monarchy as an archaic and classical Greek cultural stereotype, see McGlew 1993, 24– 35. 21. Thus, my monarchical model differs from what Dewald (forthcoming b) calls “the despotic template,” which she defines as “Herodotus’ description of the evils of autocratic rule.” The code of kingship is the broader discourse on kings that shapes and delimits the monarchical model. The subtlety of such discourse has led Flory (1987, 119– 49) to detect (wrongly, I think) advocacy in favor of a one-man rule in the text. Gray (1997) rightly emphasizes the variety of internal patterns.

50

Telling Wonders

verified by the actions of specific historical rulers throughout the narrative.22 The problematic of monarchy represents in fact much of what Herodotus’ work is about. It provides terms to be taken symbolically as  η),  imperialism, well as literally for presenting the issues of rule (αρχ transgression, oppression, and the competitive self-aggrandizing drives of both individuals and states.23 The narrative explores monarchy as a historical phenomenon and as a plausible prospect for the future of Greece in the literal sense (we tend to forget, for example, the extent to which Herodotus, with his long-range view of patterns, unwittingly predicted fourth-century monarchical outcomes). This is an especially conspicuous case, however, where the analogical system works not only horizontally but also vertically, that is, metaphorically, by assimilating to monarchy other manifestations of power, leadership, and alienation from the commonwealth.24 The bilateral equivalence between cities and kings, like the translation of rivers into moral boundaries, also largely depends on the symbolic and semic work of vertical analogy. The implicit interaction between horizontal and vertical analogy conveys the sameness of apparently different kinds of experience through a covert mode of communication akin to that of the ainos. Explicit Analogy While resemblance plays such a large role in the historical narrative, much rarer are metanarrative glosses of comparison that place two specific facts side by side and expressly say that one is somewhat like the other or, for that matter, different from the other. Yet there are other metanarrative clues. Additional allos (other) and adverbial kai (also) serve as markers of horizontal assimilation. The qualification of events in terms of their position in a series of similar items—for example, first, second, third conquests of Ionia (1.92.1, 1.169.2, 6.32)—brings out the continuity of a historical process potentially down to the time of narration (what about, for example, a fourth conquest of Ionia?). Coincidences 22. The character is Otanes in the Constitutional Debate (3.80). See especially Lateiner 1989, 167– 81. The other generalized descriptions in the debate are not equally related to the appearance of patterns in the narrative. 23. The topic of empire in the Histories has been discussed by Stadter (1992). 24. See discussion of horizontal and vertical analogy earlier in this chapter. The participation of the monarchical model in the symbolic code of the Histories is indicated, e.g., by Croesus’ words to Cyrus that Solon spoke for all men and especially for those who deem themselves happy (1.86.5).

Comparison

51

among mutually autonomous occurrences point to a unitarian historical movement and mysterious interconnections.25 Two items may follow one another in close narrative sequence on the basis of their similarity. Thus, the narrator consecutively mentions two queens of Babylon (the second “more clever than the one who reigned before,” in an example of quantitative difference highlighting likeness); another time, he recalls two Persian governors in Thrace, Mascames and Boges, both distinguished for bravery.26 A queen pattern, in the first instance, and a Persian cultural  definition of αρετ η (valor), in the second, emerge from the juxtaposition of similar historical cases. Occasionally, through a gloss of analogy, the discourse tampers with the story by bringing in a more or less extraneous element, as when the narrator compares different stages of a war.27 The comparison may attach a new small narrative to one that has just ended: thus, the report of Xerxes’ decision to send the Greek spies home unharmed (7.146.2– 7.147.1) attracts the announcement of the narrative of another, similar gnome about his releasing enemy ships captured at the Pontus (οικε . . . 28 By this compositional principle, the narrative accumulates  τ ηδε αλλη). evidence of a certain type and thereby trains the listener to generalize and interpret. A more drastic manipulation of the story occurs when the second term of comparison belongs to a different historical context than the first. The name of this woman who ruled [in Egypt] was the same as that of the Babylonian [queen] [το περ τ η Βαβυλωνι  η], Nitocris. (2.100.2) Trivial as it may seem, this comparative back reference provides a significant reading direction, one the text has already suggested by the simple juxtaposition of the two Babylonian queens in the earlier narrative 25. E.g., Cambyses is wounded in the same part of his body where he has wounded Apis (3.64.3); the battles of Himera and Salamis occur on the same day (7.166), and so do those of Plataea and Mycale (9.100.2). 26. 1.184– 85.1, 7.106– 7. In a continuous dynastic line, a king is frequently compared  . . . δε from eunomie to kakotes), to his predecessor. See, e.g., 2.124.1 (a reversal with µεν 2.127.1 (similarity), 2.134.1, 1.103.1, and the more elaborate 2.110.1– 3. 27. See 9.98.4: the strategy of Leotychides at Mycale was the same as that of Themistocles at Artemisium (see Immerwahr 1966, 256). The battle of Salamis is constantly compared to the battle of Artemisium: see 8.42 (same commander and larger fleet), 8.66.1 (number of enemy not inferior). See Immerwahr 1966, 268. Cf. 1.191.3: to capture Babylon, Cyrus did that which (τα περ) the Babylonian queen had done; cf. 3.152. 28. Cf. 7.144.1, 7.57, 7.114.2, 4.78.1.

52

Telling Wonders

(1.184– 185.1). Here Herodotus explicitly adds the Egyptian Nitocris to the pattern, as if the homonymy were the sign of other analogies—two women, both Eastern, both rulers, both shrewd, and so on. Despite certain striking differences between them, the narrator establishes the continuity of a type across time and space.29 Within the small group of glosses comparing contextually distant items, two are of particular interest. On the formal level, both transform analogy into historical action by assuming mimesis. In other words, the narrator says not that “X is like Y,” but rather that “X imitates Y.” This formulation agrees with his general reluctance elsewhere to exercise authority over diachronic analogy. These glosses almost suggest that the responsibility for bringing two disparate actions together lies not with the collector of the logoi but with the historical agents who chose to play someone else’s role.30 More importantly, both glosses belong to the same broad analogical field whose overarching prominence in the logos I have already mentioned—the monarchical model. Though they do not in themselves fulfill an indispensable function, they confirm the symbolic aspect of this analogical network in the Histories. Both glosses of comparison and the narratives they respectively introduce serve to justify the analogy between the citizen of a Greek polis and a monarchical ruler. The Monarchical Model in Athens The first gloss leads to the abrupt narrative juxtaposition of Cleisthenes, the tyrant of Sicyon, to the Athenian democratic reformer of the same name. Cleisthenes of Athens enters the logos because he belongs to the story entitled “Athenian Ordeals and Achievements after the Fall of Tyranny” (see the introductions at 5.65.5– 5.66.1). The section of this narrative where Cleisthenes plays an active role is deceptively dry. After prevailing over his political rival Isagoras and obtaining the support of the people, he replaced the four old Ionian tribes with ten new ones; he named these after ten local eponymous heroes and discarded the old denominations (5.66). 29. The queen pattern is rich in symbolic associations. For related studies, see Dewald 1981; Tourraix 1976; Flory 1987, 23– 48; Munson 1988; Dillery 1992.  30. In Herodotus, the verb µιµεοµαι refers to animate imitators in the sense of “doing as someone else does or did” and does not necessarily imply awareness or intention to imitate (see, e.g., the analogy drawn by Cambyses’ wife at 3.32.4), but in practice it almost always refers to a derivative imitation (see, e.g., 2.104.4 in an ethnographic context and 4.166.1 in a historical one).

Comparison

53

This account of the foundation of the new Athenian democracy, however, bristles with negative signs.31 It identifies Cleisthenes and Isagoras, respectively, by referring to the former’s bribery of the Delphic oracle and to the latter’s foreign origins (5.66.1). It denotes their political prominence with a verb (ε δυναστευον, “held sway”) that is appropriate to narrow oligarchic circles. Herodotus elsewhere applies it to aspiring tyrants and corrupt potentates.32 The two men “engage in stasis for the sake of power” [ε στασι ασαν περ ι δυναµιος],33 and Cleisthenes coopts  δηµον  the demos at large to his hetaireia (τον προσεταιρι ζεται). Because, in the time of Herodotus, at least, Athenian hetaireiai were exclusive aristocratic clubs formed with the aim of forwarding the political advancement of their members, the term here suggests that Cleisthenes’ democratic stance was a self-serving maneuver in the context of an aristocratic power struggle.34 A verbal parallelism confirms Cleisthenes’ resemblance to Isagoras, who aimed at establishing himself tyrant in Athens:  first Cleisthenes, when he is “losing ground” (ε σσουµενος, 5.66.2; cf.  5.69.2), and then Isagoras, when he is “losing ground” (ε σσουµενος,  5.70.1), come up with a special device (see ε πιτεχναται at 5.70.1). Just as the one elicits the friendship of the Athenian people, the other appeals to the Spartan king, his guest-friend. Once Cleisthenes, banished by Isagoras and the Spartans, fades from the story and from the narrative, the next section reports the resistance of the Athenians to the Spartan intervention in favorable terms (5.72.1– 2), and a conspicuous interpretive gloss praises the resulting democratic order (5.78). The Cleisthenes passage, however, emphasizes the reformer’s bid for personal power while representing the reform itself in such reductive terms that its democratic implications either in a practical or in an ideological sense appear unintelligible: “He gave the Athenians ten tribes, whereas they used to have four, discarding the names of the children of 31. See Strasburger 1955, 15; Bornitz 1968, 49– 50; Fornara 1971, 55. Cf. Myres 1953, 180– 82. 32. On dunasteia, see Ostwald 1969, 113, 116– 17, citing Thuc. 3.62.3– 4. Dunasteuein appears in a negative context in the Histories at 6.39.2, 6.66.2, and 9.2.3. The verb is used of Athens with similar effect (5.97.1). 33. 5.66.2 Stasis (civil struggle) is another oligarchic term (see 3.82.3) and a key negative concept in Herodotus: cf. 8.3.1, quoted and discussed in chap. 3, “The Evils of War.” 34. Georges (1994, 160) dubs this expression a “conscious paradox.” Cf. the less derogatory formulation of Arist. Ath Pol. 22.1. Herodotus’ term is anachronistic for the beginning of the fifth century. See Ostwald 1969, 142– 43. Ober (1993, 227) renders the tense and the middle form with “Kleisthenes embarked in the process of becoming the demos’ trusted comrade”; but the reference to hetairiai cannot possibly have benign connotations.

54

Telling Wonders

Ion, Geleon, Aegicores, Argades, and Hoples, and coming up with  the names of another group of heroes, all native of the land [ε ξευρων] except for Ajax” (5.66.2). The verb ε ξευρι σκειν here connotes a contrivance, something new and invented for the sake of expediency but at the same time ostensibly recovered from the traditional past.35 After the narrative has done its work through connotation, the coup de grace ˆ arrives in the form of a gloss of comparison that introduces the account of a reform by the tyrant of Sicyon by stating that it provided the model for that of the Athenian Cleisthenes. In this [i.e., the replacement of the old tribes in Athens], this   Cleisthenes it seems to me [δοκεειν ε µοι ] was imitating [ε µιµεετο] his maternal grandfather Cleisthenes the tyrant of Sicyon. (5.67.1)  The self-referential δοκεειν ε µοι identifies both Cleisthenes’ mimesis and the resulting similarity as interpretations of the narrator. The family relation and the homonymy (which we will never be allowed to forget; cf. 6.131.1) serve as corroborating evidence for the one and the other, respectively. Herodotus often attaches one story to another incidentally, by exploiting an intrinsic story connection; he adds the pretext of the axion logou, narratability, as he lets the thematic correspondences take care of themselves.36 Here, however, he openly says that the tyrant of Sicyon is brought into the logos not because the narrative about Cleisthenes of Athens provides an opportunity for talking about his famous grandfather but because of the tyrant’s thematic relevance to the present context of the democratic reform. The juxtaposition would have been enough to trigger the analogy, and the narrator did not have to be and normally is not as explicit as he chooses to be in this case. The analeptic narrative introduced by the gloss of comparison compounds the damage for the Athenian Cleisthenes, whose policy it purports to explain by analogy and derivation. The section on the reorganization of the tribes in Sicyon, which is the essential point of the similarity between the Athenian democratic reformer and his grandfather, is pre ι σκειν are often used positively in reference to a 35. The verbs ε ξευρι σκειν and ευρ people’s inventions (see 4.46.2, 8.98.1, 2.92.1) but are also applied to constructs (see, e.g., 2.23, 4.79.3). See especially Cambyses’ far-fetched “discovery” of a nomos overriding another Persian law that he does not want to obey (3.31.4). 36. For the ways in which narratives are brought into the logos, see especially Jacoby 1913, 344– 50, 383– 92; Pagel 1927, 41– 61; Erbse 1961, 243– 57; Munson 1986.

Comparison

55

ceded by another account. This relates how the older Cleisthenes outlawed Homeric recitations (the Homeric poems being full of the glory of the Argives) and managed to expel the Homeric hero Adrastus, who happened to be buried in Sicyon. Consulted on the matter, the Pythia forbade it and declared that whereas Adrastus had been the king of the Sicyonians, Cleisthenes was their lapidator (λευστηρα, 5.67.1– 2). Reference to the typical despotic action of “killing men without trial” is thus, through the voice of the oracle, inserted into an account of violations of a religious nomos.37 To circumvent the Delphic interdiction, Cleisthenes “invented” a device for making Adrastus leave on his own. The verb ε ξευρι σκω (5.67.2) connotes once again subversive manipulation of tradition, as in the case of the new heroic names of the younger Cleisthenes’ Athenian tribes. The two actions are in fact parallel, since the Athenian Cleisthenes, according to Herodotus, abolished the four heroes as namesakes of the city tribes, while his grandfather found a way to get rid of one hero as the recipient of cult in the city. The “device” of Cleisthenes   the Elder (µηχανην; cf. µηχανωται applied to the religious ruse of Pisistratus at 1.60.3) consisted in importing to Sicyon from Thebes the hero Melanippus, who had been Adrastus’ worst enemy, and in transferring Adrastus’ solemn festivals, sacrifices to him, and his choruses to the god Dionysus (5.67.2– 5). The survey of these blatant violations of the city’s time-honored traditions (cf. 5.67.4– 5) is followed by the account of how the tyrant changed the names of the Dorian tribes in Sicyon “in order that Sicyonians and  Argives would not have the same ones.” He mocked (κατεγελασε, a term crucially related to violations of customs at 3.38.1– 2) the Dorians of Sicyon, by choosing for them new denominations derived from the words for “pig,” “swine,” and “ass.” For his own (non-Dorian) tribe, however, he reserved a dignified name, “Archelaioi,” referring to his own arche, “rule” (5.68.1). This reform outlived Cleisthenes only by sixty years (5.68.2). After this pointed reference to the tyrant’s fall, the narrator reemerges to bring us back to the younger Cleisthenes in Athens with a resumptive gloss that reiterates almost obsessively analogy and its supportive story elements—imitation, kinship, and homonymy.  see Elayi 1979, 224– 27; Ogden 1993. McGlew 37. See 3.80.5. On the term λευστηρ, (1993, 16) notices that Cleisthenes’ expulsion of Adrastus constitutes a reversal of the expulsion of the deceased tyrant’s bones extra fines. In Herodotus’ narrative, it also constitutes a reversal of the expulsion of the Alcmaeonids (5.72.1: same verb ε κβαλλειν) as a result of their ancestral curse. On the political dimension of hero cults, see Boedeker 1993.

56

Telling Wonders

This is what Cleisthenes of Sicyon had done [retrospective conclusion]; and, indeed, with regard to the Athenian Cleisthenes, who was the son of the daughter of this one from Sicyon and was named  after him, it seems to me [δοκεειν ε µοι ] that he also despising the Ionians and so that the Athenians would not have the same tribes,  imitated [ε µιµησατο] his namesake Cleisthenes. (5.69.1) This second gloss of comparison further enhances the relevance of the preceding narrative about Cleisthenes of Sicyon to the younger Cleisthenes. The tyrant, for example, made sure to reserve a special place in the city for his own tribe (5.68.1): this casts doubt on the impartiality of the tribal reform of the democratic leader.38 Herodotus’ attribution of antiIonian bias to the Athenian Cleisthenes (parallel to the anti-Dorian intent of Cleisthenes of Sicyon) makes reference to the hostility among different groups of Greeks and to later hegemonic struggles.39 I have stopped to discuss this explicit comparison because the analogy it brings into the open affects Herodotus’ overall representation of modern democratic Athens—born at this historical point after the fall of tyranny and risen to greatness through the Persian Wars and beyond. Tyranny, as it has been often repeated, is viewed by fifth-century Greek thought as antithetical to Greece.40 In Herodotus, two elements favor the assimilation between Greek sixth-century tyrants and Eastern monarchs: a deliberate confusion in his use of the terms turannos, basileus, and mounarchos;41 and his attribution of similar features to kings and tyrants, despite all the variety of his representations in individual cases. The analogy between the two Cleisthenes, however, undermines the notion that the “tyrannical” stage in the political development of the Greeks has come to an end. As the creator of Athenian democracy is moved closer to the opposite tyrannical/barbarian camp, the monarchical model begins vertically to affect the new Athens as well, firstly through its leadership and secondly through the demos, both of them “tyrannical” in a broad metaphorical sense. After representing Cleisthenes’ bid for primacy as analogous to the 38. Some modern historians have tried to explain the apparent territorial oddities of Cleisthenes’ reorganization by speculating that the Alcmaeonids may have received special treatment. See Fornara and Samons 1991, 39– 56, for a discussion of the problem. 39. Here Herodotus’ analogy may not be as misleading as Macan argues (1895, 1:211). See Hall 1997, 53– 54. 40. See, e.g., Georges 1994, 37– 46. 41. See Ferrill 1978; Dewald forthcoming b.

Comparison

57

factionalism of the time of Pisistratus, the narrative partially recruits to the monarchical model subsequent Athenian democrats who seek or exercise power with high-handed methods, albeit in a constitutional setting. Secrecy and self-interest mark the essential alienation of each from the commonwealth of citizens. At the same time, however, the demos, whom Cleisthenes has coopted to his ambitions, has inherited power in internal and external affairs, including, at the time of narration, his hostile policy toward the Ionians.42 With democracy, king and city have to some extent become one, because while separate from the city and a threat to its institutional integrity, each “tyrannical” or potentially tyrannical leader at the same time often expresses the will of the demos. The city is both the victim of an individual’s schemes and his willing accomplice—both subject and ruler, in a metaphorical sense. This contradiction emerges from the cases of Miltiades and Themistocles. Miltiades plans an imperialistic venture against Paros on his own by keeping the Athenians in the dark as to its destination and deceiving them about its goals. The sovereign democratic assembly, however, decides on the expedition (6.132). When the expedition fails, they assert their authority over the leader by bringing him to trial. Miltiades’ conviction by the city and a death that appears as the result of divine punishment coincide to mark the “tyrant’s” downfall.43 Themistocles, like the Pisistratids, establishes a personal connection with Xerxes (8.109.5) and has tyrannical features of his own.44 But he also embodies the enterprising and opportunistic tendencies of the city as a whole and in most cases acts as its executive.45 The city’s own monarchical image catches up with the representation of the individual leader seeking power for himself in the state. At the end of the Histories, individuals are but pale reflections of the policy of the city. It is “the Athenians” who 42. The Athenian prejudice against the Ionians is referred to the present of narration by  See Fornara and Samons 1991, 106– 9, on the the gloss of testimony at 1.143.3 (κα ι νυν). Athenian contempt of the Greeks of Asia as a psychological justification for the aggressive policy of Athens against its allies after the Persian Wars. 43. In his expedition against Paros, which results in impiety and death, Miltiades imitates in reduced format the career of the arch despot Cambyses (see 6.132– 36; cf. 3.37, 3.16, 3.27– 29, 3.64.3, 3.66.2). See Immerwahr 1966, 191– 92; Corcella 1984, 137; Flory 1987, 115. 44. Themistocles imposes his own strategy on the Greeks at Salamis (see 8.62.2– 63, 75– 76) and compulsion on other Greeks after the battle (see 8.111– 12). See Munson 1988, 101. Themistocles’ self-interest and deception (see, e.g., 8.4– 5, 8.109– 10.1) are characteristic of the rising tyrant: see “The Monarchical Model in Sparta” later in the present chapter. 45. For the affinity between Themistocles and Athens, see Immerwahr 1966, 223– 25; Wood 1972, 185– 86; Raaflaub 1987, 227; Munson 1988, 100.

58

Telling Wonders

begin the offensive stage of the war against Persia by crossing over to the Chersonese (9.114.2). The commanders follow the instructions of the Athenian community at home (9.117). Xanthippus is not prominent in the overall narrative, so the crucifixion of Artayctes, carried out by his orders, tends to convey the barbaro-monarchic mentality of postwar Athens as a whole.46 But Xanthippus’ son Pericles, chronologically the last Athenian leader in the Histories, is significantly introduced as a descendant of Cleisthenes, the one “who established the tribes and the democracy and who bore the name of his maternal grandfather from Sicyon” (6.131.1). This gloss of identification inevitably recalls the earlier comparison between the democratic reformer and the tyrant precisely at the moment when Pericles flashes through the narrative of the Histories in the guise of a lion, symbolically completing the identification between leader and city.47 To some of Pericles’ contemporaries, the arche of Pericles in Athens resembled a tyranny.48 But Pericles himself, according to Thucydides, declared that the arche of the Athenians in Greece was “like a tyranny.” A tyrant is a “lion, mighty, ravenous” among his fellow citizens, as the Pythia says of Cypselus, (5.92β3). Pericles is the leader-turannos of the city-turannos, the representative of the sovereign demos, a lion at home and to the rest of the Hellenic world. The fluid naming of political realities in contemporary discourse represents the linguistic foundation of the analogical impact of Herodotus’ text. By glossing over substantial distinctions, the monarchic analogy implicates the Athenian demos as it does no other Greek community in the Histories and in a way that directly resonates with the fifth-century notion of the rise of a polis turannos in Greece.49 In this overarching perspective, the explicit comparison of the two Cleisthenes is a reading 46. The juxtaposition of the sacrifice of Eobazus by a Thracian tribe (9.119) and the Athenian execution of Artayctes (9.120) encourages comparison between the two actions. For the ambivalence of Herodotus’ representation of Athenian action at the end of the Histories, see Dewald 1997. 47. 6.131.2. For lion symbolism, see my chap. 4, “Wondering Why” and “Wonder and Disbelief.” For glosses of identification, see my chap. 1, “Referential Glosses.” 48. For Pericles as tyrant, see Plut. Per. 7.1– 2, 12, 16, Cratinus (frag. 240 Kock) quoted in Plut. Per. 3. Cf. Thuc. 2.65.9. For the Athenian empire as tyranny, see Thuc. 2.63.2 (words of Pericles); cf. 3.37.2. 49. For the idea of Athens as the polis turannos, see especially Thuc. 1.122.3     (τυραννον . . . πολιν), 1.124.3 (πολιν τυραννον), 6.85.1; Aristoph. Knights 1111– 12; Knox 1971, 53– 106; Raaflaub 1979; Raaflaub 1987, especially 223– 25, 241– 46. For the idea of the demos as tyrant at home, see Kallet 1998, 52– 54.

Comparison

59

direction whose importance is proportionate to the political change Cleisthenes initiated in Athens. The Monarchical Model in Sparta Elsewhere in Herodotus’ Greek narratives, the monarchical model more straightforwardly serves to dramatize an internal dilemma—the tension that exists between individual and state, even within a constitutional order as free as possible from the danger of true tyranny. Sparta not only represents the paradigmatic setting for this sort of confrontation but also provides the context for the second gloss of historical analogy, which I am now going to discuss. The seer Tisamenus of Elis performed the sacrifice for the Greek army before the battle of Plataea. The mention of his role provides the occasion for a narrative on how he and his brother had just recently managed to become Spartan citizens (analepsis: 9.33) and on the victories the Spartans subsequently achieved after he had taken permanent service with them (prolepsis: 9.35). I will focus on the first (analeptic) part, which triggers the comparison and is introduced, in terms of narratability, by the rather unusual marker of absolute uniqueness: “These were the only men in the world who became Spartan citizens.”50 Tisamenus belonged to the family of the Iamidae of Elis. On receiving a prophecy from Delphi that he was destined to win five victories of the greatest importance, he started training to become a champion in the pentathlon (the five agones of the pentathlon would give him five victories in one). This was not what the oracle meant, however, and Tisamenus did not attain the athletic success he expected. The Spartans understood the oracle and offered Tisamenus a post in their army. But when Tisamenus demanded as compensation that they make him citizen “with full privileges,” the Spartans were indignant and let him go. Subsequently, however, fearing the Persian invasion, they sought him out again and even yielded to his demand that his brother as well as himself be awarded citizenship (9.33.1– 5). The flashback I have just summarized belongs to the first of three narratives about Greek seers that interrupt the account of the battle of Plataea and mark a pause in the “real” action.51 The proleptic section of 50. 9.35.1. How and Wells (1928, 2:302) point out that this is an exaggeration in light  of 4.145, 4.149, and 7.134. On Herodotus’ use of µουνος, see n. 6 in the present chapter. 51. See Masaracchia 1975, 153.

60

Telling Wonders

the Tisamenus story serves to anticipate prophetically the outcome of the battle, because it mentions Plataea as the first of the five victories Tisamenus was destined to win for the Spartans (9.35.2). Seers are important indices on the battlefield of Plataea, since here more than ever we feel the presence of the divine.52 The stories of Tisamenus and the other seers are not, however, centered on the role of their protagonists as intermediaries between the divine world and human actions; rather, they seem to pursue a different set of themes. The first and third narratives are expanded glosses of identification for Tisamenus and Hegesistratus, the seers on the Greek and Persian sides, respectively. The second and apparently most accessory of the seer narratives brings in the mythical code, with the heroic diviner Melampus, and gives the key for interpreting the other two.53 It is introduced by a gloss of analogy that counterbalances the uniqueness of Tisamenus’ achievement of Spartan citizenship by pointing out its structural similarity with an earlier event of a different order.  In making these demands, [Tisamenus] was imitating [ε µιµεετο] Melampus, if one takes the liberty to compare/imagine a bid for  ε ι κα σαι βασιληι ην τε κα ι citizenship and one for kingship [ως  πολιτηι ην α ι τεοµενους]. (9.34.1)54 As in the glosses on the two Cleisthenes discussed earlier (5.67.1, 5.69.1), this statement expresses the analogy in story terms, although the 52. See Immerwahr 1966, 294. 53. For the famous Melampus (mentioned at 2.49.1 and 7.221) and his alter ego and brother, see especially Od. 11.281– 97 (where Melampus is not mentioned by name), 15.225– 46; Hes. Ehoiai frags. 37, 129, 130, 131 MW; Pherecydes FGrHist 3 F 33, 114, 115; Bacch. frag. 4.50– 51 SM; Pind. Paean 4.24– 26; Diod. 4.68.4– 5; Apollod. 1.9.11– 12. See also Lloyd 1976, 224– 25.  54. Α ι τεοµενους is a correction by Stein. Masaracchia (1975, 154 n. 173; 1978, 169–  70) stands by the manuscript reading α ι τεοµενος, referring it to Tisamenus and interpreting  ε ι κα σαι absolutely, in the sense of “suppose” (as at 1.34.1). He argues that with τε και , ως one must refer both βασιληι ην and πολιτηι ην to the same subject(s), with the translation “In saying this, he [i.e., Tisamenus] was imitating Melampus—so one may suppose—by asking for both the kingship and the citizenship.” My translation follows Stein and most other critics, but Masaracchia’s reading, which brings out the ambiguity of the gloss, (1) is consistent with the fact that nowhere else does the narrator apply the verb ε ι κα ζω in the sense of “liken,” to himself and (2) renders the connective force of the phrase τε και , which joins together the two terms more strictly than is expressed in the translation “if we compare people asking for citizenship and [people asking for] kingship.” I agree with Masaracchia that the narrator wants the listener somehow to envision Tisamenus as asking for the kingship.

Comparison

61

contextual link (both Tisamenus and Melampus are seers from Elis) this  ε ι κα σαι marks the comparitime remains implicit. The self-referential ως    son as an interpretation, like δοκεειν ε µοι in the Cleisthenes passage. The narrative records that when the Argives wanted Melampus to cure their women who had been seized by Dionysiac madness, he requested onehalf of the kingdom as compensation. The Argives at first refused. Since the women’s madness increased, they eventually decided to give him what he wanted. Realizing that the Argives were desperate, Melampus raised his price and demanded not only half the kingdom for himself but also another third for his brother. Reduced to a tight spot, the Argives agreed to both requests (9.34.1– 2). The concluding statement repeats the comparison: “In the same way the Spartans too, since they desperately needed Tisamenus, yielded to him in everything” (9.35.1). Mythical history is never a focus of Herodotus’ exposition; the myth of Melampus here provides an archetype to the modern story of Tisamenus and lifts it from the realm of the literal.55 An emergency places both individuals in a de facto position of power so that they obtain an exceptional political advantage that violates the city’s integrity.56 By using Melampus to interpret Tisamenus, the text emphasizes the invasive character of Tisamenus’ request and paradoxically transforms his achievement of citizenship into a metaphor for the acquisition of kingly power. As in the case of Cleisthenes of Athens and Cleisthenes of Sicyon, we are here in the presence of vertical analogy, in which the juxtaposition depends on the symbolic code. The narrative tries hard on its own to convey the idea that the position Tisamenus acquires in Sparta signifies something beyond what the story allows. Tisamenus enters the Plataea narrative as a seer, but the analeptic passage de-emphasizes that fact, even as it tells the story of how he became a seer. It does not remind the audience that the genos of Tisamenus, the Iamidae, is a famous family of seers, nor does it attempt to justify the strange fact that despite his background, Tisamenus misinterprets a prophecy, especially one concerning something as predictable as his future as seer.57 Tisamenus proceeds to train in athletics—an index 55. On Herodotus’ distinction between historical and prehistorical space, see especially 1.5.3 and 3.122.2. See von Leyden 1949– 50; Shimron 1973; Hunter 1982, 105– 6; Lateiner 1989, 118– 23. For the symbolic function of heroes in Herodotus, see Vandiver 1991, especially 68– 69, on Melampus. 56. Cf. Cobet 1971, 71. 57. The Iamidae are descended from Apollo; see Pind. Ol. 6.35– 72; Paus. 6.2.5. A “divine seer” plays a role in Simonides (frag. 11W2, line 42). It is tempting to speculate that

62

Telling Wonders

of political ambition—and not even his last-minute defeat at Olympia induces him to reconsider the meaning of the oracle.58 When the Spartans correctly interpret the oracle, the narrative implies that they understand the five victories prophesied by Delphi to be military victories to which Tisamenus would contribute as diviner. No one says so in so many words, however: “The Lacedaemonians, having realized that the proph ι ους αγ  ωνας],  ecy referred not to athletic but to martial contests [αρη tried to persuade Tisamenus to become, for a fee, leader in their wars   πολεµων],  [η γεµονα των together with those of the descendants from Heracles who were kings” (9.33.3). This job description forgets to limit Tisamenus’ new assignment to the field of religion, just as the preceding section of narrative has underplayed Tisamenus’ specialized vocation. Though the duties of the diarchs included priestly functions (see Xen. Lak. Pol. 15.2, 13) and though the vagueness of the Spartans is plausible, their reported expression nevertheless resembles a proposal that Tisamenus fulfill the same role as the Spartan kings. Neither a proven diviner from the start nor an athletic victor, Tisamenus is an ordinary man who obtains something unique and hard to achieve. Added to Tisamenus’ kingly position as “war leader,” the attainment of citizenship for himself and his brother is all that is missing for the metaphor of a second diarchy.59 Inserting the Melampus story through explicit comparison establishes that the audience should imagine (ε ι κα σαι) a vertical analogy between one who became king in a literal sense and one who became “king” in a figurative sense, because the narrator himself reads his material in this way. This gloss is therefore of some importance in endorsing the application of the symbolic code to the analogical interpretation of the Histories. As Tisamenus the citizen is like Melampus the king, so the reverse is true. Literal kingship is in Herodotus the paradigmatic manifestation of the abstract concept of personal power. Within Herodotus’ complex exploration of the ways and means of Simonides’ narrative of Plataea was somehow the source for Herodotus’ Tisamenus/ Melampus analogy, which has a poetic cast. 58. For the tyrannical connotations of athletic victories, see 5.71 (Cylon); cf. 5.47, 6.36, 6.103, 6.122, 6.125– 26. See Nagy 1990, 186– 87. On athletic victories by seers, see Pritchett 1979, 55, citing Pausanias (6.4.5, 6.17.5– 6) and Pindar (Ol. 6). 59. The brother’s name is Ηγι ας (9.33.5), which according to Macan (1908, 1.2.668) “is Ionic (and Attic) for Αγι ας or Αγι ας, a name perhaps identical with #Αγις.” The parallel with Melampus marks the end of the analeptic part of the insertion about Tisamenus. In the proleptic continuation, which has a different function, Tisamenus’ role as seer becomes again central and is explicitly mentioned (9.35.1).

Comparison

63

kingship (and metaphorical kingship), we recognize two partially distinct but overlapping patterns. The first, theorized through the words of Otanes in the Constitutional Debate and embodied in the figures of specific autocrats in the narrative, concerns the way in which the king exercises his power.60 More pertinent to the Tisamenus/Melampus sequence, however, is the second pattern, which focuses on how men who count as outsiders to the system attain kingly status. This pattern is especially important in the Histories because it provides a bridge between the inaccessible hereditary monarch figures and “every man”: any individual may upset the existing political structure by rising in status on account of chance or skill, often without the backing of force and even with the consent of those on whom he imposes himself. The royal bodyguard Gyges stumbles into the kingship under undesirable circumstances and makes the best of things (1.8– 13). Amasis, a man from the people, suddenly becomes king by the decision of the army and then proceeds to reconcile the rest of the Egyptians to his rule by means of sophie (2.162.1– 2, 2.172.2). Darius, an Achemaenid close to the throne, is nevertheless treated as a more or less ordinary Persian noble who takes advantage of a situation of crisis: he must first reinstitute the monarchy and then obtain the post for himself with a ruse (3.82– 83.1; 3.85– 87).61 The aristocrat Pisistratus drives around in a cart drawn by mules (an index of inferior status) and gains power by his tricks (1.59– 63, especially 1.59.4). By illustrating the accessibility of monarchical rule to the clever or lucky, these stories emphasize its essential illegitimacy. Thus, the founder of the Persian monarchy is the mule Cyrus, a mixed breed, the grandson of a king and a social outcast.62 Deioces, an especially insidious upstart, begins his career in a society that has achieved autonomie and eleutherie (freedom), and he leads it to permanent doulosune (slavery) and tyrannis all on his own (1.95.2,  and lusts for absolute power. Already 1.96.1). Deioces is clever (σο␸ος) highly regarded, he becomes even more so by practicing justice at a time of great lawlessness in Media. But when all depend on him for arbitration, 60. See discussion of the monarchical model earlier in this chapter. 61. Darius was the last of the seven to join the conspiracy (3.70) and was “a spearbearer of Cambyses and still a man of no great account” (3.139.2). 62. For Cyrus as mule, see 1.55.2, 1.91.5– 6. See also Nagy 1990, 335– 37. Persecuted babies who grow up to become rulers, such as Cyrus and Cypselus (5.92γ2– 4), illustrate the unthreatening beginnings of monarchical power. A marginal instance of the pattern of the rise to kingly status is the rascally thief who becomes the son-in-law of Rhampsinitus (see 2.121).

64

Telling Wonders

Deioces becomes unwilling to continue as judge; he claims that it is not expedient for him to neglect his own affairs and tend to those of his neighbors, settling their controversies day in and day out. Given the present state of lawlessness, the people decide to give themselves a king, and their choice falls on Deioces (1.97– 98.1). His actions from now on create the precondition for the exercise of absolute power enhancing the separation of the monarch from the community and his existence beyond and above the law.63 The end of the story of Deioces almost rejoins the other pattern that focuses on full-blown monarchical rule. The first part, however, describes the mechanisms of his coming to the throne and is strikingly similar to the story of Melampus. Deioces refuses to be judge among people who need his arbitrations, until they make him king. Melampus refuses to rescue the beleaguered city of Argos by his divination, unless he and his brother receive nothing less than the kingship. The narrative about Melampus reproduces the pattern of the Median Deioces in an ancient Greek setting, namely, Argos in the heroic age. But Melampus is brought in as the analogue of Tisamenus, who in turn transfers the pattern to modern Sparta. Each of these three men—Deioces, Melampus, and Tisamenus— possesses a special skill that the community needs, and each obtains a contract of power that satisfies his ambitions. The dual citizenship of Tisamenus and his brother is a term of the symbolic code of kingship as it emerges in the Histories from narratives about Sparta. Far removed from autocratic monarchy in the Eastern sense, Sparta is also historically immune from tyranny as is no other city in Greece.64 At the same time, Sparta provides monarchical paradigms through those who occupy or are close to occupying the peculiar office of the dual kingship. Thus, one issue that symbolizes the ambivalence of the Spartan kings as individuals vis-a-vis ` the constitutionality of their office is whether Demaratus is the legitimate heir to the throne or a more or less illustrious (mule-related) interloper.65 Many Spartans of royal blood in 63. Deioces’ literal isolation in the fortress of Ecbatana (see 1.98.2– 100) symbolizes his autonomy and unaccountability, such as, in the words of Otanes, “would place even the best of all men who occupies this office outside of customary ways of thinking” (3.80.3). This condition is a source of abuses, if not by Deioces himself, by his successors. 64. See 5.92α1– 2; Thuc. 1.18.1. 65. If he is not the son of his predecessor Ariston, he was fathered either by the hero Astrabacus (“the one with the mule saddle”) or by one of the household servants, the “guardian of donkeys” (6.68.2, 6.69.4). See Boedeker 1987a. For mules and kings, cf. my n. 62 in the present chapter.

Comparison

65

the Histories ardently desire this position or plot to obtain it,66 contract irregular marriages,67 or are guilty of treason for the sake of power or gain.68 The Spartan diarchy is literally different from other royal systems, but its significance in the logos rests largely on the fact that at Sparta, the men who make trouble truly bear the title of king.69 Through the particulars of Spartan history and prehistory, Herodotus’ narrative is able to shape the monarchical model in a special way that applies to the problem of leadership in the Greek states. In Herodotus, diverse pressures tear the Spartan diarchy. Given the nonindividualistic ideals that Sparta stands for (see, e.g., 9.71.3), the requirement to adhere to the ethos of the city as a whole is greater for these kings/nonkings than for the politically prominent elsewhere in Greece. At the same time, the inherited privilege of their office separates the diarchs from the rest of the city; they lie close to the dangerous sphere of autocratic transgression. The narrative connects the tension inherent to the status of the Spartan kings to the ambivalent legacy of their heroic ancestry, since the hero is both capable of the highest display of excellence and unfit to live with his peers.70 Thus, the closest Greek analogue to the despot Cambyses is the Heraclid furens Cleomenes. At the other end of the spectrum, we find the citizen-king Leonidas, in whose case the code of kingship exceptionally denies the monarchical model, just as the heroic code is able to transfer Homeric kleos to a hoplitic ethical context.71 Heroic antecedents within the monarchical model—emerge again with the introduction of Melampus as the archetype for Tisamenus (who through his own family of the Iamidae has a heroic connection of his 66. Examples are Theras (see 4.147.3), Dorieus (see 5.42.2), and Leotychides (see 6.65). 67. Examples are Anaxandridas (see 5.40.2), Ariston (see 6.62), and Demaratus (see 6.6.2). Eros for turannis (see, e.g., 1.96.2, 3.53.4) and unbridled sexual desire go together. See Hartog 1988, 330. “Doing violence to women” is one of the typical monarchical actions in the words of Otanes (3.80.5), amply illustrated in the narrative (see 1.8– 13, 1.61.1, 3.50.1, 5.92.η1– 3, 9.108– 13). 68. E.g., Leotychides takes bribes (see 6.72); Demaratus turns East (see 6.70.1– 2). 69. The issue of Spartan kingship in Herodotus is related to that of the paradoxical “foreignness” of this city. See “Implicit and Explicit Difference in the Ethnographies,” and “The Texture of Nomos” later in this chapter. 70. This is arguably one of the issues of Sophocles’ Ajax. See Knox 1961, 144– 48; Rose 1995. 71. The codes operative in Herodotus’ representation of the Spartan kingship are discussed in Munson 1993b. See also Boedeker 1987a. For the parallels between Cleomenes and the insane Cambyses, see Griffith 1988, 70– 71; Munson 1993b, 45 with n. 32. For the kleos of Leonidas, see chap. 3, “Specific Glosses of Interpretation,” in the present book.

66

Telling Wonders

own). In a minor counterpoint, Herodotus’ diviners and their heroic ancestors chime in with the kings to symbolize the relation of individual and state. This representation exploits the shared knowledge of a culture in which prophets and seers are close to the sources of power, predict and assist the rise of despots (1.62.4), influence policies (7.6.3– 4), compete with political leaders in deliberations (7.142.3– 7.143.1), and invent war strategies (8.27; cf. Thuc. 3.20.1).72 Like kings, seers, by definition, stand out from the citizen body. Normally outsiders, they become a part of the polis only to fulfill a leading public role. They are suited, therefore, to be taken as the doubles of kings, who also tend to claim special authority over prophecies.73 At the side of Leonidas at Thermopylae, we find as his equivalent Megistias the seer, who imitates the selfless action of the king by choosing to die with the three hundred; beside Leonidas, he alone is represented as making the choice (7.221). Just as Leonidas is the offspring of Heracles, so we are told that Megistias is from the stock of Melampus (7.221), that same Melampus who forces himself on Argos as king in the digressive narrative in the account of Plataea. A conflicting heroic heritage lies behind the present-day seers just as it lies behind the Spartan kings. The archetype Melampus produces two antithetical modern metaphors. One is Megistias, the citizen-seer, his descendant; the other is Tisamenus, who by becoming a citizen “imitates” Melampus, the king-seer. Implicit Analogy The Seers of Plataea In the Plataea narrative, the explicit analogy between Tisamenus and Melampus combines with an implicit but inevitable thematic comparison between the Tisamenus/Melampus doublet and a third Elean seer, eloquently named Hegesistratus, also connected with Sparta. While Tisamenus enters the logos of Plataea as the mantis of the Spartans and of the whole Greek side, Hegesistratus, an old enemy of the Spartans, now performs the sacrifices for the Persians. Hegesistratus’ appearance, like that of Tisamenus, gives rise to an expanded gloss of identification that explains how he came to fulfill his present role (9.37). Just as the unique72. Evidence for the importance of Greek seers in politics and war and for the sometimes problematic power relations between strategos and mantis is collected by Pritchett (1979, 46– 90; see 65– 70 for Sparta). 73. These include the Spartan kings: see 6.57.2, 5.90.2; cf. 5.93.2.

Comparison

67

ness of Tisamenus’ case justifies the inclusion of his story, so here superlatives advertise the narratability of the story of Hegesistratus. He belonged to another heroic family of seers, the Telliadae, and was a man “of the highest consequence” (9.37.1); at a time previous to the current events of Plataea, he performed “a deed that defies description” and “devised the boldest thing of all we know about” (9.37.2). Like the story of Tisamenus, that of Hegesistratus has an analeptic/ proleptic movement, since it begins at a time before the battle of Plataea and ends several years after it. Having suffered from this man many wrongs that remain unspecified, the Spartans had once captured and imprisoned him and were about to put him to death (9.37.1). But Hegesistratus cut off the instep of his foot to free himself from the stocks, escaped through a hole he had made in the wall of the prison, and by a painful march reached anti-Spartan Tegea. After his wound had healed,  he made himself a wooden foot and openly became an enemy (πολεµιος) of the Spartans. Hegesistratus’ hatred (ε χθος) against the Spartans did not in the end turn to his advantage, since the Spartans eventually captured him as he was prophesying in Zacynthus and put him to death (9.37.2– 4). The conclusion underlines the crucial elements of self-interest and hatred: “But the death of Hegesistratus occurred after the events of Plataea; at the time, he was sacrificing for Mardonius for no small compensation, and he did so with great eagerness on account both of his hatred [ε χθος] for the Spartans and his gain” (9.38.1). Tisamenus and Hegesistratus—the benefactor of Sparta on the Greek side at Plataea and the public enemy of Sparta serving the Persians—are antithetical figures. To the extent, however, that both dramatize friction between individual and community, they are also part of the same phenomenon. As a fugitive to the Persian side, Hegesistratus belongs to the group, well represented in the Histories, of those exiled or alienated Greeks who seek refuge in the East and in some cases “instigate incursions against their fellow citizens.”74 This type of actant includes, among others, the deposed Spartan king Demaratus (6.70, 7.3), the Pisistratids (6.94.1, 6.102, 6.107.1, 7.6.2– 5), and Themistocles (8.109.5)—all individuals who try to impose their will on a Greek community and are therefore, in this respect at least, also Tisamenus’ analogues. The Spartans capture Hegesistratus in Zacynthus, where they catch up with the fugitive Demaratus. Hegesistratus’ self-mutilation parallels the action of 74. Boedeker 1987a, especially 191– 92.

68

Telling Wonders

another royal enemy of Sparta, Cleomenes, who is also put in the stocks, obtains a knife, and proceeds to cut himself, though in a more selfdestructive way.75 In the intersection of the patterns and in the partial overlap between the otherwise opposite Tisamenus and Hegesistratus, the idea emerges that the enemy of a city will also potentially try to become its ruler; conversely, an individual who rises to power and benefits the city may become its ruler or its enemy. The last royal Spartan of the Histories to be implicated in these different stages of the monarchical model is the king-regent Pausanias. In the extratextual aftermath of Plataea, Pausanias became, like Hegesistratus at Plataea, a wanted public enemy of the Spartans, eventually suffering death at their hands (Thuc. 1.131.1, 1.134). A fifth-century Greek audience would have regarded Pausanias as a tyrannical type without any help from Herodotus. His had been a cause c´el`ebre, in which Athens and Sparta, for different reasons, had found themselves in mutual agreement in condemning the hero of Plataea.76 The result was probably a basically familiar story (though embodied in different versions somewhat varied in intensity and detail), about a “good” Pausanias before and a “bad” Pausanias after Plataea.77 Thucydides speaks of Pausanias’ eagerness to dominate Greece (Thuc. 1.128.3); he quotes a letter where Pausanias promises to Xerxes to make Greece subject to him and proposes to marry Xerxes’ daughter (Thuc. 1.128.7). The story evokes and perpetuates a special model of degeneration within the broader monarchical model.  Just as Herodotus’ Cleisthenes “imitates” (ε µιµεετο) the tyrant Cleisthenes, so in Thucydides the generalship of Pausanias after Plataea is “an imitation of tyranny” [τυραννι δος µι µησις] (Thuc. 1.94.3). Thucydides’ description of Pausanias’ brief despotic tenure at Byzantium includes predictable indices, what he calls “small matters that displayed what he wanted to do in the future on a larger scale”: inability to live in the 75. The parallels between Hegesistratus, on the one hand, and Cleomenes (see 6.75.2– 3) and Demaratus (see 6.70.2), on the other, are noticed by Macan (1908, 1.2.673, 675). 76. Fornara (1966, 266) argues that for the Athenians, the alleged aberrancies of Pausanias served to rationalize their taking over the leadership of the Greek allies; the Spartans, for their part, had put Pausanias to death (perhaps because they had evidence of his plotting with the helots; see Thuc. 1.132.4) and needed to justify that action. 77. Evans (1988) reconstructs two versions of the story of Pausanias, one corresponding to Thucydides’ account, and the other, somewhat less unfavorable, reflected in Herodotus. There is no evidence, however, of a favorable version of the story of Pausanias’ behavior after Plataea.

Comparison

69

established style, Median dress, a Persian table, foreign bodyguards, and an unapproachable temper (Thuc. 1.130.1– 2; cf. Thuc. 1.95.1). The narrator of the Histories communicates the relevance of the Pausanias story—and implicitly testifies to his audience’s familiarity with it—in two explanatory glosses in different contexts. Once, he offhandedly mentions “the hubris of Pausanias” in his interpretation of what happened after the Persian invasion (8.3.2). Another time, he brings up Pausanias’ lust for power and his Eastern marriage (though a different marriage than the one in Thucydides’ reported letter), in a gloss of identification for Megabates, “a Persian of the Achaemenid family, cousin [of Artaphrenes] and of Darius, the one to whose daughter at a later time, at least if the story is true, the Spartan Pausanias, son of Cleombrotus, arranged his betrothal, having conceived a passion [ε ρωτα] to become the tyrant of Greece” (5.32).78 Since Megabates commands the expedition against Naxos that Aristagoras has persuaded the Persian king to undertake for reasons of his own, the mention of Pausanias’ ties to the Persians via this same man reinforces the message about disloyal Greeks. In the narrative of Plataea, however, Herodotus evaluates Pausanias as the author of the “fairest victory we have ever known” (9.64.1) and surprises his audience by his apparent determination not to mention subsequent events. But the “Pausanias before” protests too much here and in terms better suited to remind everyone of the “Pausanias after” than to correct public fame.79 Confronted with three monarchic temptations of the type that Thucydides says he yielded to shortly afterward— mutilation, luxury, sexual indulgence (9.78– 79, 82, 76)—Pausanis selfrighteously rejects them all and declares himself at one with his fellow citizens (9.79.2). Though his refusal to defile the body of Mardonius is laudable, his contemptuous distinction between what a barbarian and a Greek would do contradicts the ideological thrust of Herodotus’ text and sounds excessively self-assured.80 Pausanias’ alleged contemplation of Persian riches and his experimental comparison between a Persian and a 78. See n. 67 in the present chapter: the two forms of the tyrant’s eros here become confused. The joke is noticed by Macan (1895, 1:176) and Shimron (1989, 66 n. 35). 79. Fornara (1971a, 62– 66, especially 64) calls Herodotus’ portrayal of Pausanias “a masterpiece of irony and a harbinger of tragedy.” Cf. Evans 1988, 4– 8. 80. See the narrator’s gloss declaring that Xerxes’ mutilation of the body of Leonidas was against Persian custom (7.238.2). See also Payen 1997, 174– 75. In his narratorial voice, Herodotus avoids explicit comparisons between Greeks and barbarians of the type made by his Pausanias.

70

Telling Wonders

Spartan banquet occur in a context of fascinated curiosity, ill-concealed greed, and ambiguous laughter.81 All this is not merely a favorable portrayal but rather a pretend inversion of the monarchical model that has been applied to all other Spartan kings in the Histories except for Leonidas. Like his seer Tisamenus, Pausanias gains a splendid victory for Greece. Like the seer Hegesistratus, however, he is already poised to become the accomplice of Persia against Sparta and all the Greeks. Thus, in antithesis to the pair Leonidas and Megistias at Thermoplyae, the diviners of Plataea on either side synthesize the ambivalence of Pausanias. As they do so, they symbolically carry forward and broaden Herodotus’ messages on the disharmonious combination of leadership and citizenship and on the dangers of prominent individuals for the city-states of Greece. The Last of the Seers Two more diviners appear at the end of the Histories in similar fashion as Tisamenus and Hegesistratus and demand association to the same analogical network: De¨ıphonus, who performs the sacrifice for the Greeks at Mycale; and especially Euenius, his father, “to whom the following thing happened” (9.92.2). At the end of the story of Euenius, the resumptive summarizing statement is followed by a gloss that undermines even this secondary connection to the main narrative: “but now I have also heard this: that De¨ıphonus . . . was not the son of Euenius” (9.95).  Just as Tisamenus and Melampus demand a special reward (µισθος, 9.33.3, 4; 9.34.1) and Hegesistratus sacrifices for no small salary  ο λι γου, 9.38.1), so De¨ıphonus plies his trade all (µεµισθωµενος ουκ  Ελλαδα ε ργα, 9.95). The narrative over Greece (ε ξελαµβανε ε π ι τ ην concerning Euenius separates the issue of compensation from the profession of seer, but it maintains the theme of the individual’s blackmail of the city by translating it into an ethical and juridical question of dike (justice). Euenius was one of the prominent citizens of Apollonia appointed for a year to guard at night a special flock that by daytime grazes along the banks of the local river. The flock is sacred to the Sun, and the people of Apollonia, who greatly revere it “on account of some prophecy,” give it 81. Fornara (1971, 62– 63) notices Pausanias’ interest in the spoils (at 9.80.1 and 9.82.1), and see 9.82.2. The reference to the helots at 9.80.1 may be an allusion. Pausanias’ laughter (9.82.3) is a negative index. See Lateiner 1977, especially 177.

Comparison

71

shelter at night in a cave. One night, during his turn as guard, Euenius fell asleep and about sixty of the sheep fell prey to the wolves. The citizens of Apollonia brought Euenius to trial and condemned him to be deprived of his sight. After the sentence was carried out, however, both the animals and the land became barren, so the Apolloniates consulted the oracles of Delphi and Dodona. Here they learned the reason for their calamity: they ι κως) blinded the guardian Euenius. The gods themselves had unjustly (αδ had sent the wolves; they would not, therefore, cease avenging Euenius until the people of Apollonia would make reparation by giving him any  thing he would choose and consider fair (ου προτερον παυσασθαι           αν  τιµωρεοντες ε κεινω  πρ ιν η δικας δωσι των ε ποιησαν ταυτας τας  ος εληται κα ι δικαιοι , 9.93.4). The gods, for their part, would grant αυτ Euenius a gift that would make many men regard him as fortunate. The people of Apollonia concealed this response and entrusted the matter to a group of fellow citizens, who approached Euenius. Sitting next to him on a bench, these men started speaking of various things until the opportunity came for casually asking him what reparation he would choose if the people of Apollonia should be willing to compensate him for  ελοιτο, ε ι ε θελοιεν  what they had done (τι να δι κην αν Απολλωνιηται    ε ποι ησαν, 9.94.1). Euenius, who knew δι κας υποστηναι δωσειν των nothing of the oracle, said that the gift of two fine estates and the finest house in town would constitute for him adequate compensation (δι κην   γενοµενην,  ο ι ταυτην αποχρ αν 9.94.2). As soon as he had finished saying that, those sitting at his side replied: “Euenius, the people of Apollonia pay you this compensation [δι κην] in accordance with the oracles.” When he learned about the response, Euenius made a great fuss because he felt he had been deceived. But the Apolloniates bought the properties he had chosen from their owners and gave them to him, and from that day on he also had prophetic powers and became famous (9.93– 94). I have paraphrased this story rather fully to convey its remote and idealized setting. The flock sacred to the Sun, grazing on the shores of the nameless river during the day and sheltered at night in a cave, evokes a primordial and mythical atmosphere of a community close to the gods. As with the cattle of the Sun in the Odyssey (12.127– 33),82 the violation of this herd creates a crisis. The wolves attacking the cattle, the trial, and the elders on the bench next to Euenius seem to have come out of the 82. See Vernant 1989b.

72

Telling Wonders

Homeric ecphrasis of the shield of Achilles, where two lions devour two oxen from the herd that pastures by the river, and where the elders sit in a circle on polished stones in the judgment scene (Il. 18.573– 81, 503– 4). The punishment sent by the gods to the Apolloniates for their unjust treatment of Euenius recalls the calamities by which Zeus avenges transgressions of Dike in Hesiod’s Works and Days: “great suffering, famine, and plague at once; the people perish away, women do not give birth, and households are diminished” (242– 44). In this archaic paradigm of the polis, where good government and communal deliberation prevail, a situation arises that, as in the stories of Tisamenus and Melampus, places an individual in a position of posing a threat. Here prophecy itself is a gift and a compensation from the gods, not a service in exchange for which compensation is due. But just as Tisamenus demands Spartan citizenship in exchange for his needed services as seer, Euenius, supported by prophecy of the oracle, could have demanded from the Apolloniates excessive compensation for his blindness. His reaction after learning about the oracle shows that he would have indeed done so. Kingship is not mentioned as a possibility for this price, but the earlier case of Melampus and the gift that Euenius actually requests (a choice portion of land) point in that direction.83 Unlike Tisamenus and Melampus, however, Euenius is never allowed to exceed his political status. A group of citizens delegated by the people keeps control of the negotiations and manages to correct the earlier miscarriage of justice according to divine injunction; at the same time, they preserve the city. The gods apparently do not object to the deceit, and Euenius settles down with their priceless gift of divination and with a reasonable human prize, neither of which violates the city’s institutions. He remains an ordinary citizen but is famous for his prophecy and richer than before; his new material possessions are bought at public expense, and the transaction wrongs no one. The issue of reward in this story, corresponding to the misthos (salary) of the seer in the other narratives, centers around the juridical, political, and religious notion of rightful balance, a Dike of Hesiodean stamp. 83. Cf. the “many gifts of land” obtained by Callias of Elis, another seer of the family of the Iamidae, as a reward for helping the Crotoniates in their war against Sybaris (5.45.2). Since the only other outsider who is supposed to have helped the Crotoniates in this war (though they deny it) is Dorieus, a royal Spartan, this points again to the Greek cultural notion of a metaphorical equivalence between seer and king.

Comparison

73

Unlike the former stories of seers, moreover, this one is pervaded with the words and actions of the gods—the prophecy regarding the cattle, the divinely induced curse on the land, the prophecy about Euenius’ compensation, and the gods’ gift of prophecy to Euenius. The gods supervise the human legislators and judges but leave them to their own devices in managing themselves politically. This archaic morality tale about remote Apollonia offers a vision of the righteous city and a hypothetical solution to the problem of the individual’s personal power and privileged status in the Greek polis. I have ranged far and wide on the basis of Herodotus’ two explicit analogies. In the historical narrative, glosses of comparison constitute the exception, and the conveyance of meaning does not depend on their presence. Both the explicit analogy of the two Cleisthenes and the Tisamenus/Melampus parallel, out of a mere handful of this type of interventions, confront a central issue in the Histories, that of kingship, and point to its broad metaphorical application. These rare visible stitches in a far broader analogical weave reveal the thought processes of the narrator and confirm a host of implicit analogical associations, horizontal and vertical, that the audience would have made elsewhere without prompting. Comparison in Space Although both history and ethnography advertise themselves as the report of extraordinary facts, the discrepancy between the two forms of discourse is perhaps nowhere more conspicuous than in the way in which they approach qualitative comparison. In the account of past events, particulars become mutually linked through implicit analogy; but comparison operates predominantly several levels below the surface, and the reconstruction of analogical networks is almost entirely dependent on the interpretive operations of the listener. The report of existents in faraway lands, by contrast, is from the start based on the qualitative comparison between what is to be found “over there” and what belongs to the familiar horizon of narrator and audience. In ethnography, therefore, and to a lesser extent also in geography, comparison is very much on the surface of the text even when it is implicit. Frequently it is explicitly the object of discussion.

74

Telling Wonders

Implicit and Explicit Difference in the Ethnographies Ethnographic description is designed to communicate and enhance the idea of difference.84 History follows a certain chronological and causal development, but not every historical narrative statement places the uniqueness of the fact it records at the center of attention. In ethnographic description, by contrast, relating a fact is generally a declaration of difference with respect to some other fact that the audience assumes to be in the normal order of things. They hunt locusts, and after they catch them they dry them in the sun and grind them, then they sprinkle the powder on milk and drink it. (4.172.1) Because ethnography is primarily a description of difference, it frequently expresses itself in negative statements that contradict the cultural or geographic norms to which the audience is accustomed.85 Similarly, advertisements of narratability in an ethnographic description tend to reinforce the narrative’s built-in raison d’ˆetre of reporting differences from anything Greek. These then are the customs [of the Thracians] that are most noteworthy. (5.6.2)  The customs qualified as ε πι␸ανεστατοι in the conclusion just quoted include almost complete inversions with respect to the Greek norm. For example, “the Thracians do not guard their young girls [notice the negative form] and allow them to have intercourse with any men they wish, while they strictly confine married women; they acquire the women they marry from their parents at the cost of great riches” (5.6.1); “Tattoos are considered a sign of nobility, not to be tattooed a sign of low birth” (5.6.2).86 84. See, e.g., Od. 6.119– 20, 9.105– 15. Redfield (1985, 99– 100) describes Herodotus as a tourist avid for difference. 85. For more on ethnographic negations, see chap. 3, “Cognitive Statements and Polemical Negations.” In the historical narrative, the statement that someone did not do something is an interpretive gloss that more generally contradicts the audience’s expectations according to their contextual or extratextual knowledge. 86. Inversion as a way of explaining the world “over there” is discussed by Rosellini and Sa¨ıd (1978, 985– 91) and Hartog (1988, 212– 16).

Comparison

75

Only exceptionally does ethnographic description need to state difference from Greece by explicit means.87 More frequent are Herodotus’ statements of absolute uniqueness: he refers to “the only men in the world,” “the only region on earth,” and a climate “not as among other men.”88 The incomparable natural phenomenon par excellence, both in a qualitative and in a quantitative sense, is the Nile. By virtue of some special power of its own, the Nile “is naturally opposite” (τα ε µπαλιν  πε␸υκεναι) from other rivers, since it spontaneously floods in summer when all other rivers are dry, making the Egyptians into a people who “do not plow.”89 The Nile’s unparalleled nature causes it to be something other than a river, more similar to a sea, with cities emerging on the  surface, “very similar in a way” (µαλιστα κ η ε µ␸ερεες) to the islands on the Aegean (2.97.1). Here the Egyptians sailing across the flooded plain and the Scythians driving their cart on the frozen sea (4.28.1) are complementary visions of the extraordinary.90 Implicitly but unmistakably, Scythia is the polar opposite of Egypt; and each of the two is not only different from Greece but utterly unique.91 In correspondence with the uniqueness of the nature of Egypt and its river, Herodotus declares the absolute uniqueness of the Egyptians in the programmatic introduction to the ethnographic part of the logos. The statement begins with a celebratory gloss that proclaims that Egypt is more worthy of narration than any other region because it has “the most wonders and works superior to description” (2.35.1). The comparative glosses in the next sentence explicitly connect this pure advertisement of narratability to the fact of Egyptian qualitative differences in the two 87. It does, e.g., at 1.131.1, with polemic force (see chap. 3, “Cognitive Statements and Polemic Negations”). 88. See 3.107.1 (Arabia is the only land to produce incense, etc.); 4.184.1 (the Atarantes are the only men we know who have no individual names); 2.104.2 (the Colchians, Egyptians, and Ethiopians are the only men who practice circumcision); 2.68.3 (uniqueness of the crocodile); 3.113.1 (unique sheep); 3.104.2– 3, 4.28.2 (uniqueness of climate; see Hartog 1988, 29– 30). 89. 2.19.2– 3 (cf. 2.24.1; 2.14.2; 2.12.2). Also, with respect to the abundance of its resources and volume, “no other river can be compared with it” (4.53.1; cf. 2.10.1– 3). See also 4.50.1, comparing the size of the Ister with that of the Nile. 90. Cf. 2.92.2. See Corcella 1984, 71. 91. On the implicit Egypt-Scythia polarity in Herodotus, see Redfield 1985, 106– 7. Its principal terms are hot/cold, no rain/rain in summer, changeable Nile/unchanging Ister, one river/multiple rivers, old ethnos/young ethnos, culture/nature, complex customs/simple customs, many wonders/few wonders, immobility/nomadism, centralized monarchy/multiple kings. The pairing of Egypt and the North occurs already in Pind. Isthm. 6.23.

76

Telling Wonders

specific areas of nature of the land and culture of the people. They establish, in fact, a double polarity. The Egyptians, at the same time as they have a CLIMATE that is different [ε τεροι ω  ] and the RIVER, which has a NATURE unlike  [␸ υσιν αλλο ι ην] that of the other rivers, [so] for the most part established LAWS AND CUSTOMS that are the complete opposite [πα ντα ε µπαλιν] to those of the rest of mankind. (2.35.2) The preliminary list of about twenty-three customs that immediately follows (2.35.2– 2.36.4) is designed to illustrate what the Egyptians customarily do παντα ε µπαλιν (completely opposite) not just from the Greeks (as is implied of the Thracians at 5.6, cited earlier) but from the rest of # mankind, ο ι αλλοι ανθρωποι (or ο ι αλλοι, ωλλοι). This declaration of utter Egyptian difference gives the audience a jolt by establishing a new subdivision of the world, in which the Greeks become marginalized. The opposition between the Egyptians and everybody else—rather than, in the usual way, between the Greeks and the non-Greek world, Egyptians included—is a part of Herodotus’ polemic against both the Greeks’ sense of being a special nation (even special in the sense of their being the exclusive representatives of normalcy) and the dismissive attitude they affect toward Egypt in particular.92 No gloss in the Histories proclaims the uniqueness of the Greeks, and only two passages attribute to all barbarians a nomos that the Greeks do not have.93 In a rare instance where Herodotus attributes the same nomos to all barbarians, the statement highlights the similarity between the Spartans and the barbarians.94 The Histories both presuppose as a given and discourage the commonplace notion of a Greek/barbarian polarity. Other glosses expressing the cultural uniqueness of a people vis-a-vis ` ο ι αλλοι ανθρωποι almost exclude a few remote ethnea from the mainstream of 92. Lloyd 1976, 310. 93. At 8.105.2, the barbarians of Asia are meant. The gloss at 1.10.3 (“Among the Lydians, and among almost all barbarians, even for a man to be seen naked brings great shame”) is in its context designed to emphasize similarity a fortiori rather than difference between barbarians and Greeks. On the quantitative evaluative comparison at 1.60.3 (Greeks more intelligent than barbarians), see chap. 3, “Explicit Evaluation.”  ος]  as 94. “The custom of the Lacedaemonias at the death of their kings is the same [ωυτ that among the barbarians of Asia. For, as a matter of fact, most of the barbarians follow the same custom at the death of their kings” (6.58.2). Cf. n. 161 and corresponding text in the present chapter.

Comparison

77

humanity, be it Greek or non-Greek: “their language is like no other because they squeak like bats” (4.183.4); certain peoples of the Caucasus and India “couple in the open like cattle.”95 Scholars have identified several systems that underlie Herodotus’ overarching conception of world differentiation. These schemes reflect our modern understanding of partially inherited forms of classification, which enabled the Greeks to define their own identity by supplementing and interpreting what they knew about exotic cultures in an unconsciously organized way. Thus, Rosellini and Sa¨ıd have traced the structural correspondences among the sexual, dietary, and religious customs that Herodotus attributes to the most remote and primitive peoples he describes. This demonstrates the consistency with which these three areas of culture are made to diverge from the Greek norm of monogamous marriage, on the one hand, and the consumption of cooked and cultivated cereals and boiled or roasted meat in the ritual context of sacrifice, on the other. The three marginal and mutually distant peoples said to “couple like cattle,” an index of extreme primitivity, also have alimentary practices that are equally abnormal in each case. Their sexual promiscuity goes together with a diet that diverges either on the side of omophagy (consumption of raw meat or fish, cannibalism) or on the side of vegetarianism (uncultivated, uncooked herbs and cereals).96 Redfield formulates a somewhat different set of principles by subdividing Herodotus’ peoples into “hard” and “soft” cultures. The first are represented by Scythians, Massagetae, Thracians, Ethiopians, and, relatively speaking, Greeks; the second by Egyptians, Babylonians, Lydians, and, relatively speaking, Persians. Soft peoples are characterized by luxury, the division of labor, and complexity of nomoi, especially in the sphere of religion; hard peoples are simple, harsh and fierce. Among soft peoples market exchange proliferates; hard peoples rely on gift and theft, the heroic mode of exchange. Soft peoples centralize resources through 95. 1.203.2 (κατα περ τοισι προβατοισι), 3.101.1; cf. 4.180.5. See Hartog 1988, 226. In two cases, glosses of absolute uniqueness draw attention to the individuality of the Greeks’ close neighbors, Caunians and Lycians (1.172.1, 1.173.4). Uniqueness is also attributed to the Ethiopians, in the same terms as to the Egyptians (3.20.2) and, within Persian society, to the Magi (1.140.2). 96. See Rosellini and Sa¨ıd 1978, especially 955– 60. On the Greek dietary norm, see Detienne and Vernant 1989, especially the essays by Detienne, Vernant, and Durand.

78

Telling Wonders

taxation, build monuments, are literate and organized; their politics tend toward tyranny. Hard peoples have relatively weak political organizations and tend toward anarchy. Soft peoples tend to acculturate the dead, hard peoples to naturalize them; among hard peoples women are treated as an abundant natural resource, more or less freely available, whereas among soft peoples women tend to become a commodity, disposed of by sale, through prostitution, or otherwise. Hard cultures fall short of civility; they are unwelcoming and difficult to visit. Soft cultures are confusing and seductive, difficult to leave once visited.97 This brilliant synthesis sheds light on the links between ethnography and history by refining Hellmann’s pattern of the “primitive opponent.” In Herodotus, in fact, no “soft” people conquers a “hard” people; “hard” people remain free or even conquer their “softer” aggressors.98 Other implicit systems have been detected as determining Herodotus’ description of the world. According to the conceptual map devised by K. Muller, ¨ for example, high cultures are placed in the center of the oikoumene (Egyptians, Babylonians, Lydians, Persians, Greeks), and primitivity increases as one proceeds outward, with builders, cattle raisers, and hunters/gatherers distributed in concentric circles toward the edges.99 These conceptual reconstructions provide us with a vocabulary and fundamental frameworks for discussing Herodotus’ ethnographic material. We should recognize, however, that they sometimes constitute devices by which we attempt to make sense of the apparent disorder of Herodotus’ description, more than actual distillations of the way in which Herodotus represents the world. Structuralist studies, such as those of Rosellini and Sa¨ıd, are especially prone to shift from an analysis of the discourse of Herodotus to a discussion of the mythical forms of thought that inform his material.100 The different schemes that have been proposed, moreover, cohere with one another only up to a point. They sometimes break down altogether and fail to account for numerous narrative details and metanarrative interventions, because Herodotus both is subject to cultural ways of thinking and rebels against them, is both a 97. 98. 99. 100.

Redfield 1985, 109– 10. See Redfield 1985, 112– 13; Hellman 1934, 80– 89; Cobet 1971, 104– 20. Muller ¨ 1972, 121– 22. See also Romm 1992, 46– 47. See, e.g., Mora’s criticism (1985, 60– 72) of Rosellini and Sa¨ıd 1978, 962– 66.

Comparison

79

lover of symmetry (as Redfield and others have maintained) and contemptuous of it. Differentiating from Within When glosses express dissimilarities between two foreign peoples, rather than uniqueness in an absolute sense, they emphasize what a complicated and irregular place the world is. When they mark differences between ethnea that live in the same general area or tribes belonging to the same ethnic group, they seem especially designed to discourage schematization. Based on the ethnographer’s detailed knowledge of people and places, the narrative reveals to the audience that difference manifests itself in ways they might not expect. The common stereotype of rudeness, primitivity, and poverty the Greeks frequently attached to the words Scythian or Thracian (names for “hard” cultures according to the modern critical notion) is, for instance, shown to be inadequate in the case of the Agathyrsi, who are neighbors of the Scythians but “extremely luxurious and wearers of gold” (4.104). The Agathyrsi also “practice the community of women, so that they may be brothers of one another and, being all related, not have mutual envy or hatred.” “In their other customs, however,” notes Herodotus, “they come close to   the Thracians [Θρηιξι προσκεχωρηκασι]” (4.104). The portrayal of the Agathyrsi forces us to a further distinction. In the poetic and ethnographic traditions about remote peoples, Lovejoy and Boas have long ago identified “hard primitives” and “soft primitives,” a categorization that Romm has recently expanded.101 The first group includes the tough and relatively poor Scythians, Thracians, most Libyans, and the extra-Herodotean Arimaspi. The second—Herodotus’ Argippaeans, or Bald Men; the Ethiopians; the Hyperboreans of tradition—are the simple but joyful cultures blessed with Golden Age abundance and a natural state of peace and justice.102 Presenting an idealized alternative to the phthonos (envy) among citizens in a Greek polis, the communal life of the Agathyrsi fits into this traditional type.103 But the expression that    οτατοι describes their prosperity— αβρ . . . κα ι χρυσο␸οροι [extremely 101. Lovejoy and Boas 1935, 287– 90; Romm 1992, 47– 81. 102. See Lovejoy and Boas 1935, 1– 15; Rosellini and Sa¨ıd 1978, 962– 66. 103. On the community of women and children in Greek utopia, see Aristoph. Eccl. 613– 35; Plato Rep. 463c– 465b, criticized by Arist. Pol. 2.1.4– 18 (1261b16– 1262b37). For Greek traditions about exotic family customs, see Pembroke 1967.

80

Telling Wonders

luxurious and wearers of gold]—does not suit idealized Golden Age  connotes cultural refinement, even a degree  ος savages. The adjective αβρ of effeminacy; it is only used to qualify advanced cultures (“soft” in Redfield’s sense of the word) and in particular the notoriously delicate Lydians.104 The resulting combination of three different stereotypes (hard primitives, soft primitives, and soft civilized people) creates a sense of unexpected and asymmetrical difference. In Libya, the nomads east of Lake Tritonis and the nonnomads west of it “do not have the same customs” (4.1871). Beyond the borders of Scythia, the rule of extremes at the edges accounts for the presence of the most just and peaceful of men (4.23) and, to the southwest of these, of the Androphagoi, a “peculiar people” that ignores the very notion of justice and “the only ones . . . who feed on human flesh” (4.18.3, 106). At the opposite extremity of the earth, the Libyan “Country of Wild Beasts” contains “the wild men and the wild women” mentioned at the end of the list of monstrous animals (4.191.2– 4). Yet this is also the home of the Maxyes, who wear their hair long on the right and shaven on the left and paint their bodies red: they, at least, practice agriculture, “own houses by custom,” and “say that they are descendants from the men who came from Troy” (4.191.1). The civilized features of this ethnos and their link to the heroic splendor and the past of the Greeks throws somewhat out of kilter the symmetrical correspondence with the Androphagoi on the basis of bestiality.105 Herodotus’ explicit comparisons reinforce the message of irregular differentiation both with and without the presentation of differentiating material. Though neighbors of the Scythians, the Melanchlainai are αλλο  [a different people, not Scythian].106 The narra εθνος κα ι ου Σκυθικον tive does not substantiate the gloss in this case, and elsewhere another gloss assimilates the customs of the Melanchlainai to those of the Scythians, aside from the former’s black cloaks (4.107). Also, the uniqueness of  “of the delicate feet,” when she advises him to 104. The Pythia calls Croesus ποδαβρε,  or good”  ον flee the Persians (1.55.2), who are still a “hard people” and “have nothing αβρ (1.71.4). For habrosune outside Herodotus, see Nagy 1990, 282– 85. 105. Pace Rosellini and Sa¨ıd 1978, 960. Herodotus records, without explaining them, connections between the primitives of central Libya and Greek traditions (4.178, 179, 189, 190). At 4.170 and 4.180, he speaks of customs borrowed from the Greeks, in the first passage specifically from Cyrene. 106. 4.20.2 (contra Hecataeus, FGrHist 1 F 185, who apparently defined them as  ε θνος Σκυθικον). The general tendency to group all northern barbarians in the category “Scythians” is noted by Strabo (11.6.2; cf. 7.3.9, quoting Ephorus, FGrHist 70 F 42).

Comparison

81

the Tissagetae, “a numerous and peculiar” people who live by hunting, remains elusive (4.22.1). Fully documented is the distinction between Budini and Geloni, who live in the same area but “do not at all have the η γλωσσ  η . . . ουδ ε same material culture or the same language [ου τ η αυτ η]”  and are “not at all similar [ οµοιοι]” in appearance and διαι τα η αυτ coloring (4.109.1). The red-haired and blue-eyed Budini are autochthonous nomads, while the Geloni (“which the Greeks wrongly call Budini”) are settled agriculturists with language and customs that correspond to their Greek origins.107 But an unpredictable peculiarity of the Budini   ταυτ  η] arises here: they are “the only ones in that region” [µουνοι των 108 who eat lice. Another explicit reference to the misconceptions of the general public concerns the distinction between Scythians and Massagetae, “which some say is a Scythian ethnos” (1.201). The Massagetae “wear the same dress and have the same material culture as the Scythians” (1.215.1), but in the area of social organization, the custom of having wives in common is only their own, “for what the Greeks say that the Scythians do, it is not the Scythians who do it but the Massagetae” (1.216.1). Isolated differences of various kinds also exist among tribes that belong to the same large ethnos.109 Glosses of similarity of the type “[The Gilgamae] have about the same customs as the others” (4.169.2) are not merely fillers that “make it possible to extend knowledge by moving from one group to its closest neighbors.”110 Rather, the ethnographer intervenes to say that neighbors share the same customs because he does not allow his audience 107. 4.109.1. Language is a frequent criterion of internal differentiation in this and  other parts of the world (see 4.106: γλωσσαν . . . ι δι ην; 4.23.2; 4.108.2; 4.117; 3.98.3; 1.57.3; 2.42.4; 2.105; 6.119.4; 7.70.1; 7.85.1). 108. 4.109.1. The statement has the same effect if we accept the interpretation of Stein,  who, on the basis of Photius’ definition of ␸θει ρ, understands ␸θειροτραγεουσι as “they eat pinecones.” But see, e.g., Macan 1895, 1:78, and cf. Strabo 11.2.1, 14, and 19. The case (mentioned again shortly) of the Libyan women “biting lice off themselves” (Hdt. 4.168.1) not only is decisive for the meaning of the word in this passage but also represents a pendant to this passage according to Herodotus’ procedure of balancing differences and similarities: some people eat lice; some people only bite them off. 109. They exist, e.g., among the Thracians: see 5.3.2– 5.4.1. In Herodotus, the word ethnos is used both in this sense (see, e.g., 4.197.2) and to denote smaller subdivisions within the same ethnic group. See Jones 1996; Hall 1997, 34– 40. The differentiation of nomoi within a single unitarian people (e.g., not all the Egyptians venerate the same gods in the same way; see 2.42.1) or within a system (e.g., there are different Egyptian practices concerning extraction of entrails depending on the festival; see 2.40.1) gauge the internal complexity of the system. 110. Hartog 1988, 226.

82

Telling Wonders

simply to assume that this is the case. Concerning the Adyrmachidai, who are the first group of Libyans one meets proceeding westward from Egypt, he relates that they (1) “generally follow Egyptian customs” but (2) dress “like the other Libyans” and (3) have a few customs all their own: they are “the only ones of the Libyans” whose women get rid of lice by biting them off themselves and throwing them away and “the only ones” who present to their kings the young girls to be married (4.168.1– 2). This passage illustrates Herodotus’ painstaking recording of differences and congruences among tribes that live in the same general area and his individualized portrayals of less well known ethnic groups.111 Explicit Sameness and Analogy The Texture of the Earth If ethnography and geography are first and foremost the representation of what is different for the listener, and if Herodotus seems particularly concerned with internal differentiation as well, what role does similarity play in his account of foreign lands and peoples? “The experience of difference,” says Corcella, “if it is to be a real experience with its level of intelligibility, cannot do without a certain recognition of similarities to the world of habitual experience. It is precisely such recognition which makes comparisons and translations possible.”112 In the case of Herodotus, however, we should go one step further and even talk of an active pursuit of the similar, which counterbalances his observation of difference in all areas of his ethnographic and geographic research.113 Explicit comparisons that establish that something is like something else are frequent and of many kinds, and they employ an extensive metanarrative vocabulary of similarity. The most interpretive or speculative ethnographic comparisons are occasionally accompanied by self-referential 111. Cf., e.g., 4.180.1 (“While the Machlyes grow their hair at the back of their heads, the Aseans grow it in front”), 4.178 (“Next to the Lotophagi . . . are the Machlyes, who also consume lotus, but less at any rate than those mentioned above”), 1.173.4, 4.17, 5.4.1. See also the catalogue of Xerxes’ forces (7.61– 95) in reference to the equipment of the various national contingents (7.62.1, 7.63, etc.), where the balance is between what is  ο (the same) or belongs properly to a different ε πιχ ωριον (strictly local) and what is το αυτ people. 112. Corcella 1984, 74. Herodotus’ practice of noticing correspondences is also emphasized by Muller ¨ (1972, 116). 113. Gould (1989, 11– 13) notices the physical dimension of this pursuit: the histor  travels to Thebes to verify whether the traditions there would correspond (συµβησονται) with those in Memphis (see 2.3.1; cf. 2.44.1).

Comparison

83

signs, especially the verb συµβα λλω.114 Two things that belong to different real or narrative worlds can be “put together” with respect to size or the degree to which they possess a certain feature, if their similarities in other ways furnish grounds for the comparison. Or two things can be compared qualitatively, with the reservation that excessive quantitative discrepancy in size may make the comparison invalid.115 If one is unable to “put together” (sumballein) something unfamiliar with something already known, the foreign object may appear strange indeed and aporia ensues.116 The ethnographer Herodotus is, however, a master both at recognizing difference and finding likeness. Excessive difference in nature, especially if not properly corroborated by autopsy or reliable verbal testimony, is rejected out of hand.117 For things that can be verified, geography in particular preserves the balance between the amazing individuality of a particular phenomenon and the need to integrate it into a unitarian world ruled by uniform natural laws. While Herodotus overstates his presentation of the Nile as the incomparable river/nonriver with a different phusis (nature), he nevertheless attributes its uniqueness to natural factors susceptible to observation, such that they either actually affect this or that other river as well or would do so if they occurred elsewhere.118 Herodotus’ assertions that the Nile “cannot be 114. Ε ι κα ζω as a self-referential metanarrative term in the sense of “liken” only occurs in the history at 9.34.1, where, if the reading of the passage is correct, it signals a vertical analogy that has almost the boldness of a metaphor. See n. 54 and corresponding text in the present chapter. In the narrative, ε ι κα ζω designates Gelon’s metaphor of the spring that has been taken away from the year (7.162.2), the assimilation of snow with feathers made by the Scythians (4.31.2), and the Ionian assimilation of Egyptian crocodiles with lizards (2.69.3). The imprecision of the operation indicated by ε ι κα ζω in the sense of “liken” is also present in the speculative ε ι κα ζω in the sense of “guess,” “imagine,” or “suppose,” which is a gloss of opinion used to mark a gloss of interpretation. 115. The narrator apologizes twice for putting together (συµβαλλειν) qualitatively things with widely discrepant size (2.10.1, 4.99.5). 116. For putting things together (συµβαλλειν) as a source of understanding (with συµβαλλεσθαι, “conjecture,” in glosses of opinion), see Hohti 1977. 117. E.g., the difference of the Arimaspi is rejected at 3.116.2: “Neither this do I   believe, that one-eyed men exist in nature [␸ υονται] with the same nature [␸ υσιν] as other men.” Cf. 4.25.1, 4.105.2, and the cautionary gloss of source at 4.191.4. Herodotus’ rejection of fabulous peoples represent an innovation with respect to his predecessors (see Scylax of Caryanda in Tzetzes Chiliades 7.629– 36, qtd. in Romm 1992, 84– 85). 118. For Herodotus’ scientific outlook on the Nile, see especially 2.20– 23 (criticism of previous theories on the flooding of the Nile), 2.24– 26 (Herodotus’ own explanation), 2.27 (explanation of the peculiarity of the absence of breezes). See Lloyd 1976, 91– 107. Herodotus’ insistence on conformity to physis is analyzed by Corcella (1984, 74– 84) and Thomas (2000, 135– 38). See also Donadoni 1947. For identity of natural processes in different

84

Telling Wonders

compared” (verb συµβα λλειν) in fact appear in contexts where he mentions other rivers that in one way or another resemble this one, at least “if one can put together [συµβαλειν] great things and little.”119 Comparing the Nile to the well-known Ister even leads Herodotus to theorize on the unknown sources of the former, because an observed similarity between the two rivers becomes evidence for conjecturing (verb συµβαλλεσθαι) one that is not visible.120 Both unique and similar to other rivers, Herodotus’ Nile fulfills two contradictory functions. On the one hand, unfathomable and mysterious, its sources lost beyond reach, the Nile is with Egypt as a whole the very symbol of the problems facing human understanding that the know-it-all Greeks, with their usual penchant for simplifying things, do not properly recognize.121 On the other hand, the strangeness of the Nile is itself limited by its participation in the general nature of the rivers of the world, which it cumulatively sums up and represents. Herodotus both seeks symmetry and rejects the assumption of it, because his attempt to make sense of things scientifically goes hand in hand with a fear of the ideological consequences of oversimplification.  His inference that the Nile and the Ister “are equal” [ε ξισουσθαι] (2.34.2) in conformity to a conceptual model appears to contradict his polemic elsewhere against the regularity of the shape of the earth as was represented in the early maps—a perfect circle, surrounded by Ocean and internally subdivided into the three great landmasses of Europe, Asia, and Libya (4.36.2, 4.42).122 The edges of Herodotus’ earth are parts of the world, see the gloss of similarity at 2.25.1. Herodotus’ view that something that occurs in nature is susceptible to reduplication under the same circumstances emerges in the report of the “wonder” of the skulls observed at Pelusion (3.12) and verified elsewhere: “I saw another similar case [κα ι αλλα οµοια] at Papremis” (3.12.4). 119. 2.10.1– 3. Cf. 2.29.3, 4.53.5  ε γω συµβαλλοµαι τοισι ε µ␸ανεσι   τα µ η γινωσκοµενα 120. Herodotus writes, ως  τεκµαιροµενος [as I conjecture, inferring the things that are not known from those that are  apparent] (2.33.2; see also 2.34). Lloyd (1966, 337– 44) cites Anaxagoras’ dictum οψις των  ηλων  αδ τα ␸αινοµενα [things that are apparent are the vision of things that are unclear] (Sext. Emp. VII 140 ⫽ DK 59 B 21a) as the first extant formulation of this use of analogy, widely applied by the Hippocratic writers. 121. See, e.g., the Greeks’ fanciful theories on the flood, by which they want to be “signaled for their cleverness” (2.20.1); their limited conception of the extent of Egypt (2.15– 16); their ethnologically ignorant version of the story of Psammetichus’ experiment (2.2.5) and tale about Heracles (2.45); and their “myth” concerning the Nile’s origin from Ocean (2.21, 23). 122. Cf. 2.23, 4.8.2 (see Lloyd 1966, 342; Gould 1989, 89– 90, Thomas 2000, 75– 101). See also the contradiction between the “symmetrical” assumption at 4.36.1 and the criticism of symmetry at 4.36.2 (see Romm 1989).

Comparison

85

irregular and provisional; the perimeter of the inhabited world is often merely defined by the impossibility for men to reach out and discern what lies beyond a physical obstacle.123 The large-scale subdivisions of the oikoumene are to him mere theoretical constructs with little empirical validity. Herodotus attributes tremendous importance to natural boundaries as symbols of the limits human action must respect. In this context, the separation between Asia and Europe is fundamental; ignoring it for the sake of aggression epitomizes adikie.124 A different set of epistemological and ethical principles, however, is operative in his geographical discussions. This disorganized stade-by-stade traveler refuses to interpret the earth in terms of global canonical subdivisions.125 These lead to unacceptable territorial claims and to the compartmentalization of experience.  The Persians consider as their property [ο ι κηιευνται] Asia and the foreign peoples who live in it, while they regard as something separate [ηγηνται κεχωρι σθαι] Europe and the Greek world (1.4.4; cf. 9.116.3) Herodotus rejects this sort of imperialistic allotment just as he devalues big geographical boundaries.126 Arabia does not end at the Arabian   ] (4.39.1). The notion of the three gulf except “by convention” [νοµω continents, he insists, falsely assumes the similarity of these lands in size  and shape, while “not small are the differences [τα δια␸ εροντα] between them” (4.42.1). The contours of Asia, except on the eastern side, “have υρηται  been found to be similar” [ανε οµοια] to Libya (4.44.3), but Europe is “equal” to both of them in length; in width, Europe “does not ι η]” (4.42.1). appear to me even comparable [συµβα λλειν αξ  4.36.2) for the Asia/Libya subdivision adumHerodotus’ scorn (γελω, brates an objection to continents in general, be they three or two. To him,  η γ η), and the three continents are only the whole earth is one (µι  η ε ουσ 123. There are numerous glosses of “not knowing” in reference to what lies beyond certain points: 2.31, 3.98.2, 4.16.1, 4.17.2, 4.18.3, 4.25.1, 4.27, 4.40.2, 4.45.1, 4.53.5, 4.56. See Romm 1992, 10– 37, especially 32– 37. 124. See my introduction. 125. Contrast the apparently regular percourse of Hecataeus’ Periodos Ges (see Pearson 1939, 30– 96). On the theoretical schematism of Anaximander’s map, see Van Paassen 1957, 57– 61. 126. See Thomas 2000, 98– 100. Cf. chap. 3, n. 176, on how Herodotus’ conception of geographical boundaries must adapt to ideological considerations.

86

Telling Wonders

names.127 They represent such a distorted view of the way in which lands and water are distributed that, among other things, they are even unable to take into account the existence of Egypt or at least its delta (2.16). A country, he interprets, is not a predefined geographical entity surrounded by physical boundaries but rather coincides with the area of habitation of a certain people (2.17.1). Herodotus does not replace the old schemes with a new one of his own making; instead, he uses the traditional names as a matter of convenience (4.45.5). But what travel and opsis empirically teach him of the physical world, and what he wants his audience to visualize, is rather the parceling up of the earth by multiple boundary lines and the assiduous and more or less random repetition of physical patterns and shapes, in various sizes, throughout the unitarian surface of the oikoumene.128 “Comparing large things with little” once again, Herodotus explains that “Scythia borders on the sea, to the east and to the south, just like [κατα περ] Attica,” and that “the Taurians inhabit it  ε ι ] in Attica a people  in about the same way as if [παραπλησια . . . ως different from the Athenians should inhabit Cape Sounion” or “as if in Iapygia a people other than the Iapygians were to begin at the Brentesian harbor and inhabit the cape up to Taras.” These are just two examples,  and there are “many other similar [παροµοια] promontories that Tauris 129 resembles [ο(ικε].” In physical processes, as in shapes, differences and similarities are everywhere mutually balanced. In the land of Assyria, it does not rain and the crops are irrigated by the river through hand-operated machines, “not as [ου κατα περ] in Egypt, where the river rises.” But “canals cut across all the territory of Babylon, just like [κατα περ] Egypt (1.193.1– 2).” In Scythia, the country most antithetical to Egypt, “rivers are not much fewer in number than canals in Egypt” (4.47.1). The nature of both lands is such that “men do not plow.”130 When differences between two regions are most striking, analogy operates across categories (vertical analogy) and by compensation: “Somehow, the extremities of the inhabited earth obtained as their lot the most beautiful things just as [κατα περ] Greece 127. See 4.45.2– 5; Lloyd 1976, 78, 83– 85; Bornitz 1968, 190– 92; Benardete 1969, 111; von Fritz 1936, 320– 30; Thomas 2000, 80– 86. 128. Cf. Immerwahr 1966, 316. 129. 4.99.4– 5. See also 4.156.3, 4.182, 4.183.1. 130. 2.14.2, 4.2.2. See n. 91 in the present chapter; Hartog 1988, 17– 18. See also the parallels between Babylon and Egypt (4.198.1– 3), Libya and Babylon (4.198.1– 3). On Libya and Scythia, see Benardete 1969, 121– 26; Rosellini and Sa¨ıd 1978, 973– 74.

Comparison

87

obtained the most beautifully mixed climate” (3.106.1). In a subsequent description, the eschatiai (edges) are shown to contain both the most beautiful and the ugliest of things, as when the most fragrant of Arabic spices, ledanon, is said to be found in the stinkiest of places (3.112). Yet the beautiful and its opposite are distributed everywhere, and the verbal  correspondence between τα καλλιστα [the most beautiful things] and τας    ωρας πολλον τι κα λλιστα κεκρηµενας [the most beautifully mixed climate] alerts us to a qualitative similarity hidden behind the opposition. Does Climate Determine Culture? The word somehow (κως) at 3.106.1 (just quoted) expresses the idea that the empirically observed equivalence eludes the clear cause-and-effect combination of scientific reasoning. In a similarly inexplicable way, the patterned texture of the physical earth extends to the men who inhabit it. When the programmatic introduction to the Egyptian ethnography, quoted earlier, states that the Egyptians, “together with” [αµα] their different climate and river, also have laws and customs that are the complete opposite of those of the rest of mankind (2.35.2), Herodotus is reformulating the fifth-century medical theory about the influence of climate with drastic changes. The Hippocratic Airs, Waters, Places, though not exempt from inconsistencies and self-corrections, attempts to establish a logical connection between the natural features of a land, the physique of its inhabitants, and even their moral character. The description of the marshy area around the Phasis exemplifies the mixture of the scientific and the ethical codes: here the climate is hot and rainy and the waters turbid; the inhabitants have big, thick bodies, no visible joints or veins, jaundiced complexions, and a natural sluggishness with regard to physical toil (Airs 15). All finer regional distinctions, moreover, are subordinate to the difference that according to the Hippocratic author separates Europeans from Asiatics. In Asia, the stability of the climate and the fertility of the land stunt the courage, industry, spirit, and strength of the people (Airs 13, 16). The Hippocratic writer’s interest in the scientific causes of ethnic differences is proportional to his penchant for evaluating foreignness as a set of pathologies, abnormalities, or diseases. Herodotus’ ethnographic discourse stays clear of both things. He notices that even the inhabitants of the same general area are often different, but he does not say why. Only occasionally and provisionally does the text suggest that climate may

88

Telling Wonders

partly account for the way people are.131 The introduction to the Egyptian ethnography, at any rate, does not scientifically explain the correspondence between land and people but treats it as a sort of mystery: “Egyptian customs differ from those of the rest of mankind not because Egypt’s geographical situation differs from the rest of the world, but just as it is different.”132 A vertical analogy is here established between land and people, by which one level of reality represents the other.133 The physical texture of a unitarian oikoumene finds its symbolic correspondence in the patterns of differences and similarities of nomos. The Texture of Nomos At the intersection between a people’s land and its nomoi, between the physical and the cultural, are natural resources and the manifestations of material culture that these resources determine: foods, fabrics and clothing, buildings and utensils. In this area especially, Herodotus employs the sort of comparative language that emerges in the geographic passage about the Thracian Chersonese cited earlier. Hartog emphasizes its rhetorical value for making the exotic understandable through an appeal to 131. See Lateiner 1986, 16, and now especially Thomas 2000, 103– 14. In the Histories, the idea that an infertile land makes people hardy and a fertile land produces “soft” men appears in the historical narrative, in the mouth of characters: see 7.102.1 (Demaratus) and 9.122.1– 2 (Cyrus), the second particularly close to Airs, Waters, Places 12, but not backed by the narrator’s authority and occurring in a notoriously enigmatic episode (see Dewald 1997; Pelling 1997). The narrator establishes in his own voice a connection between the nature of the environment and the people’s physical characteristics, at 2.22.3 (men south of Egypt are black because of heat) and 2.77.3 (a stable climate promotes good health; seasonal changes bring diseases). For effects of climate on a region’s fauna, see 4.29, 5.10. Elsewhere in the Histories, the influence of environmental determinism is muted. Lateiner (1989, 159) sees an implicit connection between Egypt’s stable climate and the Egyptians’ unwarlikeness, acceptance of despotism, and unchanging institutions (see, e.g., the interpretive gloss at 2.147.2), though he admits that the narrator never formulates that thought. Herodotus says that the Ionians have built their cities in the most beautiful climate of the world (1.142.1), and in some cases, he represents the Ionians as weak (see especially 1.143.2, 6.11– 12). Beautiful climate is, however, elsewhere attributed to Greece as a whole (3.106.1), and the weakness of the Ionians appears to be a matter of history and institutions (1.169.1, 4.137– 39, 4.142). Aside from the general “semiclimatic” and moralistic idea that a tough life makes a people tough, Herodotus’ message on why an ethnos is the way it is proposes a variety of partial or possible reasons, none of them with absolute and general validity. 132. Immerwahr 1956, 279. 133. The Egyptians are like their land somewhat in the same sense as Darius (albeit misguidedly) perceives that he and the river Tearos are alike and belong together, “the best and the most beautiful of all men” and “the best and most beautiful of sources” (4.91.2).

Comparison

89

the familiar.134 But comparisons of this kind are also a manifestation of Herodotus’ ideology of a unitarian world. The world can be explained in this way because it is “same,” with a limited number of possible shapes. When the river rises and the plain is flooded, a great number of lilies, which the Egyptians call lotus, grow from the water. People pick them and dry them in the sun and then ground what they extract  to poppy seed, and make from the middle, which is similar [ε µ␸ερες] from it breads baked on the fire. There is also a root of this lotus, which is edible and rather sweet, of the size of an apple [µεγα θος  to roses, κατα µηλον]. There are also other lilies, similar [ε µ␸ερεα] that grow in the river, the fruit of which is found in a separate calyx that sprouts on the side from the root, very similar in appearance   οµοιοτατον] [ ι δ εην to a wasp’s honeycomb: inside this are several  grains as large as [ οσον τε] an olive, which can be eaten fresh or dry. (2.92.2– 4) Each object that is “different” becomes disassembled into parts—roots, flower, fruit—and explained as a combination of similarities.135 The comparative discourse visually recalls familiar appearances, even in those cases when size is what is actually being compared.136 The second terms of comparison include natural and cultural products from the internally diversified Greek world: Cyrenaic lotus (2.96.1), Boeotian sandals (1.195.1), Lesbian craters (4.61.1). The Egyptian Hephaestus in Memphis is “very  similar [ε µ␸ερεστατον] to the Phoenician Pataiki.” If someone has never seen a Pataiki, another analogue is available from a different region: it is the “imitation” [µι µησις] of a pygmy (3.37.2– 3). Exotic objects that resemble things found in Greece or in other foreign lands are not merely 134. Hartog 1988, 225– 30. See also Corcella 1984, 69. 135. So also are the hippopotamus (2.71), the ibis, and the water snake (2.76.1, 3). The procedure is similar to that for describing hybrids, but the effect is familiarizing rather than the opposite. Cf. 2.68.2– 3, 4.23.3, 4.177. See Hartog 1988, 249– 50. Another type of gloss follows the pattern “similar in everything, except such and such” (see 4.183.3, 4.61.1, 4.23.3).   (4.192.1) 136. In Libya, there are oryxes “as big as oxen” [µεγαθος . . . κατα βουν]  ησι ε µ␸ερεστατοι]  and crocodiles three cubits long and “very similar to lizards” [τ ησι σαυρ (4.192.2); the giant ants in India are smaller than dogs and larger than foxes, and they dig in

αυτ

τροπον]   ον the sand “in the same way” [κατα τον as Greek ants, to which they are “very similar in appearance” (3.102.2). See also 2.67.2, 2.73.2, 3.100, 3.106.2, 4.192.3.

90

Telling Wonders

easier to describe; they reveal the patterned character of cultural artifacts all over the world. Human industry, like nature, is always different but imitates itself incessantly. Reminders of this fact shorten the distance between the rest of the world and Greece in ways that, given the historical framework of Herodotus’ descriptions, may suggest more intangible points of vertical analogy. The simple notice that the outermost wall of Median Ecbatana is

Αθηνεων   “just about the size of that of Athens” [κατα τον κυκλον  µαλιστα κ η το µεγαθος] puts two imperial cities side by side.137 Sparta momentarily intrudes in the Scythian narrative when the narrator stops to notice that the cauldron built by king Aryantas at Exampaios in Scythia is “six times as large as” the crater Pausanias dedicated on the mouth of the Black Sea.138 Just as quantitative comparisons concerning size also suggest analogy in shape, so comparisons between two objects cause an overlap of their respective contexts.139 Foreign objects may also be explained in terms of their similarity to something that appears to lie outside their own category. If we look at the explanatory analogies in texts that are not concerned with the description of the world (e.g., most of the Hippocratic treatises), we notice the special poetic quality of Herodotean comparisons.140 These set side by side war, sailing, hunting, agriculture, and building as branches of a unitarian, device-producing field of human activity. A boat can be compared to an ox to be curbed (2.29.2) or to a shield (1.194.2), because it constitutes in itself the manifestation and symbol of many different sorts of effort. The most famous of these Herodotean transferences, in an ethnographic gloss describing the relay system of Persian postal couriers,141 has an impact not unlike that of a Homeric simile: “the first rider delivers his charge to the second, and the second to the third, and thence it passes from hand to 137. 1.98.5. See Georges 1994, 140. 138. 4.81.3. This comparison occurs in a narrative full of implicit analogies with Spartan realities discussed later in the present chapter; (“The Sameness of the Scythians”). 139. For other comparison of man-made items, see 2.7.1 (the Egyptian road from the sea to Heliopolis is almost equal in length to the road that leads from the altar of the twelve gods in Athens to the temple of Olympian Zeus in Olympia); 2.170 (the pond in the precinct of Athena at Sa¨ıs is about the same size as the Round Pond at Delos); 5.59 (Cadmean letters are similar to Ionian letters); 4.74 (Scythian hemp is very similar to linen). 140. Hippocratic analogies tend to assimilate physiological processes of the human body to technological processes or the life of plants. Lloyd (1966, 345– 60) gives many examples. See also Lateiner 1986, 13. 141. On ethnographic glosses, see chap. 1, “Referential Glosses.”

Comparison

91

hand, just as the torch-race that the Greeks perform in honor of Hephaestus” (8.98.2). The function of technical explanation and illustration cannot be separated from the ideological message it communicates metaphorically: the moral equivalence between two different kinds of performances, with different practical goals, occurring in different cultural contexts. From the experience of “same shape, same size” or “same procedure,” Herodotus’ pursuit of the similar extends to the observation of “same function”: “X is to ‘them’ what Y is to ‘us’.”142 Just as a linguistic translation must overcome a conceptual discrepancy, so the ethnographer translates culture by adjusting the difference to point out equivalence: the Issedones clean out and gild the skull of a dead man and then use it “as a    λµατι], making solemn sacrifices every year: sacred image” [ατε αγα “each son does this for his father, just as the Greeks celebrate the Genesia  [κατα περ Ελληνες τα γενεσια]” (4.26.2). The natural and the literal have a greater role among the Issedones than among the Greeks, but the purpose of foreign and domestic custom is the same.143 Glosses of this type make explicit a procedure Herodotus follows as a matter of course. Every time he describes a practice that occurs with no altar, no fire, no libation, no flute, no fillets, no barley meal (1.132.1), or a ritual in which the victim is tripped from behind and strangled (4.60), and calls each of these a θυσι α (sacrifice), he posits an equivalence that turns otherness into otherwiseness.144 The practice of interpretatio Graeca does not in Herodotus have an ethnocentric impulse; it is rather the foundation of his relativism. I am now ready to discuss the role of comparative glosses noting that two or more peoples are in certain respects similar or follow similar practices. 142. See Hartog 1988, 227. 143. For the Greeks, this degree of contact with the dead body may have been regarded as polluting. See Garland 1985, 41– 47. On the Genesia and other Greek commemorative practices for the dead, see Jacoby 1944; Garland 1985, 104– 8. For other Herodotean equivalences, see 1.202.2 (Caucasian primitives inhabiting the islands on the Araxes River have discovered [verb ε ξευρι σκω] a particular fruit, which they burn and inhale to become inebriated, “as the Greeks with wine,” the local equivalent of a symposium), 3.98.4 (the Indians of the marshes wear a mat made of reed “like a breastplate”). Cf. 4.75.1 (for the equivalence between a Scythian hemp sauna and a Greek steam bath), 1.193.5 (for the equivalence between Assyrian cultivation of palm trees and Greek cultivation of fig trees). 144. Cf. Mora 1985, 57.

92

Telling Wonders

As a rule, similarities between different ethnic groups are taken as a sign of mutual contact,145 common origin,146 or borrowing.147 Herodotus points out the debt of the Greeks toward various barbarian nations much more frequently than the other way around.148 When emphasizing the debt of Greek culture to Egypt in particular, he creates a corollary to the greater antiquity, wisdom, and moral authority of the Egyptians.149 The facts that ritual practices of certain Greek sects “are in agreement”   [ οµολογ εουσι] with Egyptian religious regulations150 and that the Egyptians celebrate some of the festivals in honor of Dionysus “about in the

παντα Ελλησι] (2.48.2;  α σχεδον same way as the Greeks” [κατα ταυτ cf. 2.49.2) are presented as cases of Greek borrowing from Egypt.151 A gloss of this type records that the Egyptians depict Isis with horns 145. In the case of ethnea living close to each other or tribes belonging to the same large ethnos (e.g., Scythians, Thracians, or Libyans), such interventions serve to confirm the assumption of internal differentiation (see discussion of differentiation earlier in this chapter). Proximity is the easily inferred cause of similarity between Lydians and Greeks (1.35.2, 1.74.6, 1.94.1, 7.74.1), Indians near Caspatyrus and Bactrians (3.102.1), Libyan nomads and Egyptians (4.186.1). 146. See 2.104.1– 5 (Colchians and Egyptians), 4.108 (Geloni and Greeks). 147. This is true even when the historical circumstances of the borrowing are unknown, as in the case of the language of Caunians and Carians (1.172.1) and that of the dress of Sigynnae and Medes (5.9). On diffusionism, see Muller ¨ 1972, 19– 20; Lloyd 1975, 150. 148. The Greeks contribute the technique of welding iron to world culture (1.25), pederasty to the Persians (1.135), most of their customs to the Asbystae (4.170), and a ritual for Perseus to the Egyptians of Chemmis (2.91.4). 149. According to Herodotus, the Egyptians invented the year and its subdivisions (2.4.1); geometry, which was then introduced into Greece (2.109); altars, statues, temples, animal sculptures, and use of the names of the twelve gods, the last of which the Greeks took from them (2.4.2; cf. 2.43.2, 2.49, 2.50.1); divination (2.49.2, 2.52.2) and oracular shrines (2.54– 57); festivals and processions (2.58; cf. 171.2– 3); and hemerology (2.82) and the theory of transmigration of souls (2.123.2– 3), these last two followed by some Greeks. The Athenians, thanks to Solon, have adopted in perpetuity what Herodotus evaluates as an “impeccable” Egyptian law (2.177.2; cf. 2.160). Herodotus says that the round shield and the helmet have come to the Greeks from Egypt (4.180.4). 150. 2.81.2. On the relation that Herodotus establishes (here and at 2.123.2– 3) between Egyptian cult and Pythagoreanism and Orphism, see Froidefond 1971, 187– 89. 151. According to Herodotus, Greek borrowings from other peoples include, “interestingly enough” (αρα), from the Libyans, the dresses of the statue of Athena, the aegis, the ritual cry of women at sacrifices, and the practice of yoking four horses (4.189.1), as well as the god Poseidon (2.50.3); games from the Lydians (1.94.1– 2); hoplitic equipment invented by the Carians (1.171.4); the Dioscuri, Hera, Hestia, Themis, the Charites, the nereids, and ithyphallic Hermes from the Pelasgians (2.50.2– 2.51.1); the solar clock, the meridian, and the subdivision of the day into twelve parts from Babylon (2.109.3); writing from the Phoenicians (5.58); the Ionian dress, which is actually Carian (5.88.1).

Comparison

93

“just as the Greeks do” [κατα περ Ελληνες] (2.41.2). The similarity to the Greeks here works as a sort of protective shield under which the text raises for the first time the difficult issue of Egyptian theriomorphy. The identification of gods with animals and the worship of living animals appear to interfere with Herodotus’ agenda of demonstrating to a Greek audience the superiority of Egyptian purity, religiosity, and dedication to the divine.152 His account is fuzzy in this area, for it simultaneously communicates several overlapping and contradictory messages: (1) Egyptian animal worship is “different” and disturbing; (2) it is (a) purely symbolic and therefore (b) similar to some Greek representations of the divine; (3) it is “Good,” a special sign of the Egyptians’ closeness to the gods and of a profoundly instructive view about the mutual relationship among different forms of life in the cosmos.153 This conflicted attitude first comes to the fore in the excess of discourse that accompanies Herodotus’ account of Egyptian representation of the divinity the Greeks call Pan (2.46.2). He begins by underlining similarity (2b). In Egypt, painters and sculptors represent Pan with a goat face and he-goat legs, just as the Greeks do [κατα περ Ελληνες]. Next, he reassures by negation (2a). . . . not because they believe him to be of such form—no, they believe him to be the same as the other gods.    [ουτι τοιουτον νοµι ζοντες ε"ιναι µιν αλλ οµοιον τοι σι αλλοισι  θεοι σι]. Finally, he admits to some uneasiness with this type of thing (1). For what reason they depict him in this form, it is not too pleasant  ε στι λεγειν].  ον  for me to say [ου µοι ηδι 152. Egyptian zoolatry was repulsive or ludicrous to outsiders. See Lloyd 1976, 291– 96, especially 293– 94 for Greek and Roman evaluations. In a fragment of Anaxandrides (39 Edmunds) an Athenian character comically declares the incompatibility between Greeks and Egyptians largely on the basis of the latters’ zoolatric practices. 153. Cf. Froidefond 1971, 202– 3. The narrator’s ambivalence does not at any rate mean that the analysis “the humanity of the Greek gods makes the Greeks superior to the Egyptians” (Benardete 1969, 46) represents the message of the text.

94

Telling Wonders

The negative program in the sentence just quoted is consistent with the general introduction, where the narrator declares himself not eager   . . . προθυµος) (ουκ to give detailed report of “divine things” he learned, on the grounds that what different cultures know about the gods is “equal” [ ισον]; he will therefore avoid the topic unless absolutely com154 Both  ο του λογου   pelled by his argument (υπ ε ξαναγκαζοµενος). times, Herodotus expresses himself subjectively in terms of personal distaste rather than blame and invokes similarity or equivalence. One passage where he breaks his rule of silence, evidently compelled by the logos, produces an apologetic effect: it etiologically explains the Theban prohibition to sacrifice sheep and rams to Zeus and the statues representing Zeus with the face of a ram, with a local tradition that is comfortingly similar to the familiar Greek myth of Zeus and Semele. Heracles wanted to behold Zeus directly, but Zeus did not want to be seen by him. Since Heracles insisted, Zeus skinned and decapitated a ram and then showed himself to Heracles wearing the ram’s hide and holding its head in front of himself (2.42.3– 4). The myth proves appealing because it reduces theriomorphy to a disguise; it communicates the message that the Egyptians do not really identify Zeus with a ram, just as the interpretive gloss at 2.46.2 explicitly says that they do not really believe Pan to be  τοιουτον—a he-goat.155 The reminders that the Greeks also represent Isis with cow’s horns and Pan with the face and legs of a goat similarly have the function of diffusing strangeness. Yet the “difference” represented by the Egyptians’ peculiar relationship to the animal world is also treated as clearly axion logou. In the account of Pan’s cult in the district of Mendes, the explicit denial that Mendesian Pan is a goat competes with the implication that he really is one. The logical discontinuity of the interpretive discourse barely camouflages the idea: “This is why the aforementioned Egyptians do not sacrifice he-goats and she-goats: the Mendesians reckon Pan to be one of the eight gods” (2.46.1). Later on, the narrative makes a revealing transition from goats in general to one special τρα γος, venerated above all the others (2.46.3– 4). A gloss intervenes, to show that the identification between god and animal is inscribed in the language: “In Egyptian, both 154. 2.3.2. The need to protect the relativistic position is clearly one of the issues here, though not the only one. See Linforth 1924; Gould 1994, 92– 93, 103. Cf. the refusal to explain why animals are sacred in Egypt at 2.65.2 and 2.47.2. 155. For ram cults in ancient Egypt, see Lloyd 1976, 190– 94. Herodotus’ reductionism in explaining Egyptian theriomorphy is discussed by Gould (1994, 202– 3).

Comparison

95

the ram and Pan are called Mendes” (2.46.4). Finally, Herodotus provides a terse narrative of what he introduces as a shocking event and apparently interprets as a public ritual. In this district, in my time, there took place a prodigious thing  [τερας]: a he-goat coupled with a woman. This came about as a public exhibition [ε πι δεξιν].156 Among the entries in the preliminary list of Egyptian customs that are “opposite to those of the rest of the world,” one records, “the daily life of   other people is defined as separate [χωρ ι ς . . . αποκ εκριται] from that of beasts, but the Egyptians live theirs together with their animals” (2.36.2). Later on, an evaluative statement introduces a survey of sacred animals (2.65– 76) and directs the listener to interpret it as an illustration of the extraordinary Egyptian piety. The narrative describes the role of animals in religion, the symbiosis of the Egyptians with their animals, and the phusis (physical characteristics) and modes of behavior of the animals themselves. Like human beings, animals perform goal-directed action; at the same time, they seem to enjoy a special connection with the supernatural world. Thus, male cats contrive (σο␸ ι ζονται) to couple with the females depriving them of their young (2.66.2), while on other occasions 157 As the land of  cat behavior is divinely inspired (θεια πρηγµατα). thomata, Egypt once again puts into focus fundamental problems of analogy and difference. Elsewhere in the Histories, the observation of primitive cultures raises the issue of the cosmic relationship between the realm of animals, human custom, and the divine.158 But here is a developed society where the most cultured and pious of all people worship animals and where animals enact those same elementary impulses that 156. 2.46.4. The form of the discourse here (summary introduction, narrative, concluding gloss) resembles 7.57.1, but the interpretive gloss in the conclusion is opaque; see chap. 1, “What Is Metanarrative?” in the present book. On the tradition of bestiality in the Mendesian nome, see Lloyd 1976, 216. 157. 2.66.3. See Smith 1992, 7– 15, 96– 107. Cf. the funeral ritual of the phoenix (2.73) and the ibises defending the pass at Buto against the winged serpents of Arabia (2.75). See also the alleged wolf ritual at 2.122.3 and the cow statue with the attached story of Mycerinus’ daughter at 2.129.3– 2.130.1. 158. The notion of a divinely and naturally determined analogy between the animal and the human world is implied in the section on the extremities of the earth and especially at 3.108; see especially Pagel 1927, 30– 33, on this passage as a description of nature that parallels the historical process.

96

Telling Wonders

among human beings translate into nomos. Herodotus’ view of a vertical correspondence between different levels of reality, which elsewhere expresses itself in animal symbolism and analogy, is susceptible to historie especially in Egypt.159 In book 2, the contradiction between the sameness and otherness of the world emerges, then, as a theoretical principle, which encompasses the cosmological, the geographical, and the ethnological spheres. The initial introduction to Egyptian culture attributes to it customs that are all opposite to those of the rest of the world (2.35). In the course of the narrative, this statement is partially confirmed and partially corrected, as Egypt becomes archetypal in both historical and symbolic senses.160 Other similarities reveal an ethical affinity between Egypt and Greece that cannot be accounted for in terms of contact, influence, or borrowing (see, e.g., 2.92.1). In the section just preceding the narrative on Egyptian animal worship, this affinity between Egypt and Greece in opposition to the rest of the world is surprisingly formulated in terms of a clear-cut separation, first institutionalized by the Egyptians, between animals and humans. Almost all other peoples, except Egyptians and Greeks, have intercourse in temples and go to the temple after intercourse without having washed, considering that humans are like other animals  [κατα περ τα αλλα κτηνεα]. (2.64.1) Sparta, which throughout the Histories represents what is most desirable from the point of view of social and political ideology, is also the city of Greece where Herodotus finds the greatest number of correspondences with non-Greek societies, including the Egyptian society.161 The origins and causes of such correspondences remain indefinable. Similarly, in his discussion of the special honors the Egyptians reserve for their hereditary warrior caste, the Machimoi, the narrator stops again to notice a similarity with Greek, and especially Spartan, values and declines, to interpret it as yet another sign of Egyptian influence. 159. Froidefond (1971, 193– 94) underlines Herodotus’ philosophical interpretation of Egyptian religion as expressing the notion of a solidarity among living beings not without similarities with Orphic and Pythagorean doctrines. 160. See Benardete 1969, 47. 161. See 2.80 (respect for the elders; see Froidefond 1971, 176); 6.60 (hereditary assignment of professional occupations). Sparta is compared to foreign nations also at 6.58.2 and 6.59.

Comparison

97

I cannot specifically determine whether the Greeks have learned this also from the Egyptians, seeing that also Thracians, Scythians, Persians, Lydians, and almost all the barbarians regard as less worthy of honor than the other citizens those who learn a trade and their descendants, considering noble instead those who abstain from manual work and especially those who devote themselves to war. This is something, to be sure, that all the Greeks have learned and especially the Spartans, though the Corinthians are those who despise artisans the least. (2.167.1– 2) The two cultural features involved in this comparison—contempt of manual labor and admiration for the pursuit of war—find specific formulation in the list of Thracian customs (5.6.2). Of the other foreign peoples named in 2.167.1, the Scythians are described as warlike throughout, and the Persians appear even more aristocratically contemptuous of petty trades than do the Greeks themselves (1.153). The inclusion of the Lydians in the list is surprising, however—even more than the fact that the gloss is attached to the description of the unaggressive Egyptians. It shows the extent to which Herodotus wants to stretch the similarity.162 Wherever the nomos comes from, and regardless of its cross-cultural and intracultural variations (e.g., the fact that in Greece it is followed more by the Spartans than by the Corinthians), it represents an area of ideological agreement between and within two internally differentiated groups— “almost all the barbarians” and “all the Greeks.” Many other glosses in the Histories underline resemblances between a foreign culture and the Greeks or between one foreign culture and another. Formulae of the type “about the same, except for . . . , ” also found in the description of exotic animals and plants,163 place the narratable difference in the context of the overall similarity of certain customs (see, e.g., 1.74.6). Such formulae also point out the overall similarity of two people in a specific area of culture, with the exclusion of a smaller group within a larger ethnos. The similarity must remain pure discourse and devoid of narrative content, while the difference is narrated, often with an unforgettable visual detail (see 4.190). In most cases of this sort, once again, similarity remains unexplained. Nasamones and Massagetae live at the opposite ends of the world, but 162. On Lydian unwarlikeness, see “The Sameness of the Lydians” later in the present chapter. 163. See end of n. 135 in the present chapter.

98

Telling Wonders

(predictably) they are characterized by about the same degree of primitivity; as a consequence, they have similar marriage customs, just as the Babylonians share certain cultural practices with the equally developed Egyptians.164 But Herodotus also establishes connections between societies with discrepant levels of culture.165 The frequent cross-cultural references convey two distinct, but not incompatible, attitudes. Firstly, Herodotus’ historie teaches that long-range coincidences, recurrences, and overlaps are something we can count on in the sphere of culture, just as it verifies the more or less random repetition of small patterns on the physical surface of the earth. This assumption of similarity can even provide some corroboration for the existence of an unseen cultural phenomenon, if one knows about the existence of one like it somewhere else.166 Secondly, however, recurrence and similarity, though likely, are underrated phenomena, which the narrator points out as worthy of great interest. The tradition of ethnographic discourse to which Herodotus is the heir is overdetermined in the other way. Difference is the standard of narratability and what the audience is readiest to expect and perhaps best trained to recognize—difference, most especially, from themselves. Particularly in the sphere of custom, therefore, Herodotus is generally more intent on showing the mutual equivalency of foreign practices and their functional character and normalcy than on letting his audience gape at their strangeness. When, instead of difference or equivalency, actual similarity occurs, it constitutes a “wonder,” that is, a profoundly satisfying discovery that invites reflection.167 The Egyptians follow their ancestral customs rather than acquiring new ones. Among other notable customs, they have also a particular song, “Linus,” which is also sung in Phoenicia, Cyprus, and other countries but changes name from culture to culture, though it  ος

συµ␸ ερεται  happens to be the same [ωυτ ε"ιναι] as the one the 164. On the Massagetae and Nasamones, see 4.172.2; cf. 1.216.1. See Rosellini and Sa¨ıd 1978, 975. On the Babylonians and Egyptians, see 1.182.1– 2, 1.198. 165. He establishes connections between Babylonians and the Lycians of Patera (1.182.3), Babylonians and the Eneti of Illyria (1.196.1), Babylonians and Cyprians (1.199.5), Ethiopian Macrobioi and Egyptians (3.24.2), Satrai and Greeks (7.111, where one item is part of an explicit comparison and another suggests implicit analogy). 166. See 4.33.5, 4.195 (cf. 2.150.2– 4). On the principle that “things that are apparent are the vision of things that are unclear,” see n. 120 and corresponding text in the present chapter. 167. See Lloyd 1975, 147; Corcella 1984, 90– 91; Giraudeau 1984, 121.

Comparison

99

Greeks sing; so, among the many things in Egypt that I am in  ζειν µε], there is also the question of  wonder about [αποθωµα where they found this “Linus.” They seem always to have sung this song. (2.79.1– 2) The Egyptian ethnography begins by presenting Egypt as the land possessing the greatest number of thomata, as well as nomoi and ethea that are all opposite to the rest of humankind (1.35). In the passage just quoted, a minute similarity is ranked among the Egyptian wonders: it connects not only Egyptians and Greeks, or Egyptians and Phoenicians, whose historical contacts can be reconstructed, but Egyptians and many other peoples from whom the Egyptians are isolated and to whom they are, as a rule, opposite. The feature involved in the similarity is moreover a specific—and one could almost say, unessential—ritual, something that belongs entirely to culture. Its peculiarity enhances one’s surprise at finding that it is shared. Pre-Socratic ideas concerning the basic homogeneity of humankind lead up to the principle, later in the fifth century, that the differences among people are a matter of convention, while uniformity of phusis (nature) is what truly counts.168 But Herodotus’ discovery about the ubiquity of “Linus” shows that for him the cultural uniformity one can perceive in the midst of the obvious differentiation of nomos is especially interesting, because it provides evidence for a connection between nomos and phusis and therefore for the equal importance of both in human events. Herodotus does not express himself in terms of an antithesis between the two and is far from proclaiming the importance of the one to the detriment of the other. The narrator or his characters qualify as     “human nature” (ανθρ ωπινος and ανθρωπ ηιος) human problems, sufferings, and limitations; what is humanly possible in contrast to what is not; and all the objective constraints that constitute not so much humanity itself as the condition that is imposed on all people of all ethnea.169 The 168. See, e.g., Antiphon, DK 87 B44, frag. 2. See Thomas 2000, 132. For discussion on ancient ideas of the unity of humankind, see especially Guthrie 1969, 160– 63; Baldry 1965, especially 23– 29.   169. Cf. το ανθρ ωπινον (human nature) as pattern of behavior in Thucydides (e.g.,   1.22.4, 3.82.3). In Herodotus, the phrase η ανθρωπη ι η ␸ υσις once refers to what a person     is capable of doing (3.65.3); the adjectives ανθρ ωπινος and ανθρωπ ηιος tend otherwise to  be referred to human affairs (πρηγµατα) and what happens to people (παθεα). See Powell  1938, s.vv. The notion of limitation is also implicit in Herodotus’ use of the term ␸ υσις in

100

Telling Wonders

complex of collective human responses to these constraints—we would like to say, the natural and at the same time divinely ordained sphere of such responses—is the “Nomos king of all” of 3.38.4, the embodiment of human impulse toward self-regulation and culture that generates all the different nomoi.170 A major task in Herodotus’ study of the internal differentiation of Nomos consists in verifying and pointing out to his audience the cross-cultural correspondences, equivalences, and similarities of the nomoi. Religious customs represent an especially important field of investigation in this respect, because they are less immediately functional and apparently less contingent on a people’s external circumstances than are other customs. They are, moreover, always based on theoretical and unverifiable constructs that vary from ethnos to ethnos. When the narrator states, “what all men know about divine things is equal [ ισον],” this generalizing comparative gloss couples the idea of contingent differences with that of essential uniformity. Herodotus continuously assumes or points out the cross-cultural correspondence of major figures of gods by translating divine names; for him, this constitutes an important proof that human societies tend to conceive of the divine in somewhat similar ways. The consensus of different ethnea over a minor ritual expression, such as the song “Linus,” carries the proof of the unity of cultures one step further. It is amazing precisely because it does not appear to be either mutually learned or a fundamental and predictable expression of human “nature” and the human experience of the world. Implicit Similarity and Analogy The Sameness of the Lydians In the ethnographic sections, glosses of similarity counterbalance the representation of difference. Further, like the far less numerous pointers of reference to one’s character affecting behavior, especially courage (5.118.2, 7.103.4,  8.83.1) or anger (7.16α1); in an ethnological sense, ␸ υσις denotes the temperament of a people (more or less aggressive; see 1.89.2, 2.45.2), which is distinct from, but not in   antithesis with, their νοµοι (in the plural; see 2.45.2). Otherwise in Herodotus, ␸ υσις is something concrete and uncontroversial, such as the family from which one is born (7.134.2) or the physical characteristics of people (3.116.2, 8.38), animals, or lands. Only at  4.39.1 does the implicit idea of ␸ υσις, in the sense of “natural lay of the land,” come into  antithesis with νοµος, in reference to the traditional (and inadequate) way of speaking about continents. See Heinimann 1945, 13– 41; Giraudeau 1984, 131– 32. 170. See chap. 3, “Funeral Customs and Other Nomoi.”

Comparison

101

analogy in the history, these cooperate with the effects of implicit analogy that the text achieves through narrative means. When it comes to foreign peoples and places, Herodotus talks to an audience of disbelievers. Even harder for the Greeks to accept than what may sound like the tall tales of travelers (1.193.4)—and more crucial for Herodotus to display—is the similarity of the other. After the assassination of an usurper to the throne, seven noble Persians come together to reconsider the form of government that would be best for their state. Can Persians discuss political theory and come to a reasoned choice as Greeks would do? Herodotus corroborates, “Speeches were made, unbelievable [απιστοι] to some of the Greeks, but they were nevertheless made” (3.80.1; cf. 6.43.3). The way in which the narrator marks the episode reveals his program to thematize similarity and overcome the prejudice of his listeners. Frequently in the text, an unexpectedly familiar feature is strategically planted in the midst of alien foreign actions or customs. A case in point is the statement “the Egyptians call barbaroi all those who do not speak the same language as themselves” (2.158.5). At a different level, the narrative elicits the audience’s self-identification, for example, by recording a special custom of the Trausians: “The relatives sit around the newborn lamenting the misfortunes of which he, since he has come into this world, will have to fill the measure and enumerating all the   . . . παθεα]. As for the sufferings that are the lot of men [ανθρωπ ηια dead, they bury him in the earth among celebrations of joy and merriment, considering all the evils from which he has been freed so that now  he is entirely happy [ε ν πα σ η ευδαιµον ι  η]” (5.4.2). The Trausian custom as such reverses birth and funeral and is opposite to the Greek practice. But at the same time, it represents the ritual enactment of a perception of human experience widely theorized by the Greeks as a true and superior insight.171 The cultural knowledge encoded in this custom raises an entire foreign people to the rank of those who are, by the audience’s own standards, especially wise.172 Toward the end of the work and within the narrative of Persian 171. The maxim that death is better than life is expressed in the Histories by Solon (1.31.3) and Artabanus (7.46.4). Asheri (1990b, 149) cites the parallels of Theognis 425– 28 and especially Hesiod frag. 377 MW: “et Hesiodus natales hominum plangens gaudet in funere.” 172. Asheri (1990b, 149) and others assume that the Trausian custom is based on a belief in immortality, such as Herodotus attributes to the Getae (4.93), but in Herodotus’ description of the custom, the eudaimonie attributed to the dead person seems to consist only in the deliverance from the evils of life.

102

Telling Wonders

debacles, the enraged Persian commander Artayntes almost kills Xerxes’ brother Masistes, who has accused him of having exercised his leadership  (9.107.1). An in a manner “worse than a woman” [κακι ω γυναικος] ethnographic gloss intervenes to explain that “among the Persians, to be called ‘worse than a woman’ is the gravest of insults” (9.107.2). What is the point of translating a cultural code that is exactly the same as that known by the audience?173 The narrator never uses it in his own voice but attributes it throughout the Histories to various characters, especially Persian, at the same time as he is accumulating evidence of vigorous female actions and of male inadequacies vis-a-vis ` women.174 At this particular stage of the narrative, the gloss underlines the convention to remind the listener that the current humiliating state of affairs for the Persians in the face of the victorious “masculine” Greeks is not the result of that society’s fundamental values. Rather, it derives from a historical process that has interfered with those values and prevented them from becoming actualized. Herodotus’ representation of the Persians proceeds, in fact, on two parallel tracks. On the one hand, they are throughout— from the times of Cyrus and the narrative time of the Persian ethnography to the battle of Plataea and the narration present of that same ethnography—masculine, tough, courageous, and dedicated. On the other hand, their “enslavement” to an autocratic ruler, the expansionism of their kings, the consequent acquisition of material goods, and the corruption of their leaders by luxury have cooperated to produce an inferior performance. This has happened even though the Persians value and cultivate what the Greeks call andreie, “masculine valor,” just as (implicit analogy) the Greeks do. The issue of andreie as a traditional standard of differentiation between Greeks and Eastern barbarians is prominent in Herodotus’ representation of the unwarlike Lydians, the first barbaroi who enter the Histories as an ethnographical subject, and the barbarians who, the text insists in explicit terms, most resemble the Greeks. In the regocentric narrative about Croesus (1.6– 91), one of the few passages that features 173. Greek characters and narrators in literary texts proclaiming that men should be “men and not women” and so on span a broad range, from the Homeric heroes to Socrates and beyond. See, e.g., Il. 2.235, 2.289, 8.163, 22.124– 25; Soph. Trach. 1071– 72; Thuc. 4.27.3; Plato Apol. 35b2– 3; Xen. Lak. Pol. 11.3. 174. 1.155.4, 1.207.5 (with 1.212– 14), 2.102.5, 7.11.1, 7.210.2, 9.20, and especially the whole Artemisia sequence (7.99.1, 8.68α1, 8.88.3). On women in Herodotus as foils for men, see Dewald 1981.

Comparison

103

the collectivity of the Lydians, a historico-ethnographic gloss with quantitative comparison, helps exonerate the Lydians from the defeat in the battle of Sardis: “At that time there was no people in Asia more virile     [ανδρηι οτερον] or stronger [αλκιµ ωτερον] than the Lydians. They fought on horses, carried long spears, and were good riders” (1.79.3). Later on in the history, we learn that after their conquest by Cyrus and subsequent rebellion, Croesus suggested to Cyrus that he impose on the Lydians a cultural change, so that they would not rebel again. The reform included a ban on all weapons and the prescription that the Lydians wear tunics and buskins, play string instruments, and bring up their children to  be shopkeepers (καπηλευειν). In this way, Croesus said, they would soon “become women instead of men” (1.155.4). This is a crude etiology for the effeminacy of the present-day Lydians according to the Greek stereotype; it dramatizes the consequences of their political enslavement to the Persians.175 But positioned between the two passages we have just seen, an ethnographic section provides another model of development and an intelligible context for what the history depicts as an abrupt transformation of the Lydians into “women instead of men.” This insertion, which I shall call the Lydian appendix (1.92– 94), also has another simultaneous—and one might say contrasting— agenda: that of establishing both explicitly and implicitly the cultural similarity of Lydians and Greeks and the status of Lydia as almost a nonforeign country. This explains why the passage, compared to the other ethnographies in the work, contains very little geographic and ethnographic information—no wonders, no lay of the land, no data on climate and production. According to the programmatic introduction, Lydia is the opposite of Egypt, with few thomata and little to tell (1.93.1; cf. 2.35.1). The city of Sardis, Croesus’ prosperous capital (1.29.1) later joined to Susa by the Royal Road (5.52), here receives no description. The golden dust washed down by the Pactolus (see 5.101.2; cf. 5.49.5, 6.125.4) is barely mentioned. The river Halys, Croesus’ fundamental “ethical” boundary in the historical narrative (1.6.2, 72, 75), is entirely absent.176 How does the information Herodotus chooses to include help to explain present-day Lydians while supporting the notion of their affinity with the Greeks?   175. See 1.94.7 (ε δεδουλωντο). Cf. ε ξανδραποδι σασθαι (1.155.1) and ανδραπο δισθεντας πρηθηναι (1.156.1). 176. See Jacoby 1913, 332, 339– 40; Lombardo 1988, 172– 81.

104

Telling Wonders

The Lydian appendix, subdivided into three clearly marked sections, occurs at the point in the logos when the story of Croesus has definitely ended; a dramatic aftermath (1.86– 91) has invited reflection on its complex set of meanings. The first section (1.92.1), on “Croesus’ other offerings in Greece,” is a delayed addition to the narrative of the king’s dedications to Delphi, which was prominent in the history of his policies (1.46– 55). The “other offerings” remind us of Croesus’ ambiguous alterity: on the one hand, his respect for Greek gods and sympathy with Greek culture; on the other hand, the gold of the dedications, which points to the Asian splendor of his rule. A historical gloss at the end of the section enhances the monarchical code: the offerings were partly financed from the estate of a political enemy who supported Croesus’ half brother, Pantaleon, in his claims to the kingship and was subsequently tortured with a carding comb and executed (1.92.2– 4). This entirely new story contains the only act of royal mutilation attributed to Croesus: it rectifies the domesticated portrayal of the Lydian king in the preceding narrative and regularizes his membership in the analogical category of absolute rulers that dominates the rest of the Histories. After the offerings of Croesus, Herodotus only chooses to describe the tomb of Alyattes. He advertises it as the one great εργον that can almost stand comparison with those of Egypt and Babylon (1.93.2). By way of this oversized mark of royal power, the discourse finally arrives at the hitherto much neglected community of the Lydians. They have commissioned the monument, and on the summit of the structure, consisting of a base of huge stones surmounted by a mound of earth, five pillars say which parts have been paid for by merchants, artisans, and prostitutes, respectively (1.93.2– 3). The three professions are emblems of Lydian “femininity” (not effeminacy, as in the passage on Cyrus’ reform). Banausic activities are related to women’s work and prostitution is another form of commerce.177 The ability of women to be agents in the place of men—unlike Greek women, the daughters of the Lydians raise their own dowry (by practicing prostitution) and give themselves in marriage (1.93.4)—points in the same direction. So does the emphasis on building (εργα), a feature we find notably in Babylon, where the commerce of women in some form or another is also traditional (1.196.1– 4, 5, 1.199). Most important, however, is the ethnographer’s choice of the 177. See Xen. Lak. Pol. 1.3. For prostitution as one of the indices of a soft culture, see Redfield 1985, 109– 10, quoted earlier in this chapter.

Comparison

105

monument to represent and symbolize the society as a whole. Standing next to another Mermnad landmark, the Gygaean Lake (1.93.5), the sema (tomb/sign) of Alyattes is a sema of the Lydian people and their subjection to the king. “Femininity” in the sociocultural sense, prostitution, and subjection to monarchical rule are all features that differentiate the Lydians from the Greeks, but the narrator insists on cultural similarity (1.94.1). The section that follows, the last of the ethnography, consists mainly of a long historical gloss that finds the roots of affinity in the past. The Lydians were “the first men we know” to coin gold and silver and the first to become shopkeepers (καπηλοι, 1.94.1), and it is said that they have invented various games the Greeks also now play (1.94.2). Their innovations (ε ξευρ- occurs five times) connote intellectual resourcefulness, or sophie, which the Greeks claim as their national heritage.178 Here they also represent a display of hardiness. During a famine in pre-Heraclidean times, the Lydians “looked for remedies,” and “one man devised one thing and another that.” This occasioned the discovery of popular games as a sort of distraction, so that by playing the whole day and not taking any food for one day out of two, they lasted for eighteen years (1.94.3– 4). The mention of the famine leads in turn to the subordinate narrative of the Lydian migration to Tyrrhenia (94.5– 7), reminiscent of familiar stories of later Greek colonizing expeditions, such as those the text reports elsewhere concerning Therans, Phocaeans, and Teans.179 The Lydian king divides the population into two groups and draws lots (as the Therans do), to decide who should stay and who should go; he then places his own son at the head of the latter group (1.94.5). Unlike Alyattes and Croesus, this king stands outside the monarchical model: he rather resembles the archaic Greek basileus in his role as founder.180 Thus, seen as a whole, the Lydian appendix pursues the idea of similarity and difference with the Greeks by tracing in reverse order, through description and embedded historical narrative, a telescoped evolution of 178. Cf. 1.60.3. See Diller 1961, 67– 68. 179. After seven years of drought, the Therans draw by lot a party from themselves, at the rate of one son out of two, and set off to Libya under the leadership of Battus (see 4.151– 53). The Phocaeans and the Teans abandon their homes and sail for the West rather than face Persian subjection (see 1.164– 68). A few generations later, the Athenians apparently consider doing the same (see 8.62.2). See also the Ionians at 1.170 and 9.106. 180. The traditional figure of the founder, as the embodiment of the trials and tribulations of the city and the antithesis of the tyrant (though also with the potential of becoming a tyrant), has been examined by McGlew (1993, 22– 26, 157– 82).

106

Telling Wonders

the Lydian people—from the remote struggles for survival of a small and courageous nation, to their resourceful inventions and economic development through banausic and “feminine” trades (καπηλοι at 1.94.1 corre sponds to καπηλευειν at 1.155.4), to the growth of their monarchy’s wealth and power. In this representation, the Lydians are fundamentally similar to the Greeks, not Hellenized barbarians. Croesus is Hellenized, but his rule also marks the maximum distance between the collectivity of the Lydians and the Greeks. The difference between the two ethnea is in fact a by-product of the Mermnad monarchy’s development. By the time of Alyattes, and even more so with Croesus, Lydian society has become increasingly passive and subjected: it expresses itself through the king, his monuments, his wealth, his personal prestige and connections, and his more or less well-advised policies. In the first section of the Lydian appendix, Croesus is dominant, and the people are ethnographically absent, just as they were politically irrelevant in the preceding historical narrative. In the chapter on Alyattes’ tomb, they place their activity in the service of their king. Among the early Lydians of the third section—who are the closest analogues of the Greeks and historically explain the affinities that still exist between the two peoples—those who left and changed their name to “Tyrrhenians” perhaps avoided this development. Like the Phocaeans and Teans, at any rate, and unlike the rest of the Ionians, they were spared what eventually befell those who had stayed behind.181 As the narrator abruptly concludes, “the Lydians, then, were enslaved by the Persians” (1.94.7). The alleged transformation into “women” under Cyrus will represent the next stage. The insistence on similarity and sameness in Herodotus’ representation of foreign peoples fits in with the analogical thrust of his historical account. A major concern of the apodexis as a whole is to describe the world so that the histories of foreign peoples may be comprehensible in light of their specific cultures. At the same time, however, each model of explanation, each way of being and becoming, is to some extent also applicable to everyone else, including the Greeks. The signals of similarity between different ethnea both conform to the more general principle of the patterned unity of the world and suggest specific ways in which historical processes can reproduce themselves cross-culturally. I am speaking here not of firm historical laws or inevitable cycles but of recognizable 181. Cf. Lombardo 1988, 202 on Herodotus’ Lydians as a negative paradigm for the Ionians.

Comparison

107

models of likely human behavior and of likely consequences. Herodotus’ Persians, Scythians, Lydians, and so on, are each different from the Greeks and every other people. The history of each is to a great extent not repeatable in other cultural settings. Yet specific features shared by different cultures and the sameness of the anthropinon connect all of these in an untidy system of mutual allegories.182 The Sameness of the Scythians The portrayal of the Scythians in the Histories constructs a people so profoundly different from the Greeks that they constitute the virtual embodiment of the other.183 Yet among the ethnea described by Herodotus, the Scythians are a particularly unstable paradigm of alterity: they are alien but also familiar. Their affinity with the Greeks is partially based on their shared experience of finding themselves on the receiving end of Persian aggression and being able to confound the efforts of that vastly superior power.184 But Herodotus’ implicit suggestion that the Greeks can recognize themselves in the Scythians—crude nomads, living in an area of the world where there is little to admire (see 4.46.2)—has a wider scope. It aims at displaying the sameness of what is most distant and foreign. It is part of his overall pursuit of worldwide cross-cultural links. The antecedents of Herodotus’ Scythians belong to an old and conflicted tradition of ethnographic representations of northern pastoralists that goes back to Homer. The idealized “hard primitive” first appears in a passage of the Iliad where Zeus averts his gaze from the painful battle of Trojans and Achaeans and looks into the distance, over “the land of the Thracian riders of horses and of the Mysians, who fight at close quarters, and the noble Mare Milkers, drinkers of milk, and the Abii, most righteous of men.”185 Theoretically warlike, but also just and naturally disinclined to bloodshed (a feature represented by their milk drinking), remote societies suggest relief from the troubles of the more civilized Homeric world of fighting and ships. In his discussion of representations of the Scythians down to his day, Strabo introduces the Homeric passage 182. On the allegorical modes of modern ethnography, see Clifford 1986. 183. See Hartog 1988, passim, especially 11. 184. Hartog (1988, 35– 38) gives a complete catalogue of the points of analogy between Herodotus’ narrative of Darius’ Scythian expedition and that of Xerxes’ Greek expedition. But see Immerwahr 1954, 262– 63. 185. 13.1– 6. See Lovejoy and Boas 1935, 288, citing Riese; Romm 1992, 53. Levy 1981; Marincola 1997b, 4. For “hard” and “soft” primitives according to the definition of Lovejoy and Boas, see “Differentiating from Within” earlier in the present chapter.

108

Telling Wonders

right after reporting a statement by Ephorus that testifies to contradictory ethnologic views in the fourth century b.c. Ephorus says . . . that the ways of life of the Sauromatae and the  and other Scythians are not all alike, for some are harsh [χαλεπους] even eat humans, while others abstain from eating any other living beings. The other writers, he says, talk about their savagery [περ ι   οτητος] της ωµ because they know that the terrible and wonderful are striking; but one should tell the opposite facts and make them into paradigms [παραδει γµατα ποιεισθαι], and Ephorus himself will therefore describe only those who follow most just modes of behavior [δικαιοτα τοις ηθεσι]; for there are some of the Scythian nomads who feed on mare’s milk and surpass all men in justice [τ η  η πα ντων δια␸ ερειν]  δικαιοσυν and are mentioned by the poets . . . (Strabo 7.3.9 C302 ⫽ Ephorus, FGrHist 70 F 42) This passage illuminates the ideological uses of the antiprimitivist and primitivist positions. While the cruel Scythians are savages in antithesis to whom we civilized people define ourselves, the just Scythians are the superior others, paradeigmata who teach us about our own shortcomings.186 We are here in the presence of a moralistic form of ethnographic discourse. It also briefly appears, for example, in Herodotus’ account of the mutually antithetical Androphagoi and Argippaeans, but Herodotus does not normally idealize his primitives, whether hard or soft.187 Ephorus would no doubt rank Herodotus’ Scythians among the savage 186. On the tradition of “inverse ethnocentrism” and “ethnologic satire” in the Greek representation of various peoples on the edges, see Romm 1992, 46– 48, 49– 60 (Ethiopians), 61– 67 (Hyperboreans), 67– 77 (Arimaspians and Scythians). See also Levy 1981, 57– 59. 187. Thus, for the traditionally idealized Hyperboreans (see Romm 1992, 61– 67), Herodotus is not certain of their existence and is only interested in a custom he can compare to another (4.33.5). His Androphagoi are, however, negatively idealized “hard” primitives (4.106). The Argippaeans, or Bald Men, are positively idealized “soft” primitives: they feed exclusively on the milk from their abundant flocks and on the fruit of certain local trees that also provide them with shelter. Their congenital baldness shared by men and women alike points to a utopian equality between the sexes; it also recalls the holiness of the shaven priests in Egypt and is symbolic of nonviolence: “No one among men does them wrong, for they are said to be sacred, and they do not possess war weapons. In the first place, they are the ones who settle disputes among their neighbors, and secondly, if someone who is a fugitive seeks refuge with them, no one does him wrong” (4.23.5). For the connection of hair with belligerence, see Xen. Lak. Pol. 11.3, 13.8; Plut. Lyc. 22; Tac. Germ. 38.4. See also Loraux 1977, 119. Cf. Hdt. 1.82.7 and 7.208.3. For equality between the sexes as an index of justice, see Herodotus’ Issedones (4.26.2).

Comparison

109

and cruel kind. Though they are not cannibals by custom any more than are the Persians or Medes (1.73.5; cf. 1.119.3), they drink human blood, make human sacrifices, and scalp their enemies. The point of Herodotus’ account, however, is objectively to represent the foreignness of the Scythians and at the same time to overcome it by promoting the audience’s discovery of their affinity with them. Herodotus’ work is made easier by the contemporary Greek ambivalence toward the Scythians as it emerges from fifth-century texts. Here the stereotype appears to be more complex than Ephorus’ simple opposition of good and bad Scythians. For the most part, the Scythians of fifthcentury political discourse are intractable xenophobes living at the extremities of the earth, but this very isolationism also makes them autarchic and impregnable.188 In the Eumenides, a passage reflects on what is desirable for the safety of a state by pairing up Scythia and Sparta in a sort of political kinship. When Athena proclaims that a new court of law will be established in Athens on the Areopagos, the hill where the Amazons attempted to establish a rival city (685– 93), she promises that this court will preserve rightful fear within and provide “a bulwark for the land and a means of salvation for the city such as no one among men possesses, not even among the Scythians or in the places of Pelops” (Aesch. Eum. 700– 703). Identification and separation are here operative at the same time. To the extent that Scythia is viewed as a society founded on a rigorous order, it is comparable to Sparta, on the one hand, and to Athens, on the other.189 Like Sparta and Athens, and in antithesis to the wild Amazons of this passage, the Scythians occupy a political space “neither without rule nor ruled by a master.”190 At the same time, the coupling of the land of Pelops with Scythia in the Eumenides points to their difference from Athens. Discipline and dedication to war make both communities, one 188. See Hartog 1988, 12– 13, on representations of Scythian savagery outside of Herodotus. Examples are Aesch. Prometheus Bound 709– 14; Aristoph. Acharnians 702– 3; the Scythian archer in Thesmophoriazusae 1070– 1175, discussed by Long (1984, 138). On the setting of Prometheus Bound, see especially Hall 1989, 113– 15. The Hippocratic treatise Airs, Waters, Places (17– 22) is idiosyncratic in representing the Scythians not as a “hard” culture but as a people made “soft” and sluggish by climate and effeminate by their riding practices. See Thomas 2000, 68– 71. 189. For Scythian paradigmatic justice, cf. Aesch. Sept. 727– 33.   190. Aesch. Eum. 696. Cf. frag. 198 Nauck, αλλ ι ππα κης βροτηρες ευνοµοι Σκυθαι (Strabo 7.301), where ευνοµοι could mean “possessing a just and well-ordered society.” See Levy 1981, 61.

110

Telling Wonders

Greek and one barbarian, invincible; but they will be no match for Athens, a city defended from within by a juridical body of righteous and godfearing citizens armed with ballots (707– 9). A society that is placed under the aegis of deliberation is here contrasted with the sort of order, common to both Scythians and Spartans, that privileges war.191 Herodotus’ representation of the Scythians is similarly based on an interplay between otherness and partial identification with Spartans or Athenians. This thematic seesaw is programmatically reflected in the first two semiautonomous narratives at the beginning of book 4, one (“Scythians blind their slaves,” 4.2) serving as a long ethnographic gloss within the other (“return of the Scythians from Asia,” 4.1.3, 4.3– 4). The latter is announced first by a summary introduction: when the Scythians came back to their land after ruling Media for twenty-eight years, they had to confront “a trouble not inferior to the Median trouble,” namely, opposition from their slaves, whom the Scythian women had married in their absence (4.1.3). The second narrative is attached at this point to the mention of slaves. With a shift to the ethnographic present, it reports how the Scythians, who are milk drinkers and nomadic, blind “whomever they capture” and put them in charge of milking their mares. Though the connection the discourse establishes between the Scythian method of milking, the practice of blinding the slaves, and nomadism is opaque, the mutilation has presumably functional aims.192 Scythian culture is at once comfortable and deprived, living in abundance (4.47.1, 4.53.2– 3, 4.58, 4.59.1) but with few resources. In the absence of a more painstaking sort of labor, the Scythians indiscriminately use whatever they have on hand.193 While pointing out their drinking of milk (a feature of a “good savage”), the text here also first brings to the fore the brutal pragmatism that is the principal mark of the alterity of the Scythians with respect to the Greeks. 191. Cf. Thuc. 2.97.5– 6, which praises the Scythians for their strength in battle but   disparages their ευβουλ ι α and ξυνεσις. For the lethal warlikeness of the Scythians, see also Aesch. Choeph. 161– 63. 192. See Macan 1895, 1, 2; How and Wells 1928, 1:303; Legrand 1946, 4:48; Benardete 1969, 100– 101; Hartog 1988, 18. Mutilation is normally envisioned as an affirmation of despotic power or as a punishment. See chap. 3, n. 54, in the present book. 193. Bones are used as firewood, the ox’s stomach as a cauldron (4.61). Human skin is fashioned into clothes, and hand-skins with the nails attached serve as lids for quivers; enemy scalps become napkins, and skulls become drinking cups (4.64– 65); as hunters, the Scythians do not always make fine distinctions between animal and human quarry (see 4.134.1; cf. 1.73). See Hartog 1988, 40– 44.

Comparison

111

The continuation of the story of the Scythians’ return juxtaposes next to the functional action we have seen (the blinding of the slaves) a symbolic action that offers a rather different perspective on Scythian culture. The sons whom the Scythian women had borne to the slaves tried to block the returning Scythians and confronted them in battle. The Scythians overcame the young men’s resistance only after hearing the following advice from one of their own. What are we doing fellow Scythians! Fighting with our slaves, we are ourselves killed and become fewer; and killing them, we will in the   ι τε κτεινοµενοι future rule over fewer men [αυτο ε λασσονες      γινοµεθα κα ι ε κει νους κτει νοντες ε λασσονων το λοιπον αρξοµεν]. I propose that we set aside spears and bows and that each take instead the whip for his horse and approach them. For so long as they see us with weapons, they will believe themselves to be equal to us   ι ων]; but if they see us and born from equals [ οµοιο ι τε κα ι ε ξ οµο with whips instead of weapons, they will understand that they are our slaves and, recognizing this, will not stand their ground against us. (4.3.3– 4) This episode entails a sudden shift of Hartog’s metaphorical mirror in Herodotus’ representation of the Scythians.194 The barbarians of the previous narrative are given to strange practices and almost unintelligible brutality in the pursuit of their elementary daily living; here they suddenly reflect in a direct way the Greek audience’s ideology of freedom and mastery over the symbolic forms of their status as free men. Freedom as a Scythian value will be a fundamental element in the historical narrative of Darius’ expedition. The Scythian king Idanthyrsus  claims to Darius that he recognizes as masters (νοµι ζω δεσποτας, 4.127.4) only his ancestor Zeus and Hestia, the queen of the Scythians; this parallels Demaratus’ statement to Xerxes that the Spartans have no other master than their law/custom (7.104.4). The Scythians hold the Ionians in contempt on the grounds that by the standards of free men, they 194. See the study and critique of Hartog’s approach and metaphor (Hartog 1988) in Dewald 1990, 218, 220– 21: according to Hartog, “by looking at how a Greek constructed the Other, we also see much more clearly how a Greek understood that which distinguishes the Same: hence the ‘mirror of Herodotus.’” But Dewald cautions: “the Same unexpectedly becomes the Other”; “to extend the governing metaphor that Hartog uses, Herodotus warns us . . . that his mirrors are not bolted on their walls.”

112

Telling Wonders

are the worst and most unmanly of beings, but “if one speaks of them as slaves, they are the most master-loving and nonrunaway of human stock” (4.142). This evaluation anticipates the Greek Artemisia’s discussion later on about good and bad slaves (8.68γ). Like no other people in the Histories besides the Greeks, the Scythians are represented as a people who define themselves in opposition to literal and political slaves. But the story of the Scythians’ return at 4.3– 4 conveys a more specific parallel. We find again the political situation whereby slaves take charge of the state and marry the citizen women in the absence of the men, in Herodotus’ story about the depopulation of Argos after the battle of Sepeia (6.83; cf. 6.77). It belongs to a pattern of historical traditions concerning Greek states like Sparta that have a system of slavery of the helot type.195 In the passage quoted earlier (see especially the underlined phrases), the rhetoric of the anonymous Scythian reveals the pointedly political character of the anecdote for a Greek audience.196 In contrast to their slaves, the free Scythians perceive themselves as “equals and born   from equals” [ οµοιο ι τε κα ι ε ξ ο µοι ων] (4.3.4). In the Histories, οµοιος  alludes to the political system of the Spartan Homoioi on at least three occasions.197 In two out of the three cases, somewhat as in the Scythian episode, the word helps to express the idea that valor in battle is connected with citizen status.198 All three cases allude to the Spartan ideology 195. The pattern is identified and analyzed by Vidal-Naquet (1981, with numerous examples), though he does not mention Herodotus’ Scythian story as an analogue. 196. It is considered a piece of political theorizing by Macan (1895, 1:3). See Finley 1980, 118– 19. Corcella (1993, 230) cites later parallels. 197. In the strictly technical Spartan sense, the word first occurs in Xenophon (Lak. Pol. 10.7; 13.1, 7; etc.), but as Finley (1968, 146) observes, this fact is not very significant for the meaning of the word earlier on. 198. At 7.234.2, Demaratus says to Xerxes that Sparta is a city of about eight thousand  men and that these are all οµοιοι to those who fought at Thermopylae; the rest of the  Lacedaemonians are brave men, though not οµοιοι. At 3.55.1, an interpretive gloss attached to the narrative of the Spartan expedition against Samos states that “if those of the  Lacedaemonians who were there that day had been equal [ οµοιοι ε γι νοντο] to Archias and Lycopes, Samos would have been captured.” Both passages exploit the ordinary sense of the word to refer to the Spartan code. See Shimron 1989, 61; 1979, 132. At 7.136.2, Xerxes replies to Sperthias and Boulis, who offer to expiate the Spartan murder of the Persian  heralds, that he will not be equal ( οµοιος) to the Lacedaemonians, who have overturned basic human laws. This passage (as also 3.55.1) is an ironical reference to the discrepancy between ideology and reality. Herodotus’ Xerxes (who of course does not come from a society of homoioi; see, e.g., 1.134.1), finds the Spartan notion especially amusing; see 7.103.1, where he brings forth a contradiction in the system by a joke on the Spartan kings being worth double. On the problem of inequality of performance and in the political sphere among the Homoioi, see Finley 1968, 147– 49.

Comparison

113

of equality among themselves and their moral and political superiority vis-a-vis ` everyone else. As an exclusive community of equals, the Scythians are implicitly “like” Spartan citizens. The display of whips suggests a method of psychological conditioning of slaves to which Sparta provides the closest parallel;199 the narrator’s assessment that handling their slaves was no less of an ordeal for the Scythians than was fighting the Medes (4.1.3) hides a comparison with Spartan difficulties with the helots after the Persian Wars. A Spartan model is also confirmed by the anonymous Scythian’s mention of the problem of becoming fewer,200 by the avoidance of any reference that may suggest private ownership of the blind slaves,201 and by the narrator’s corroboration later on of the report of a Scythian invasion into present-day Scythia in preference to traditions of autochthony.202 In a later section of the narrative, we also find the statement that beyond the river Gerrhos, and “as far to the east as the ditch dug by the children of the blind men,” are the so-called royal territories and “the best and most numerous Scythians, who consider the other Scythians as their slaves” (4.20.1). The narrative does not provide much help for reconciling the existence of the blind slaves—described in the ethnographic present at 4.2 and remembered here—with these newly mentioned putative slaves of the royal Scythians.203 The blind men are prisoners of war (and therefore possibly non-Scythians), while the others, who probably include the agricultural Scythians (4.54), are in a state of political subjection. Despite the lack of explicit coherence, it is at least clear  that we are supposed to envision a privileged ruling group of αριστοι in the eastern part of the country, and two different types and degrees of servitude, roughly on the tripartite model of Spartiates, perioikoi, and helots.204 While the Spartans are citizen-hoplites, the Scythians are cityless, horse199. On the systematic degradation of Spartan helots, see Myron of Priene, FGrHist 106 F 2 ⫽ Athenaeus 14.657d; Plut. Lyc. 28. See also Garlan 1988, 153– 55. 200. 4.3.3. On Spartan oliganthropia in the fifth century, see Figuera 1986, 165– 81.  201. Contrast, for example, Persian ο ι κεται (see 1.137.1), and cf. 4.72.1 on the absence of slaves “bought with money.” 202. 4.11, cf. 4.5– 10. On the tradition of the Dorian invasion, see Hall 1997, 56– 64. 203. On the historical, geographical, and logical difficulties of Herodotus’ description, see especially Macan 1895, 1:2– 3 and 14; Macan 1895, 2:1– 30; Kothe 1969, 71– 80. 204. Darius’ opponents appear only to include nomadic and free Scythians. See Hartog 1988, 194– 98. For a distinction between royal/free and other Scythians, see also 4.120, 4.59.1, 4.110.2.

114

Telling Wonders

riding nomads; they avoid pitched battles even against invaders of their own territory.205 But somewhat like the Spartans, the Scythians do not allow the enemy to interfere with their customary way of life, never flee, and will fight when necessary.206 The ethnographer, to be sure, must translate the notion of courage into Scythian cultural terms: they reward as an  ηρ  αριστος  αν (brave man) the warrior who has collected the greatest  number of scalps (4.64.1); they demonstrate what “they call ανδραγαθ ι η [valor]” by showing off to their guests cups made of the skulls of enemies killed in battle.207 But by emphasizing the social control that among the Scythians surrounds the warrior’s achievement, Herodotus again enhances the parallel with Sparta. Xenophon reports that at Sparta “anyone would be ashamed to take a coward into his mess or be matched against him in a wrestling match” (Lak. Pol. 9.4). Among Herodotus’ Scythians, those who have not killed any enemies may not partake of the local version of the symposium sponsored annually by the governor of each province: they sit apart from the others, dishonored and ashamed, while those who have killed a great number of enemies are invited to drink double (4.66). Regulation of drinking on the Spartan model here replaces anarchical excess and the habit of drinking straight wine normally attributed to the Scythians and even mentioned by Herodotus’ Spartan sources.208 Scythians and Spartans are also similar in certain aspects of their respective codes of communication. Herodotus’ Scythian narrative devotes special attention to the language of the Scythians, which is first of all peculiar to them and consistent with other aspects of their reduc205. See Hartog 1988, 50– 53. On Greek—especially Spartan hoplitic—arete (valor), see 7.104; Tyrtaeus frags. 12.13– 20 and 11.31– 34 W. In Plato’s Laches, Socrates bridges the gap between Greek commitment “to stay at one’s post and face the enemy and not run away” (190D– E) and the Scythian method of fighting by flight and pursuit, by recalling the Spartan performance at Plataea (191A– C). 206. See 4.127.1– 3. Flory (1987, 103) remarks that Idanthyrsus’ statement that what he does in this war is only “what I have been accustomed to do” parallels Demaratus’ explanation to Xerxes that arranging their hair when they are about to risk their lives is the Spartans’ custom (7.209.3). I discuss the Scythian and Spartan defensive conception of war in chap. 3, “The Evils of War.” 207. 4.65.2. Contrast Plut. Lyc. 22, which says that the Spartans thought “it was shameful to cut to pieces those who had conceded defeat.” 208. 6.84.1– 3. Scythian drinking was notorious: see Anacreon frag. 11b Page (⫽ PMG 356) in Athen. 10.427b and Theog. 825– 30. For Spartan moderation in drinking, see Xen. Lak. Pol. 5; Plato Laws 637a; Critias frag. 6 West (⫽ Athen. 432d– 433b); DK 88 B33 (⫽ Athen. XI 463E). The sources are collected and discussed by Fisher (1988). Cf. Plut. Lyc. 12.

Comparison

115

tionistic culture. Their simplified code is illustrated by the mouse, frog, bird, and five arrows they send to Darius in response to his request of earth and water. This message turns out to mean that since the Persians are not birds, mice, or frogs, they are trapped in Scythia and will be pierced by these arrows (4.131– 32). The flattening out of the syntax is here truly amazing, showing that Scythian communication is as unique as their entire way of life and system of warfare.209 To the extent, however, that their “language” is spare and concrete, it resembles Spartan discourse. The Scythian displays of whips, scalps, and skulls find their parallel in countless visual images the Spartans use to convey the idea of their power.210 It is perhaps not entirely by chance that the description of King Ariantas’ crater, which measures the Scythian population in an approximate “Scythian” way (i.e., by size rather than by number), attracts a comparison with the crater of the royal Spartan Pausanias (4.81.3). The Spartans in Herodotus dislike long speeches, literalize metaphors, and mistrust abstractions.211 The Scythians designate snow with the word for “feathers” because of the two objects’ resemblance (4.31.2), and they use an ancient akinakes as the sacred image of Ares (4.62.2). In both cultures, the use of signs as a device for economical communication goes hand in hand with an anti-intellectualistic attitude, contempt for a certain type of sophie, and a materially simple way of life. The analogy between Spartans and Scythians in the sphere of discourse comes to the surface of the text through the reported judgment of the Scythian sage Anacharsis: he decrees that the Spartans are the only ones   ολους of the Greeks not “busy pursuing all sort of cleverness” [ασχ . . . ε ς  πασαν σο␸ ι ην] and able “to send and receive speech with good sense”    ι τε κα ι δεξασθαι  [σω␸ρονως δουνα λογον] (4.77). Anacharsis’ words make a distinction between sophie (intelligence/cleverness/wisdom) and sophrosune (good sense/wisdom). The first, which here connotes theorization and fancy rhetoric, is rejected by both Spartans and Scythians but is, 209. See West 1988. Contrast metaphorical prophecies of the type “When a mule sits on the throne of Media . . . . ” (Hdt. 1.55), where decoding the message requires that one replace the sign “mule” with the correct referent, but the referent occupies the same synctactical position as the sign. 210. Examples are red cloaks and long hair. See Powell 1988a on Sparta’s use of the visual in communication. 211. See 3.46 (Spartan response to Samian refugees), 7.226.1– 2 (Dieneces), 7.135.3 (Sperthias and Boulis). See the Spartan economic use of words at 9.11.2 (cf. Thuc. 4.40.2). For Laconian brevity, see also 9.90– 91, and cf. Sthenelaidas at Thuc. 1.86.1.

116

Telling Wonders

implicitly, a specialty of Ionians and Athenians. Sophrosune, by contrast, is valued by both Scythians and Spartans and displayed by them through economical speech.212 The narrator identifies this Anacharsis narrative as an invention of the Peloponnesians (4.77). If we believe that Herodotus’ glosses of source are historical, this points again to the existence before his time of a Greek tradition representing the Scythians as in some way analogous to the Greeks and in particular to the Spartans.213 In relation to speech and other spheres of culture, Herodotus exploits existing traditions for the purpose of establishing his own implicit analogy between the two cultures. Starting from the early episode of the whips, he does so in a context not of positive or negative idealization but of an ostensibly objective and scientific ethnography. The analogy serves therefore neither to praise nor to disparage; rather, it forms a part of the broader message that trends of similarities invariably link ethnic groups with widely discrepant levels of culture and differing customs. The Scythians are not at any rate connected only to Sparta. The multiple traditions concerning the origins of the Scythians relate them to various other ethnea of the world.214 Hartog has thrown light on the implicit analogy Herodotus’ narrative establishes between Scythians and Athenians, especially owing to the nonhoplitic role both nations played in their defensive war against the Persians and to the nonhoplitic way with which the Athenians of Herodotus’ time chose to respond to the Peloponnesian invasions.215 Athenian strategy in the earlier and more recent past has come to define the Athenians culturally, just as the Scythians’ strategy in war, Idanthyrsus says (4.127.1), coincides with their custom in peace. A major interpretive gloss that connects Scythian strategy and nomadism even enhances the implicit analogy between Scythian 212. See Georges 1994, 146. Cf. the Spartan Archidamus at Thuc. 1.79.2, 1.84.3, 1.84.2. 213. On Anacharsis’ and the Scythians’ mistrust of speech, see also Pherecydes, FGrHist 3 F 174; Hesiod frag. 150 MW; Levy 1981, 60. For the Anacharsis legend and apophthegmata, see Kindstrand 1981. 214. See 4.5– 12. On the self-contradictory quality of these multiple accounts, see Hartog 1988, 19– 30. Greek Hellenocentric traditions that make barbarian peoples derive from characters of Greek mythology (see Bickerman 1952; Georges 1994, 1– 9) are again utilized by Herodotus for the purpose of relating barbarians and Greeks. See Corcella 1993, 232; Vandiver 1991, 169– 80. 215. Hartog 1988, 39– 40, 49– 51. Cf. especially 4.122 with 8.41, 4.127.2 and 4.120.1 with Thuc. 1.143.5.

Comparison

117

and Athenians by attributing to the Scythians sophie, the very quality they elsewhere despise.216 The Euxine Sea, where Darius’ expedition was directed, of all the   regions contains the most ignorant peoples [εθνεα αµαθ εστατα]. For within the Pontic region, we cannot mention any people as  excelling in cleverness/wisdom [σο␸ ι ης] or any learned man [λογιον  ανδρα], except for the Scythian people and Anacharsis. In the one matter that is of the greatest importance for man, the Scythian race  made the cleverest/wisest discovery of all that we know [σο␸ ωτατα       πα ντων ε ξευρηται των ηµεις ιδµεν]. For the rest, I do not admire it, but they have devised this one most important thing, that no one who goes against them can escape and it is not possible to catch them if they do not want to be found. For how could men who have no constructed houses or walls but carry their homes with them and are all archers, who live not from agriculture but from livestock, and who have their homes on carts not be invincible and impossible to deal with in battle? (4.46.1– 3). The sage Anacharsis is not here “Laconic” and sophron, as in the   Peloponnesian story considered earlier. He is rather a λογιος ανηρ. Through him, the Scythians partake of the sort of sophie consisting of knowledge and of the ethical wisdom that derives from it, which are elsewhere displayed by the Athenian Solon and possessed in the highest degree by the Egyptians.217 The Scythian people as a whole, however, possess at least sophie of a cunning sort, which is also considered “Athenian.”218 The Scythians are nomadic and primitive, and these two ethnographic characteristics that most identify them as other with respect to the Greeks in general also create the preconditions for their sameness. So far as they are nomadic, with the peculiar strategy that their way of life entails, they resemble the Athenians, autochthonous city dwellers. They resemble the Spartans in their social war ethics and spare way of life, because they are primitive. At the same time, they are clearly neither Spartans nor Athenians nor Greeks, just as they are not Egyptian or even, 216. For the stereotype of the lack of intelligence of northern people, see Arist. Pol. 7.6.1 (1327b23– 25); Hall 1989, 122.  217. See 1.29– 33 (Solon), 2.77.1 (the Egyptians as λογι ωτατοι). 218. I am following here the distinction made by Dewald (1985, 52– 55) between two kinds of knowledge that Herodotus represents.

118

Telling Wonders

as Herodotus is at pains to specify, Budini or Geloni. They are entirely themselves, that is, Scythians. Identification with The Other: Anacharsis and Scyles An ethnographic-historical sequence tells about how first Anacharsis and later Scyles were killed by the Scythians for adopting foreign customs. This set of stories illustrates the role of implicit analogy in forcing the audience to reflect themselves in an alien people.219 Herodotus attributes one trait to all ethnea of the world: subservience to the constraints of their own culture. Formulated in abstract terms in the far-ranging interpretive gloss at 3.38, Herodotus’ theory of universal cultural chauvinism is borne out again and again by specific cases. Thus, a gloss of interpretation discloses the meaning of the stories of Anacharsis and Scyles by  introducing them as evidence (δι εδεξαν) that also [the Scythians] utterly avoid following foreign customs, both those of other peoples and most especially those of the Greeks.
    [ξενικοι σι δ ε νοµαι οισι κα ι ουτοι α ι νως χρ ασθαι ␸ευγουσι, µητε      τεων αλλων, Ελληνικοι σι δ ε κα ι ηκιστα.] (4.76.1) Adverbial κα ι (amounting to a gloss of similarity) perhaps refers first and foremost to the Egyptians. They are an implicit term of comparison throughout the Scythian narrative, including in the immediately preceding passage on Scythian σο␸ ι η.220 Unlike the chauvinism of the Egyptians, however, the Scythians’ dislike of foreign nomoi manifests itself in the summary violence that characterizes their culture as a whole. The two narratives that show this ferocious protectionism differ, however, in one important respect: the story of Anacharsis promotes identification with the victim; that of Scyles, with the Scythians. The Scythian sage Anacharsis embodies Spartan sophrosune in the subsidiary Peloponnesian narrative already considered (4.77). In the story about his death, he is connected with Athenian sophie (4.76). Here, like the Athenian Solon, he travels all over the world for the purpose of sightseeing 219. The set of narratives is discussed by Hartog (1988, 62– 84). 220. See Benardete 1969, 99. For Egyptian chauvinism, see the totalizing statements at 2.79.1 and 2.91.1, as well as specific evidence, e.g., at 2.41.3. On Scythia and Egypt, see n. 91 and corresponding text earlier in the present chapter.

Comparison

119

and acquires (or displays?) much wisdom.221 On his way back to the 222 he   “haunts,” or “customs,” of the Scythians (ε ς ηθεα τα Σκυθεων), lands at Cyzicus, where he makes a vow that if he returns home safe, he will celebrate the Mother of the Gods as the Cyzicenes were doing. He does so in the relative privacy of the Hylaea, but the Scythians spy on him, and their king kills him with his bow (4.76.2– 5). The story ends with a unique explanatory gloss self-referentially marked by historie: “As I heard from Tymnes, a deputy of king Ariapeithes, Anacharsis was the uncle of Idanthyrsus, king of the Scythians, and the son of Gnurus, son of Lycus, son of Spargapeithes.” From this genealogy, which the source Tymnes has provided as if oblivious to its $ implications, the narrator draws his own inference: “If, then [ων], Anacharsis belonged to this family, let him know [ιστω] that he was killed by his brother. For Idanthyrsus was the son of Saulius, and it was Saulius who killed Anacharsis” (4.76.6). The pathetic appeal to the dead Anacharsis in the third-person imperative constitutes the closest thing we find in Herodotus to the Homeric apostrophe to a character in the second-person singular.223 Without making recourse to explicit evaluation, the narrator underlines for his audience the horrible fate of a wise and pious man, caused by the ferocious intransigence of a savage people. We recall the typical Scythian of the ethnography, showing off to his guests the skulls of family members with whom he has feuded (4.65.2).  In the structurally similar story of Scyles that follows (παραπλησια, 4.78.1), a shift in perspective renders Scythian protectionism less alien from a Greek point of view. Unlike Anacharsis, Greek-raised Scyles does  ουδαµ   ηρ  εσκετο  not like the Scythian way of life (διαι τ% η µ εν ως  Σκυθικ% η). His Hellenization, which he pursues during surreptitious visits to the Greek city of Olbia, involves different areas of custom, both secular and sacred (4.78.4). In the sphere of the sacred, it culminates in Scyles’ desire to become initiated to the rites of Dionysus. But rather than promoting sympathy for Scyles’ piety, here the text sends negative signals.  µενος καταυτ  θεωρησας    ην  σο␸ ι ην πολλην  (4.76.2), 221. In γην πολλ ην κα ι αποδεξα  the second participle could derive either from αποδε ι κνυµαι, “display,” or from the far less   frequent αποδ εκοµαι, “receive.” The ambiguity is enhanced by the implicit analogy with Solon, whose sophie derives from his “sight-seeing” or (θεωρ ι η, 1.29.1, 1.30.2), but who is also represented as displaying his sophie to Croesus. The same is true for the narrator,   οδεξις) whose work is a display (απ of historie conducted all over the world. 222. See Hartog 1988, 65; cf. 4.80.1. 223. See, e.g., Il. 1.146. See also de Jong 1987, 13, 60.

120

Telling Wonders

At the metannarrative level, a gloss of anticipation of doom directs the audience to interpret Scyles’ initiation to Dionysiac religion as the precipi tating event, or προ␸ασις, in a career already bent on self-destruction, “since he was bound to incur a bad end.”224 The narrative, for its part, reports a sign of divine disapproval: while Scyles is preparing to undergo the ritual, “the god” strikes Scyles’ house in Olbia with a thunderbolt and sets it on fire, a warning that Scyles ignores (4.79.1– 2). In the tradition of resistance myths, which will eventually produce the alien and threatening Stranger of Euripides’ Bacchae, Dionysus comes to a Greek city from abroad, displays his power to those who resist him, and exacts worship from the entire population.225 In the story of Scyles, a foreigner embraces Greek Dionysus, and the divine opposes resistance. Here the author of the portent is not the god whom the Scythians refuse to recognize but a less culturally determined entity.226 Almost as a divine representative of the “Custom king of all” of 3.38.4, this power objects to an individual’s asocial adoption of a religion not his own by nomos and repulsive to his people as a whole. The Scythians reject Dionysus for reasons that are strikingly similar to those of the god’s opponents in the Greek resistance myths. A cognitive statement added as an explanatory gloss to the narrative reports, “the  to go find [ε ξευρι σκειν] Scythians say it is not reasonable/natural [ο ι κος] a god like this one, who drives people mad.”227 Described in these terms, the Dionysiac cult appears as un-Greek—or, as Herodotus himself else where acknowledges, not “consistent” (οµοτροπον) with the rest of 228 Greek culture —as it is un-Scythian. The same thing could indeed be said about the cult of the Mother of the Gods, which Anacharsis adopts in the preceding story.229 In the Scyles narrative, however, a new empha γενεσθαι.  224. 4.79.1: εδεε ο ι κακως For this type of interpretive intervention, see chap. 1, n. 64, in the present book. 225. See Guthrie 1950, 167– 70. In Herodotus’ version of one of the resistance myths (9.34), Dionysiac madness causes a social and political crisis in Argos. 226. It is Zeus according to most commentators. See Linforth 1928, 219; Hartog 1988, 73. 227. 4.79.3. On cognitive statements, see in the present book, chap. 1, n. 40; chap. 3, “Cognitive Statements and Polemical Negations.” On the negative sense of ε ξευρι σκειν, see n. 35 and corresponding text in the present chapter. 228. 2.49.2. For Dionysus in the Bacchae as a symbol of “the blurring of distinctions between Greek and barbarian,” see Segal 1982, 124. 229. For the parallel roles of these two Greek divinities in the Anacharsis and Scyles narratives, and for their equal ambivalence in the Greek perception, see Hartog 1988, 70, 72, 74– 83; Hall 1989, 153– 54.

Comparison

121

sis appears on the irrational aspects of the divinity in question,230 Scyles lacks a good motive for his initiation and disregards a divine sign, and the Scythian criticism of Dionysus sounds reasonable even from a Greek point of view. All these factors force the listeners to blame Scyles as they would not have blamed Anacharsis. Like Anacharsis, Scyles is finally decapitated by his own brother (4.80.4– 5). Here, however, the narrative delays the last act of the drama by a series of functions focalized through the Scythians as an ethnos and body politic. It represents the humiliating mockery to which they are subjected by a foreigner (4.79.4), the grief they experience in the face of  µεγαλην ε ποιησαντο,  their own king’s display of madness (συµ␸ορην 4.79.5), their deposition of Scyles and appointment of a new king, and the negotiations with a foreign power for the extradition of the criminal (4.80.1– 4). By the time the characteristic ferocity of the Scythians even against family members is fully brought back to the fore with Scyles’ decapitation, the Scythians have emerged almost as a civilized and earnest  community, concerned with preserving (περιστελλειν in the conclusion at 4.80.5) their order and integrity. The situation recalls the alarm the Spartans experienced at the adoption of foreign ways and erratic behavior by one of their own who was almost a king. Herodotus has sufficiently “turned the mirror” so that the Greeks of the audience might recognize themselves in the Scythians. The Pausanias model is not a far-fetched subtext to this story.231 The secular and more strictly “civic” side of Scyles’ adoption of foreign nomoi is represented by Scyles’ change of dress, building of a house in Olbia, and marriage to a Greek woman (4.78.4– 5). These cultural shifts go hand in hand with a separation of Scyles from his people, which is also, since Scyles is the king of the Scythians, a stepping down from his royal position. Thus, every time Scyles leads the army to Olbia, he leaves it outside the city, and after going inside the wall and closing the doors, he walks about the marketplace dressed like an ordinary Greek, “not accompanied by bodyguards or anyone else” (4.78.4). But though Scyles intermittently declines to be a king at home and becomes a private citizen abroad, his behavior fits in the symbolic pattern of monarchy we have described in our discussion of analogy in the history. The verb referring to his irrational desire to become initiated into 230. This is noticed by Hartog (1988, 75 n. 47). 231. See “The Seers of Plataea” earlier in this chapter.

122

Telling Wonders

 Dionysiac religion, ε πεθυµησε, is a term of the monarchical code.232 In response to Scyles’ behavior in the sphere of religion, the divine thunderbolt incinerates precisely the house in Olbia “that I have mentioned shortly before this,” as the narrator recalls to reemphasize the main symbol of Scyles’ civic violation. Since it was “a big and rich house, around which stood sphinxes and griffins made of white stone” (4.79.2), this supposed index of Greekness would signal to Herodotus’ audience a northern variation on the theme of Oriental extravagance.233 As Artabanus will say to Xerxes, it is against the biggest houses that the god strikes with his thunderbolt (7.10ε). The transition from nomadic king to Greek polites is therefore made to resemble a climb to despotic rule in the monarchical pattern of the history, with the same sort of metaphorical paradox that equates Tisamenus’ acquisition of Spartan citizenship with an acquisition of the kingship.234 This time, like Pausanias, who would roam in Byzantium in Persian dress and make plans to marry a Persian woman, and like Deioces, enclosed in his new royal palace, Scyles comes to embody the alienation of the individual from the community, an alienation interpreted as some form of self-exalting lapse into barbarism. Through symbols and vertical analogy, the monarchical pattern can even be stretched to apply to a case like this one, where the literal kingship actually represents the traditional good order of society, while the potentially disruptive element is an individual who excludes himself from this institutional status. Scyles, moreover, becomes a foreigner to the Scythians by becoming Greek, yet the way in which he does so resembles a barbarization in the Greek sense. Placed in the middle of a logos where the representation of a profoundly alien people is carefully designed to show the Greeks intermittent glimpses of their various selves, the Anacharsis/Scyles sequence ends up demonstrating that what from the Greek point of view represents the ignorant rejection of their own civilized customs on the part of the barbarians is really analogous to the Greek contempt for barbarian practices that are repulsive to them. As a narrative of past events concerning a foreign people’s customs  (to desire) in Herodotus, two refer to 232. 4.79.1. Of twenty-five cases ε πιθυµεω sexual lust and the instinct of animals, respectively. One occurs in Artabanus’ gnomic saying that it is wrong to desire many things (7.18.2). Of the remaining twenty-two occurrences,  qualifies Marfifteen describe monarchic or aggressive desires. At 7.6.1, ε πιθυµητης donious’ imperialistic ambitions. 233. See Macan 1895, 1:53; Hartog 1988, 73. On the austerity of Greek private houses in the fifth century, see Morris 1998, 67– 75. 234. See “The Monarchical Model in Sparta” earlier in this chapter.

Comparison

123

and institutions, moreover, the Anacharsis/Scyles unit also reveals the congruence of Herodotus’ analogical interpretation of the world in the sphere of history, on the one hand, and of ethnography, on the other. Synchronically, the story of Scyles conveys the attachment of any society to its own cultural values, whatever these may be, as well as the similarity of some of those values among Scythians and Greeks. Diachronically, through the monarchical model, it shows the danger of social and cultural change in a familiar form: in one way or another, such a change has overtaken the Lydians and the Persians and threatens the Greeks as well. Similarities and overlaps counterbalance the synchronic diversity of humankind, so that we are able to discern similarities and overlaps among historical processes that unfold at different times and in different settings of the world. We can make the experience of others our own because, in some respects at least, the others are same. The Other Is Same: Making Peace with the Amazons That the other is same constitutes part of the underlying message of another ethnographic-historical narrative in the Scythian logos, the story of the birth of the Sauromatian nation from Scythians and Amazons.235 Since the Greek concepts of alterity and self in the fifth century are closely related to a political ideology of power, we should not be surprised to find that Herodotus’ narrative combines the reassessment of such views with a recipe for correct foreign-policy relations. The exemplary behavior of the protagonists, who are strictly collective and non-Greek, suggests the prescriptive character of the Amazon story in Herodotus, in contrast with the predominantly epideictic function of the Anacharsis/Scyles sequence. The Amazons are hardly a politically neutral subject. Unlike all the other peoples mentioned in the Histories, they are an extinct race, and the text reports no logos, reliable or unreliable, that assigns to them a place in Herodotus’ map of the present-day oikoumene.236 They are characters of Greek traditions about the remote 235. 4.110– 17. The most important discussions of the episode are by Dewald (1981) and Hartog (1988, 217– 24). See also Flory 1987, 108– 13. According to Cole (1967, 143– 45), the story derives from a fifth-century source that bears connection with the source of the social history in Polyb. 6. Such an origin is well suited to the different but nevertheless theoretical use Herodotus makes of the anecdote. 236. See especially Strabo 11.5.4. Cf. Diod. 17.77.1– 3. On the extinction of the Amazons, see Lysias 2.6.

124

Telling Wonders

past. As far as we are able to determine, they play only one role, that of being repeatedly defeated, subdued, or captured by various heroes, like other bestial or lawless opponents of the mythical ancestors of the present-day Greeks.237 In the political discourse of the fifth century, the story of the Amazons becomes a prominent mythical model for the affirmation of Greek superiority vis-a-vis ` the barbarians. Eventually Athens appropriates it as a charter for the justification of its imperialistic policy after the Persian Wars.238 The visual art of the period depicts the Amazons side by side with Centaurs, Giants, Trojans, and, at least on one occasion, the Persians and Greek enemies of recent history, to represent the forces of disorder confounded by the representatives of civilization—Lapiths, Greek heroes, Olympic divinities, or modern Greek hoplites, as the case may be.239 The ideological message encoded in the combination of these unrelated struggles on a single monument is based on the well-established polarities and analogies between Greek and Barbarian, male and female, human and animal.240 After the Persian Wars, the myth most prominently includes the Amazon invasion of Attica for the purpose of conquest, an invention that seems to have supplanted the preexisting tradition of heroic expeditions to their part of the world, to serve as the antecedent and analogue of 237. For the Amazons versus Bellerophon, see Il. 6.152– 206; Pind. Ol. 13.63– 92. For the Amazons versus Heracles, see Pind. Nem. 3.36– 39; Pindar fr. 172 SM; Hellanicus, FGrHist 4 F 106; Eur. HF 408– 18; Eur. Ion 1144– 45; Apollod. 2.5.9. For the Amazons versus Theseus and Heracles, see Eur. Heracl. 215– 17. For the Amazons versus Theseus, see Pherec., FGrHist 3 F 15, F 151 and F 152; Pindar fr. 176 SM; Eur. Hipp. 10, 305– 9; Plut. Thes. 26– 28. See Gantz 1993. 238. See Tyrrell 1984; Loraux 1986, 147– 48; Tyrrell and Brown 1991, 159– 215, especially 198– 200; DuBois 1982; Castriota 1992, 43– 58; Holscher ¨ 1998, 167. 239. Attic Amazonomachies appeared paired with a Centauromachy in the paintings of the Theseion at Athens; in the Stoa Poikile, juxtaposed to depictions of captured Troy, the battle of Marathon, and the battle of Oinoe against the Spartans (see Paus. 1.15, 1.17; Aristoph. Lys. 672– 80); on the western metopes of the Parthenon (a Centauromachy, a Gigantomachy, and an Iliupersis are on the other three sides); and on the outer surface of the shield of Pheidias’ Athena Parthenos, with a Gigantomachy on the inner surface (see Plut. Per. 31.4). See especially Castriota 1992 (33– 63, 76– 89, 134– 83) and the iconographic study by von Bothmer (1957). Amazonomachies also appeared on the Athenian treasury at Delphi (juxtaposed to other exploits of Heracles and Theseus; see DuBois 1982, 57– 71), on the temple of Apollo Daphnephoros at Eretria, on the temple of Hephaestus in Athens. The temple of Apollo at Bassae in Arcadia, built probably in the last quarter of the fifth century by the architect of the Parthenon, Ictinus, testifies to the popularity of the myth outside of Athens. See DuBois 1982, 64– 66. 240. See Thales (or Socrates) in D.L. 1.33, quoted by DuBois (1982, 4– 6).

Comparison

125

the recent barbarian aggression.241 The passage from the Eumenides that compares Scythians and different groups of Greeks also mentions the Amazons; it shows how they could be invoked when speaking of the dangers that threaten the polis from abroad and of the hubris of her opponents.242 The Amazons are first of all presented as antithetical to the pious and law-abiding Athenians, whose new court will sit precisely on the Areopagos, which the Amazons once occupied. They are also contrasted to the mutually analogous Spartans and Scythians, both of whom rely on war, rather than on political institutions, for their defense, but who nevertheless, unlike the Amazons, possess civic order. Herodotus’ Sauromatian logos does not contradict the notorious “historical” events featured in the myth, but it radically revises its ethical slant. Greek authors of Herodotus’ time unanimously represent the Amazons as the fulfillment of female nature out of control: because they were aberrant and wild, driven by lust of domination, and posed a threat to the civilized world, one needed to fight and conquer them.243 Herodotus, by contrast, portrays the Amazons as a people who possessed certain peculiarities but otherwise were not all that alien from other ethnea or abnormal with respect to the moral sense of the rest of humankind, Greeks 241. See Castriota 1992, 46– 47. In Aesch. Suppl. 234– 37, 277– 90, the Amazons represent a generalized type of aggressive barbarian women hostile to marriage. Aristoph. Lys. 672– 80, where the chorus compares the rebellious Greek women to both the Persian ally Artemisia and the Amazons as depicted by Micon, confirms the public’s interpretation of the Amazon invasion on their city’s monuments as the analogue of the Persian invasion, as well as conveying the idea that the Amazons embody the threat of the female. For the gender-related aspect of the myth, see especially Tyrrell 1984, 22, 113– 28. In fourthcentury oratory, the Athenian defeat of the Amazons in Attica is a conventional topos of the list of glorious Athenian achievements that culminates in their historical defeat of the Persians. See especially Lysias Epitaph. 2.4.– 26; Isoc. Paneg. 4.68– 70; Demos. Epitaph. 60.4– 8. The topos goes certainly back at least to the time when Herodotus was composing his work, as is demonstrated by its occurrence in his version of the Athenian speech at Plataea (9.27.4). On the tradition of the epitaphios in Athens, see Loraux 1986; Tyrrell and Brown 1991, 189– 215. 242. Aesch. Eum. 681– 706. See especially 685– 93 (cited earlier in this chapter), where the Amazons are the archetypal enemy of the polis. In Aeschylus’ Oresteia, the Amazons are also models for the man-killer Clytemnestra (Agam. 1231– 32; Eum. 625– 28). See Zeitlin 1984, 160– 70, especially 163. 243. Hellanicus mentions that the Amazons removed their right breasts, and he calls them “a golden-shielded, silver-axed, female, male-loving, male-infant-killing host” (FGrHist 323a F16 and F17); cf. Hippocr. On Joints 53 [see Thomas 2000, 61– 62]). Other details about Amazonian society do not appear in fifth-century sources and may be later elaborations.

126

Telling Wonders

included. This representation is consistent with Herodotus’ approach to foreign peoples throughout the ethnographies, but as far as our evidence for the Amazons goes, it is unique in literature.244 In the logos reported by Herodotus, a group of Amazons have survived the defeat at the Thermodon River (one of the traditional events in the myth) and are carried off by the Greeks as prisoners. They prove true to their Scythian name “Mankillers” (4.110.1) by dispatching their captors on the ships. Since they find themselves without pilots and are inexperienced in navigation, they drift about until they land on the shores of the Maeotis Lake.245 Here they make their way toward the inhabited area, take possession of some horses, and start raiding the territory (4.110.1– 2). This accidental but aggressive arrival of an Amazon contingent in Scythia seems designed to replay in reduced and modified form the story of the Amazons’ invasion of Attica, which is not mentioned. Replacing the Greeks, the local inhabitants naturally undertake to defend their land. The theme of fighting goes hand in hand with the question of difference. In wonder at the language, dress, and ethnic identity of the Amazons, the  Scythians cannot “put the matter together” [συµβαλεσθαι το πρηγµα] (4.111.1). They cannot, that is, do what the narrator of the Histories frequently does in the course of his ethnographic and geographic research: find grounds for comparing a new phenomenon with something already experienced, to conjecture about its nature. The Scythians perceive the Amazons as entirely different from themselves, except for thinking that the Amazons are young men. Eventually, when they realize from the bodies of the Amazons dead in the battle that they are really women, the Scythians discover a difference that suggests complementarity rather than conflict. They stop the fighting immediately and send a group of young Scythian men, wishing to have children from the Amazons. The differences between the two ethnic groups that now confront one another are considerably reduced. Both are detachments from their respective societies, equal in number (4.111.2) and occupying the same marginal space in the wild.246 They resemble one another in appearance, 244. In art, it is perhaps paralleled by the beautiful Amazons from the sanctuary of Artemis at Ephesos. These, according to Holscher ¨ (1998, 173), “represent an act of selfassertion against overpowering Athens.” 245. These Amazons’ inexperience of ships and navigation points to their idealized primitive status. See Romm 1992, 74– 75. Herodotus’ other Amazon figure, Artemisia, is the opposite (see 7.99.3). See chap. 4, “Vertical Analogy.” 246. On the ephebic character of Scythian youths and Amazons, see Hartog 1988, 217, 219– 20 (cf. 54– 55); Tyrrell 1984, 64– 87.

Comparison

127

since the Amazons look like young men and these particular Scythians are   πρωτην   young men (ανδρας τ ην η λικι ην εχοντας, 4.110.1; τους  νεωτατους, 4.111.2). The boys have, moreover, received instruction “to do what [the Amazons] do”: withdraw when they pursue and draw near when they stop, thereby avoiding all hostile confrontation (4.111.2). Just like the Amazons, the young Scythians have no material possessions ex εζωον τ ην  cept their horses and weapons: both lead “the same life” [ζοην  ην]  of hunting and plundering (4.112). The assimilation of the two αυτ groups parallels their physical rapprochement.247 Their camps come closer and closer each day, and at one point a meeting ensues between one of the Scythians and one of the Amazons. Other individual encounters follow (4.112, 4.113.1– 3). Eventually all the young Scythians pair up with the Amazons, and the two groups, “having joined their camps, permanently live together, each man having as wife the woman to whom he had originally joined himself” (4.113.3– 4.114.1). Fusion and the acceptance of complementarity entails further assimilation and compromise on both sides. The men do not become women and the women men, as some scholars have maintained.248 Reversal is to some extent inherent to the conception of a masculine woman, but within the boundaries of this idea, Herodotus radically modifies the system of polarities in the myth. The Amazons and the young Scythians achieve a society without the inequalities of conventional marriage, since both groups play the male role in identical fashion, while the necessary (though on principle undesirable) female functions are distributed between them across gender lines. The Amazons first of all consent to femininity in the sphere of sexual relations. The words by which the narrator designates intercourse unmistakably connote the submission of the passive partner, in conformity to the asymmetrical Greek conception of the sexual union.249 The Amazon of the original one-on-one encounter proved willing to become a little less Amazonic and culturally male when she “allowed the young Scythian to  have his way with her [περιειδε χρησασθαι]” (4.113.1). In the final stage of the fusion of the two ethnea into one, “the remaining young men 247. The action at 4.111– 15 is marked by alternate verbs of separation and approaching. 248. See DuBois 1982, 36; Tyrrell 1984, 42; Brown and Tyrrell 1985; Hartog 1988, 216– 24. All the scholars here cited emphasize the pattern of reversal, though Hartog realizes the difficulty of applying it to this episode. 249. See Halperin 1990, 29– 36.

128

Telling Wonders

 tamed [ε κτιλ ωσαντο] the remaining Amazons” (4.113.3). In this particular area of behavior, there is really no alternative to the female’s assumption of a role that Herodotus, like his Greek contemporaries, interprets as subordinate.250 Strabo’s Amazons will react to this necessity by resorting only to momentary and furtive unions for the sake of reproduction; those of Diodorus compensate by keeping their mates in an inferior—even crippled—state in all other respects.251 Herodotus’ Amazons find a less radical solution. In the tradition of heroic myths, marriage to an Amazon means primarily conquering her in war, subjecting her in a social as well as in a sexual sense, and taking away her Amazon identity by integrating her into a patriarchal order. Thus, when Heracles kills the Amazon queen for the sake of her belt, this symbolically prefigures Theseus’ abduction to Greece of the Amazon Antioche or Hippolyte, whom, according to the earlier versions of the myth, he has raped.252 In Herodotus’ story, the Amazons’ concession to the physiology of sexual intercourse leads to a stable union. Their subjection, however, is both voluntary and limited to the sexual sphere. It is compensated by the assumption of a share of the social female role on the part of the men, who agree to leave their parental homes, bring the equivalent of a dowry, and be monogamous (4.114.1– 4.115.1). Both groups equally assume the masculine social role, just as had been the case before the fusion: “the women of the Sauromatae follow the old way of life [διαι τ% η]; they regularly go to the hunt, both in the company of the men and separately from them, and also go to war and wear the same dress as the men” (4.116.2). In the Scyles narrative, the national dress is the symptom of a more profound and insidious cultural change. At the beginning of this story, it represents an external and deceptive sign of alterity (4.111.1). The eventual assimilation of ideologically unproblematic features, such as dress and language (4.113.2, 117), is here made 250. See Rosellini and Sa¨ıd 1978, 999– 1000. Brown and Tyrrell (1985), who want to interpret the story as representing the triumph of the female over the male, consider the use  of the word ε κτιλ ωσαντο as a contradiction. 251. Strabo 11.5.1; Diod. 2.45.1– 3, 3.53.1– 2. 252. Eur. Hipp. 305– 9. Cf. the rape depicted on the pediment of the temple of Apollo Daphnephoros in Eretria (ca. 510 b.c.), of which the scene on the amphora of Myson is probably an imitation. See Tyrrell and Brown 1991, 166– 67. Theseus’ capture of the Amazon is mentioned by Plutarch (Thes. 26, on the authority of Pherecydes and others; cf. Pherecydes, FGrHist 3 F 151 and F 152) and by Pindar (fr. 175). The Amazon’s zoster carried off by Heracles in his ninth labor (Apollod. Bibl. 2.5.9) is a war belt but has ambiguous sexual connotations.

Comparison

129

possible by the fact that in a more fundamental sense, the ethnos that appeared to be alien was actually both similar and complementary. Thus, the fusion of the two peoples, based on their resemblance, preserves their mutual equality of status and their integrity. Among the Scythians, no warrior will partake of the distribution of spoils if he does not bring a head to the king (4.64.1), and he is excluded from the annual symposium if he has killed no enemies (4.66). An analogous nomos forbids a girl of the Sauromatae to marry before killing an enemy (4.117). This implicit parallelism means that for the Sauromatian adult women, not only is marriage conceived as a privilege, but their social prestige and full integration into the society is linked to their contribution as warriors, not wives.253 The report adds that some of them die unmarried in old age because they have not been able to fulfill the nomos; this testifies to the determination with which the women of the Sauromatae guard their Amazon identity as much as possible in a society founded on marriage. In Greek society, marriage, for a woman, not only is the counterpart of war but also excludes war. The Amazons in Greek tradition make war and reject marriage. The Amazons of Herodotus’ Sauromatian logos have accepted marriage as a carefully circumscribed change in their customs. Finally, among the Sauromatian female descendants of the Amazons, marriage remains secondary with respect to the Amazon activity of war. If the cultural differences between Amazons and Scythian unmarried men prove ultimately to be imaginary or unimportant, the gulf separating the Amazons from Scythian women (and therefore also from the Scythian men who are married to them) is real and insurmountable. Because they are better ethnologic observers than are the Scythian youths, the Amazons realize this clearly. When their new husbands propose that they all stop leading “this sort of life” in the wild and go and live with the multitude of Scythians, where the young men have families and property, the Amazons describe their difference in a series of negations: “We could never live with your women, because we do not have the same customs   α νοµαια]. [τα αυτ We fight with our bow and spears and did not learn women’s work. Your women do none of the things we have mentioned but rather do women’s work sitting inside their carts and do not go out to 253. The Hippocratic Airs, Waters, Places (17), describes a similar initiation for Sauromatian women as Herodotus does, but with the crucial difference that marriage puts an end to their war activity.

130

Telling Wonders

 the hunt or anywhere else. We could not get along [συν␸ ερεσθαι] with them” (4.114.3). Aside from the fact that they live in wagons instead of houses, the Scythian society of families is here described in Greek terms—women indoors and men outdoors (cf. Xen. Oec. 7.23). The Amazons differ from the Scythians in this one important respect. The articulate and civilized character of the Amazons’ deliberations with the Scythian young men, the lucidity of the Amazons’ reasoning, and their appeal to conventional  and shared notions of justice (see δικαιοτατοι at 4.114.4) are designed to revise the traditional notion that the Amazons are wild women and the generalized negative other, the opposite of normality in every way. Nevertheless, the one peculiarity that identifies them as an ethnos cannot be imported into a patriarchal culture. Peaceful cohabitation between Amazons and the Scythian young men is possible, but not within the society of Scythian families. “Let us live on our own,” say the Amazons, and their husbands are persuaded.254 Herodotus’ Sauromatian logos conveys two lessons. First, on objective consideration, such as both the ethnographer himself and the characters of this story are able to exercise, the other may turn out to be same. By exposing as illogical the Amazon myth that masculine women are antithetical (rather than analogous) to men, the story undermines the notion of alterity for which the myth stands. Second, it prescribes that a real incompatibility is to be dealt with peacefully through separation. The exemplary behavior of the characters in fact denies the ideology, part and parcel of the traditional Amazon myth, that war is necessary because the other is the enemy; if we do not conquer it, it will conquer us. This inescapable alternative, conquer or be conquered, is mentioned by Xerxes to justify his expedition against Greece and is formulated again by Cyrus at the end of the work.255 But to Herodotus, war is an evil second only to “intertribal” struggle (στα σις . . . εµ␸υλος).256 In the Sauromatian logos, the collectivities involved, though all warlike to an extreme degree, seem remarkably determined to avoid both war  254. 4.114.4– 115.1. Notice the noncoercive nature of the deliberations: ε ι βουλεσθε    (4.114.4); ε πει θοντο (4.114.4, 115.3); ε πειτε αξιο υτε, ␸ ερετε (4.115.3). 255. 7.11.3, 9.122.3. See Raaflaub 1987, 228, for varied formulations of the same idea in Herodotus, Thucydides, and other texts. The third option, which Otanes favors in the  . . . αρχειν   vertically analogous context of a society’s internal organization, is ουτε ουτε  αρχεσθαι (3.83.2). 256. 8.3.1; see chap. 3, “The Evils of War.”

Comparison

131

and intertribal struggle. After the initial battle, the Scythian adults and the Scythian young men do not fight with the Amazons, nor do the Amazons fight with them. Hoping to coopt the Amazons as child bearers, the Scythian adults find themselves deprived of the sons they already have, but they let them go without resistance. The young men even obtain from the Scythians their share of family property before leaving for good (4.115.1). More directly emphasized, for obvious corrective reasons, is the behavior of the Amazons themselves. The Amazons realize that their initial sea crossing into Scythia has entailed a number of more or less voluntary and aggressive violations that now put them at risk. We are afraid and frightened to have to live in this place, because, firstly, we have deprived you of your fathers and, secondly, we have greatly devastated your country. So they plan to undo the invasion and leave. Since you think fit to keep us as your wives, do the following  together with us. Come, let us move away [ε ξαναστεωµεν] from  this land and, crossing [δια βαντες] the river Tanais, live over there. (4.115.2– 3) In the Histories, the fearless men who cross rivers (with the verb διαβαι νειν) do so to subdue and conquer, appropriate what belongs to others, add to their rightful share, and, in concrete metonymic terms, acquire more land.257 Here the Amazons and their new husbands, who once again “were persuaded” (4.116.1), abandon a country that is not or is no longer their own and cross the river (δια βαντες), removing themselves—at a distance of a three-day journey away, to be exact—from conquest and war.258 According to the symbolic code of the Histories, the Amazons’ crossing of the Tanais is a violation of boundaries in reverse and a spectacular display of sophrosune. The subtext of the Amazon myth determines the 257. See my introduction, n. 26 and corresponding text. For words of the διαβαι νω family marking “unwise imperial ventures,” see Lateiner 1989, 131– 32; Payen 1997, 140. 258. The “three days” may belong to the contemporary political code. Diodorus (12.4) reports a clause of the Peace of Callias that stipulated that Persian armies should not come nearer to the coast than three days’ march. The crossing of the river in reverse is also noted by Flory (1987, 112– 13).

132

Telling Wonders

polemic intent of the story and enhances its participation in the general message of Herodotus’ ethnographic historie. According to the traditional image, the Amazons were invaders who lusted for conquest, while the autochthonous Athenians were and are by nature just and would never deprive another of his land.259 Representing the product of Herodotus’ historical and ethnographic research, the Sauromatian logos shows that the Amazons were neither wild, nor violent, nor cowardly, nor gutless, nor eager to enslave, nor ignorant of justice, nor the enemies of the race of men. Here, as elsewhere in the work, female is not the antithesis of male, barbarian is not the antithesis of Greek, and the alternative of conquering or being conquered appears invalid. Herodotus’ pursuit of the similar within his representation of difference confounds mythical constructs of alterity. His scientific ethnography teaches that difference pervades the world, to be sure, but not according to the schematic intellectual map devised by the Greeks. Conclusion I have begun by exploring the extent to which the narrator’s explicit comparative interventions confirm the cohesiveness of the Histories and throw light on their meaning. In the historical narrative, I have argued, the rule of diachronic similarity predominates on its own by implicit analogy. The juxtaposition of narrated data in itself conveys that similar events and features recur (horizontal analogy) and that certain actions or objects figuratively represent more abstract qualities and general processes (vertical analogy). Thus, the narrator rarely points out in his own voice, “Such and such is analogous to such and such.” Two cases of this sort provide to the recipient of the narrative additional help on shocking or particularly obscure connections. In doing so, they confirm the correctness of the listener’s interpretation of the history through both horizontal and vertical analogy. To the extent that these glosses make patent the narrator’s own process of interpretation, they also reveal the encompassing range of the analogical field that revolves around kingship as both historical reality and historical symbol. If historical events emerge as being like each other diachronically, 259. See the Athenian ambassadors at 7.161.3. The Amazons’ lust for conquest is already implicit in Aesch. Eum. (685– 90). See also Isoc. Paneg. 4.68– 70; Lysias 2.4. In Demosthenes (6.4– 8) the connection between autochthony and justice is immediately followed by the mention of the Amazons’ invasion. See Tyrrell 1984, 114– 16.

Comparison

133

ethnography and geography describe synchronic difference around the world. Without difference, there is nothing to narrate. Glosses explicitly declaring the difference of a phenomenon from another or its uniqueness periodically underline this inherent presupposition of the narrative. Glosses of difference are not always advertisements of narratability, however. Rather, in certain cases, they aim at breaking down ethnographic categories into smaller, if often more elusive, groupings. This maneuver seems designed to contradict and scramble excessively schematic notions, much in the same way as when Herodotus devalues the conventional subdivision of the earth into large sections. Conversely, explicit glosses of difference that engage as whole categories the Greeks, on the one hand, and the barbarians, on the other, are almost entirely absent. This constitutes again a move away from convention. Herodotus is reluctant to theorize the Greeks as special or even as the norm in an absolute sense. Most important is the frequency with which the whole representation of difference is counterbalanced by glosses that explicitly state that an ethnographic or geographic phenomenon is like another one somewhere else, either foreign or Greek. These statements scan the objective account of specific facts and add plausibility to its intended direction. When Herodotus describes how various ethnographic subjects differ from the Greeks and emphasizes their separate identities—the different ways in which they differ from the Greeks—this also conveys the different ways in which they resemble the Greeks or different groups of Greeks. The glosses of similarity compensate for the propensity of ethnography to result in a discourse on alterity, especially the alterity of the barbaroi as a whole to the Greeks as a whole. They are reminders of an ideology of sameness that manifests itself also in the unmarked representation of shared features in the context of the objective description of difference. The identification with the other, like the partial devaluation of geographical boundaries, participates in Herodotus’ overarching idea of a world that is differentiated and homogeneous at the same time. This in turn is designed to spoil for the Greeks the pleasure of contrast effects, to uncover for them surprising paradigms, and also to deny them the separate role of spectators of barbarian woes, such as Aeschylus’ Persians had allowed them to take. Each ethnos possesses its special identity and history, but pervasive and unexpected likenesses among all ethnea guarantee on principle eventual resemblances in their historical experience, as happened with the Scythians and Greeks. The synchronic patterning of the world confirms and explains the predictability of the patterns of history.

Chapter 3

Interpretation and Evaluation

A narrative text encourages its recipient to understand the events it recounts in a certain way through strategies that are largely indirect, such as the selection of the facts it narrates and the words by which it chooses to represent them. But when we hear the narrator express the main point about the facts of a narrative, speculate on its less obvious aspects, express a judgment, or draw the moral of the story, we have an interpretive gloss. This is a very broad metanarrative category, potentially coinciding with the range of referential metanarrative itself. Among explicit comparisons, a bold case of interpretation appears in the gloss that points out that Cleisthenes of Athens reproduced in his democratic reforms certain actions of his grandfather, the tyrant of Sicyon. The gloss “just like the Greeks” attached to the factual statement of Egyptian monogamy bids the listeners to think of the Egyptians in terms of affinities with themselves, rather than differences. Both these elements of the discourse provide, as do interpretive glosses in general, directions for “reading” (in the sense of decoding) the narrative. The narrator also enhances the listener’s understanding of what is being narrated by expressing or implying his approval or disapproval. Evaluation of worth is always based on certain interpretive assumptions about the meaning of the action that is praised or blamed; conversely, to explain meaning even in neutral terms often promotes a judgment in terms of “Good” or “Bad.” The distinction between evaluation and interpretation, in other words, is bound to lack rigor, but its usefulness for the purposes of the present book has to do, once again, with the differences we notice in the approaches of history and ethnography, respectively. History first and foremost investigates the meaning of events—why they occur and what general laws one can derive from the accumulation of factual data. We therefore provisionally subsume evaluation to interpretation. Herodotus’ ethnographical descriptions, in contrast, often leave unanswered (and unasked) why people customarily do what they do (e.g., 134

Interpretation and Evaluation

135

sacrifice) in a certain way. Even in cases that may appear controversial to us, Herodotus tends to report the meaning of actions as a matter of fact.1 Although cumulatively ethnography leads Herodotus to an understanding of the world, the main problem in individual ethnographic statements and sections is rather the extent to which his discourse either valorizes difference or encourages the assumption of its inferiority. As I examine the ethnographies, therefore, I will begin by subsuming interpretation to evaluation and focus on the ways and means by which Herodotus promotes approval or disapproval of his ethnographic subjects. Evaluation in the Ethnographies Explicit Evaluation I have argued in the preceding chapter that Herodotus’ insistence on the strands of cultural similarity connecting different ethnea around the world contributes to undermining the notion of a clear-cut separation between Greeks and non-Greeks and questions the validity of those categories. But the canonical antithesis pervades the logos and implicates the histor. The Spartan ambassadors in Athens declare that “with barbarians there is no faith or truth” (8.142.5). Pausanias tells the Aeginetan Lampon that to mutilate the corpse of Mardonius would be an action “befitting barbarians rather than Greeks” (9.79.1). Both these judgments occur in ironical contexts and have no authority from the point of view of the text.2 When Xerxes lashes and brands the Hellespont, however, the evaluation of his words as “barbarian and impious” [βαρβαρα τε κα ι

 σθαλα] (7.35.2) comes directly from the narrator and testifies to his ατα acceptance of the negative ethical connotations of the word barbaros. The exclusive application of the word in this sense to the most symbolic monarchical transgression in the entire logos helps to explain the partial dissociation between the barbaroi of the ethnographies and the concept of barbaron in the history. Because of the regocentric character of the historical narrative of non-Greek actions from Croesus to Xerxes, 1. E.g., from our point of view, Herodotus “interprets” the ritual described at 4.62 as a sacrifice to Ares. Mora (1985, 124– 25), e.g., disagrees on both counts. But from the point of the text, this is a sacrifice to Ares. 2. On irony in the sequence of speeches at 8.140– 9.11, see Fornara 1971a, 84– 86; Raaflaub 1987, 240. Spartan deceitfulness is a contemporary stereotype, at least in Athens. See Hdt. 9.54.1, 9.11, 6.108; Powell 1988a.

136

Telling Wonders

entire cultures tend to become identified with their less typical specimens and bear the burden of providing the negative paradigm as either ruler or ruled. Monarchy carries such powerful negative connotations that it cannot help affecting the evaluation of the peoples who accept and perpetuate this form of government.3 This evaluative constraint also has a certain carryover in points of representation of a nonpolitical nature. Thus, in the ethnography, the Lydians are authors of inventions later adopted by the Greeks (1.94.1– 2), but in the history, the Greek Thales scientifically predicts the eclipse that appears as a terrifying portent to the Lydians and the Medes (1.74.2). In the ethnography, the Egyptians do things earlier and more intelligently than the Greeks, yet in the history, a Greek doctor proves superior to the Egyptian physicians (3.129– 30). The main feature of the discrepancy is, however, that the historical narrative of the actions of kings is likely to be the locus of orientalistic representation.4 The ethnographies—which describe a wider range of cultures and take a broader outlook on culture—go a long way to repair the damage. Indeed, one might argue that from the point of view of the overarching message of the Histories, this is one of their primary functions. Nowhere in the logos, at any rate, do we find the narrator evaluating a known historical people with a statement remotely resembling the following statement that Thucydides appends to one of his narratives, despite Thucydides’ more limited opportunities for characterizing foreigners. For the Thracian race, like the bloodiest of the barbarians, is even more so when it has nothing to fear. (Thuc. 7.29.4, trans. Crawley in Strassler 1996). Herodotus applies negative evaluations to a few remote peoples, who play a very small role in the historical narrative and whose customs he cannot regard as morally equivalent to those of more advanced ethnea.5 An extreme and unique case is represented by the utterly unregulated Androphagoi, who illustrate Herodotus’ identification of law/custom/ culture with morality: they “have the most savage ways of all men 3. See the implicitly evaluative interpretive gloss to the report that the Egyptians, after gaining their freedom, set up twelve kings, “for at no time they were able to live without a king” (2.147.2). In the historical narrative, cf. 3.143.2, on the Ionians of Samos, and 1.62.1, on some Athenians in the time of Pisistratus. 4. On orientalism, see Sa¨ıd 1978, 5– 9 and passim. 5. E.g., he describes peoples who couple “like cattle” (1.203.2, 3.101.1, 4.180.5).

Interpretation and Evaluation

137

  [αγρι ωτατα ηθεα], and they do not practice justice as a custom nor do   ουδεν  δι κην νοµι ζοντες, ουτε  νοµω

they have any law/custom [ουτε ι 6  χρε ωµενοι].” Two evaluative statements that explicitly praise a people’s morality are designed to overcome the “Bad” impression that the report of their customs may have created. The funeral practices of the Issedones involve cannibalism and the fashioning of the dead man’s skull into a cup. Here, Herodotus’ effort to promote in his audience a relativistic attitude has already expressed itself in the comparative gloss that made these practices manifestations of piety on the same level as a Greek ritual. In the subsequent gloss that “otherwise these also are said to be just and to enjoy equal rights, both men and women” (4.26.2), also (και ) assimilates the Issedones to the Bald Men, who are at the opposite end of the spectrum to the Androphagoi and paradigmatically just.7 More impor tantly, otherwise (αλλως) insists on the principle that justice is a virtue that transcends cultural differences; but it also reveals the difficulty of the relativistic position. Similarly, the totalizing statement that the Getae are  

the “most courageous and just” [ανδρηι οτατοι κα ι δικαιοτατοι] of the Thracians (4.93) counterbalances, on the one hand, the information that they were conquered by the Persians and, on the other hand, the description of their cruel ritual in honor of their god Salmoxis, whose absurdity the text can hardly conceal.8 A different application of this compensatory strategy consists in punctuating the nonjudgmental account of practices that are bound to strike a listener as savage with very specific praise of certain items of material culture. In his description of the brutal Scythian sacrifice, for example, the narrator conveys the idea that despite differences and deficiencies, things nevertheless get done in a satisfactory way. Since the land of Scythia is  “terribly poor” in wood, the Scythians “have invented” [ε ξευρηται] a special procedure for cooking the meat by lighting the fire with the bones. They cook the meat in local cauldrons, “if they have them, very similar to Lesbian craters, except that they are much bigger.” If they do not have a cauldron, then they throw all the meat into the stomachs of the victims: “The meat cooks beautifully [αιθεται καλλιστα], and the stomachs 6. 4.106. As Redfield explains (1985, 98– 99), animals also have ethea, but only human beings have nomoi. 7. See 4.23.2– 5. Equality between the sexes also obtains among the Bald Men. See chap. 2, n. 187 and corresponding text. 8. 4.94; see Hartog 1988, 84– 109; Lateiner 1990, 235– 45.

138

Telling Wonders

 contain the meat easily [ευπετ εως] after it has been boned. And so the ox cooks itself, and the same goes for the other victims.”9 Sophie constitutes a conspicuous field of ethnographic evaluation.10 It is a diverse quality that may include resourcefulness, practical or theoretical intelligence, learning or wisdom; when it is attributed to societies, it lacks the ethical dark side it occasionally connotes in individuals. The dogma of Greek superior intelligence is the subtext that makes all the explicit or implicit attributions of sophie to foreign nations indices of similarity to the Greeks. At the only time when the narrator formulates the dogma directly, however, he encases it in a demonstration of Athenian foolishness (1.60.3). Conversely, the totalizing statement that evaluates the group of ethnea around the Pontos as “the most ignorant people”

 [εθνεα αµαθ εστατα], occurs in the context of Herodotus’ praise of the   Scythians for making the “cleverest discovery” [σο␸ ωτατα ε ξευρηται]— 11 a mode of life that enables them to elude subjection. Several ethnographic evaluations of worth are marked by self-referential signs I have grouped in the broad category of “opinion.”12 Some of them, however, express a more subjective and less reasoned inclination  than a γνωµη. They place the individuality of the narrator in the foreground by emphasizing his entitlement to a subjective reaction to something foreign, and they appear to constitute one of his responses to the tension between a relativistic ideology and an evaluative impulse. Thus, concerning the Scythians, the narrator points out the paramount strategic value of nomadism, but aside from this, he adds, “nothing else pleases  

αγαµαι]  me” [τα µεντοι αλλα ουκ (4.46.2). While acknowledging the argument of those societies that allow intercourse in temples, he still proclaims his personal dislike of the practice: “I just do not like it”  ].13

αρεστα

[εµοιγε ουκ Couched in objective terms, in contrast, is the negative evaluation of the Babylonians’ “most shameful custom” (1.199.1). This includes not 9. 4.61.1– 2. The narrator also praises Scythian and Thracian hemp (4.74), the Scythians’ lustral steam bath (4.75.1– 2), their aromatic body pack for women (4.75.3), and the availability to them of essential resources (4.59.1). See Romm 1998, 111. 10. See, e.g., Herodotus’ representation of the Assyrians, discussed in my introduction. 11. 4.46.1– 2. The importance Herodotus attributes throughout the logos to a nation’s ability to maintain its freedom motivates also evaluations concerning the population size and military strength of Thracians (5.3.1) and Ionians (1.143.2). See Evans 1976. 12. See chap. 1, “Self-Referential Glosses.” 13. 2.64.2. Cf. the subjective formulations of 2.3.2 and 2.46.2, discussed in chap. 2, “The Texture of Nomos.”

Interpretation and Evaluation

139

merely intercourse in a sacred place but the obligation for all local women to subject themselves to ritual prostitution in the temple of Aphrodite once in their lifetime. The narrative provides a neutral but eloquent account of the humiliation of the wealthiest among the women, each compelled to follow the first stranger who comes along and throws whatever amount of money he wishes at her feet.14 The narrator underlines the uniqueness of the occasion with a direct challenge to the listener: after the woman has returned home, “you are not going to get her no matter how much money you give.” We also receive a vivid impression of the ordeal of the less attractive women, each of whom must sit in the temple even for three or four years until some man chooses her and allows her to fulfill the ritual (1.199.1– 5). This exceptional case of explicit negative moral evaluation in an ethnographic context begins to suggest a distinction between customs that are oppressive and customs that are welcome and helpful to the people who follow them. The exact counterpart of “the most shameful custom of the Babylonians” is one that Herodotus introduces as “their most intelligent    η µετερην]  custom, in my opinion” [σο␸ ωτατος . . . κατα γνωµην τ ην (1.196.1) and which he later calls “their most beautiful custom” [ο  καλλιστος νοµος] (1.196.5), thereby signaling moral approval as well. This practice also involves a mercantile exchange of women, with the attendant problem of an inequality between the more and less beautiful. This time, however, the institution consists in a yearly auction of brides for the achievement of stable marriages, carefully designed to compensate for preexisting disadvantages and to make everyone happy. In Aristophanes’ Ecclesiazosae, a roughly comparable utopian arrangement is called a  δηµοτικη γνωµη (631). This “democratic” quality lies at the basis of the excellence of Herodotus’ Babylonian nomos, unfortunately superseded since the advent of hard times by the definitely “undemocratic” and exploitative practice of prostitution (1.196.5). A similar criterion of evaluation also emerges from the Babylonian custom judged “second in sophie” (1.197), where the marketplace principle is applied to the field of health care. Since the Babylonians have no professional class of doctors, they bring all the sick people to the agora, where passersby proffer advice and prescriptions, each according to his personal or indirect experience of this or that ailment. In the case of the 14. A form of ritual prostitution also existed in Greece, but not for citizen women. See Halperin 1990, 104– 7, for sources.

140

Telling Wonders

market of brides, various prohibitions were in place to prevent one from eluding the system (196.3– 4); so also in this case, no one is allowed to walk by a sick person without asking him why he is suffering. Both nomoi provide for the exchange of resources according to need, and both preclude individualistic violations. The Babylonian custom of gathering the sick in the center of the city for the purpose of caring for them implicitly contrasts with the Persian practice of banishing from the city altogether whoever suffers from leprosy or the “white disease.”15 To the explicit praise of the Babylonian custom, however, there corresponds no explicit negative evaluation of the Persian. What we find instead is a justification of the (implicitly “Bad”) Persian practice in terms of native knowledge: they banish the sick because “they say” that these particular diseases afflict those who have sinned against the Sun. The fact that explicit praise is more frequent than explicit blame points to a search for whatever “Good” features a certain culture has to offer from out of its own store of nomoi, institutions, or resources.16 In the description of the highly hierarchical Persian society, where there is no trace of the demotikon that Herodotus is able to find among the Babylonians, Herodotus chooses two nomoi of an entirely different sort on which he dispenses personal praise through the per “I approve” (1.137.1). Both these customs aim at formative verb α ι ν εω, protecting the patriarchal head of the family and state from succumbing  to excessive emotions—respectively, grief (αση) at the possible death of  against servants and an infant son (1.136.1) and impulsive anger (θυµος) subjects (1.137.1). As in the case of explicit comparisons, evaluations point to the ideological direction in which the text travels by implicit means. In the marketplace or other community space, people can observe what they like or need among different items, as in the wisest and second wisest Babylonian practices: this setting also serves as the metaphor for the activity of the ethnographer, who places “in the middle” a variety of different peoples and their nomoi for the benefit of his audience. Look15. 1.138. The antithetical character of the two customs is observed by Asheri (1988, 380). 16. See the praise of the Persian postal system (8.98.1). See also statements praising foreign natural resources (1.193.3, 4.198, 4.194, 3.106.1, 3.112, 3.113.1), the “correctness” of certain barbarian names (4.59.2, 6.98.3), and a people’s physical health (4.187.3, 2.77.3). No negative evaluations concerning health occur, as they do in the Hippocratic Airs, Waters, Places.

Interpretation and Evaluation

141

ing on entails evaluation and possibly a negative judgment, since people like their own customs best (3.38.1). Herodotus recognizes this inevitable tendency through his own evaluations but also keeps it under control.17 He models for his listeners an attitude of charitable observation; when he does not lead them to the realization of unexpected likeness, he promotes the discovery of understandable difference and creative solutions.18 This attitude dissociates the notion of barbarian from that of the barbaric and replaces a generalized contempt for alien customs with a more self-conscious definition of what must necessarily be the furthest limits of one’s tolerance. Strategies of Evaluation Revising Greek Traditions Herodotus’ descriptions of foreign cultures frequently imply a context of Greek ignorance and prejudice and thereby signal the ethnographer’s corrective aims. In the field of interpretation and evaluation, one of Herodotus’ strategies consists in refuting or countering Greek stories that perpetuate damaging stereotypes of the barbaroi. In book 2, passages that criticize Greek traditions concerning Egypt or the Egyptians are a part of the polemic against different branches of Greek knowledge (chronology, geography, the gods), which Herodotus pursues while more broadly demonstrating the limited power of historie for all sides involved.19 Programmatic in this respect is the initial set of two logoi (2.2.1– 5). To the experiment of Psammetichus as a positive, though imperfect, barbarian instance of historie in the narrative, the metanarrative juxtaposes a negative Greek one: according to the Greek version of the story, for the purposes of his experiment, Psammetichus cut off the tongues of his servant women. This is one of the “many frivolous things [µαταια πολλα ] that the Greeks say.”20 Similarly, at the end of his radical 17. Contrast the evaluative mode of the Hippocratic author of Airs, Waters, Places or of Strabo/Ephorus (see chap. 2, “Does Climate Determine Culture” and “The Sameness of the Scythians”). 18. For the concept of “charity,” see Asad 1986, especially 147 in reference to modern ethnographers’ tendency to give a favorable account of native phenomena. 19. See Hunter 1982, 50– 92, especially 61, 64– 65, 67, 69– 70. 20. On Psammetichus in this passage as a model histor and analogue of the histor Herodotus, see Christ 1994, 184– 86, especially 185. On the limitations of both Psammetichus’ interpretation of his experiment and the Greek version of the story, see, e.g., Benardete 1969, 33; Dewald 1998, 615.

142

Telling Wonders

revision of Greek traditions concerning Heracles in light of information acquired in Egypt (2.45.1– 3), the narrator dubs the Greek story that the Egyptians attempted to sacrifice Heracles another typical foolishness  

ι ρους) of Egyp(ευ ηθης . . . µυθος). Here the Greeks’ inexperience (απε  tian culture and nature (␸ υσις in the sense of “national character”) contrasts with Herodotus’ experience of these things (see δια πειραν at 2.77.1); it leads them to believe in an occurrence that even contradicts  nature (␸ υσις in the sense of “what is humanly possible”). As at other times in Herodotus, custom and nature, or disregard and ignorance thereof, are here closely related.21 While the Greek version of Psammetichus’ experiment reveals a generalized Greek stereotype of barbarian cruelty, the tale of Heracles’ sacrifice is based on a contemporary prejudice that is specifically tailored to Egypt. Confronted with Egyptian aloofness, the Greeks constructed a fantasy of Egyptian hostility toward guests, as is represented outside of Herodotus by the legend of how the Egyptian king Busiris habitually sacrificed travelers arriving to his land.22 The Busiris model of xenophobia also lies behind the Egyptian version of a Greek story that Herodotus reports to provide an implicit reversal of the Busiris myth. In this passage (2.112– 20), both the historical knowledge and the national character of the Egyptians are vindicated through no less a topic than the archetypal Greek saga of the fall of Troy. Herodotus has learned from the priests of the sanctuary of Proteus in Memphis that Alexander and Helen landed in Egypt on their way from Sparta to Troy. The slaves of Alexander sought refuge in Egypt, as suppliants in a nearby temple of Heracles, and told the priests about their master’s abduction of Helen. The warden of the mouth of the Nile, a fellow named Thonis, informed King Proteus. After confronting Alexander, Proteus banished the wrongdoer from his land but kept Helen, her treasure, and the suppliant slaves 21. Cf. the rhetorical questions “How does it accord with nature that Heracles, being only one and a man, should kill many tens of thousands?” (2.45.3) and “How could a dove speak with a human voice?” (2.57.2) in Herodotus’ refutation of the Greek story of the foundation of the sanctuary of Dodona, another tradition born from the Greeks’ inexperience of ethnic difference. Herodotus also indicts Greek “incorrect speaking” in his refutation of the story that the courtesan Rhodopis built the pyramid of Mycerinus (2.134.1– 2.135.6). 22. On Busiris, see especially Froidefond 1971, 177– 79; Lloyd 1976, 212. The earliest testimonies of the story date to the sixth century: see [Hesiod] frag. 378 MW; Pherecydes, FGrHist 3 F 17; Panyassis frag. 26K (⫽ Athen. 4.172D); and the late sixth-century b.c. hydria from Caere in Vienna (3576).

Interpretation and Evaluation

143

under his protection until the time when Menelaus would come and claim them back (2.113– 15). Alexander sailed back to Troy, where the Greek army was already gathered. The Greeks requested the return of Helen, but their embassies bore no results: the Trojans kept saying that Helen was in Egypt, but the Greeks did not believe them. The war went on until, finally, the city was captured; only at that point did the Greeks realize that Helen was not at Troy and send Menelaus to look for her in Egypt.23 Herodotus again calls what “the Greeks say” (which here means the  canonical Homeric version) a µα ταιος λογος (2.118.1), and he corroborates this Egyptian story in three different ways. First, he reports his independent conjecture, based on autopsy, that the shrine in Memphis uniquely named after “Foreign Aphrodite” or “Aphrodite the Guest” must be a shrine of Helen (2.112.2). Second, he examines four epic passages for the purpose of demonstrating that Homer knew about the sojourn of Helen in Egypt, though he followed the incorrect version because it was “appropri ε ποποιι ην ευπρεπ

 (2.116.1). Finally, he ate to epic poetry” [ε ς τ ην ης] refutes the Greek version, showing its logical flaws on the basis of the likely (2.120.1– 4). The Egyptians prove again to be superior sources, even on a fundamental issue of ancient Greek history.24 The truthful Egyptian version has the advantage that it more clearly reveals a fundamental pattern of history, as I shall show when I discuss the most generalized level of Herodotus’ interpretation.25 But from an ethnographic standpoint, which concerns me here, it counters “foolish” Greek prejudices with a correct representation of Egyptian phusis and nomoi. The theme of xenie (hospitality) links the myth of the Egyptians attempting to sacrifice Heracles (with its Busiris subtext) and the story of Helen in Egypt, as Plutarch noticed.26 In the story of Helen, the stem ξενoccurs a total of twelve times, and the stem δικ- (justice) occurs eight times. A great deal of character text discusses what is or is not a just and  holy (οσιος) treatment of guests. Proteus reproaches Alexander in terms 23. 2.118.2– 4. On this story, see Fehling 1989, 59– 65; Hunter 1982, 52– 61. Austin 1994, 118– 36. 24. At 2.118.1 and 2.119.3, the text refers to Herodotus’ Egyptian sources with the same vocabulary of inquiry and knowledge that Herodotus elsewhere applies to himself:

 “to know through inquiry” [ ι στορι ησι . . . ε πι στασθαι], “knowing precisely” [ατρεκ εως

ε ιχον ει πειν]. For the narrator’s evaluation ε πιστα µενοι], “they were not able to tell” [ουκ  of the Egyptians as λογι ωτατοι (most learned), see 2.77.1; cf. 2.4.1. For the real Herodotus as a logios, see Nagy 1990, 221– 24; Evans 1991, 94– 98. 25. See 2.120.5 and “Divine Retribution” later in the present chapter. 26. Plut. De Malign Herodot. 12 ⫽ Mor. 857A– B. See also Froidefond 1971, 178– 79.

144

Telling Wonders

that characterize this Egyptian king as a reversal of the guest-killer Busiris: If I did not consider it of the greatest importance not to kill any of the guests who are driven by the winds to my land, I would seek from you retribution on behalf of that Greek, O most wicked of men, you who after receiving hospitality have perpetrated a most impious deed: you have come to the wife of your own guest . . . and this was not enough for you, for you come here after plundering the house of your guest. But now, since I consider it of the utmost importance not to kill guests, I will at least prevent you from taking away this woman and the treasure, and I will hold them in my keeping for the Greek stranger/guest [ξει νω  ] until he wants to come himself and take them back (2.115.4– 6). The presence of a temple with a special nomos protecting suppliants enhances the representation of Egyptian piety. The role of the warden Thonis ensures that Proteus’ righteousness will be perceived as also typical of ordinary Egyptians. The impeccable Egyptian hospitality contrasts with Alexander’s “injustice” and “impious deed” (2.113.3; 2.114.2; 2.115.3, 4; 2.120.5). But the story gains a special polemical edge (again not lost on Plutarch) when the  Greek guest in this story, Menelaus, “behaves unjustly” [ε γενετο ...  

αδικος] and “devises an impious action” [ε πιτεχναται πρηγµα ουκ  οσιον] (119.2). After receiving Helen and her treasure back with hospitable gifts from Proteus (ξενι ων . . . µεγαλων, 2.119.1), Menelaus sacrifices two local children to obtain favorable winds for his sailing. Adapted to the xenie theme of the narrative, this reenactment of the sacrifice of the Greeks at Aulis produces a host-sacrifice that reverses the alleged guestsacrifices of the Egyptian Busiris.27 Like the anti-imperialistic Amazons in the Sauromatian logos, the hospitable Egyptians of this story set the record straight with regard to the representation and evaluation of the other. Native Voices Through native logoi, the text revises the assumptions of the Greeks about barbarians. A strategy for “challenging evaluative preferences for Greek characteristics over barbarian,” in comparison, consists in focaliz27. Fehling (1989, 62) notices the analogy between Menelaus’ sacrifice of the Egyptian children and Agamemnon’s sacrifice of Iphigenia at Aulis.

Interpretation and Evaluation

145

ing Greek and foreign customs through the barbarians.28 Appearing in the logos in the guises of characters, ethnographic subjects, or epichoric sources, “they” participate in the din of voices that resound on many different issues throughout the work. When these speakers talk at different narrative levels about the way in which they or others lead their lives, their utterances serve the purpose of ethnographic evaluation. In Culture and Truth: The Remaking of Social Analysis, anthropologist Renato Rosaldo explains, “If you ask an older Ilongot man of northern Luzon, Philippines, why he cuts off human heads, his answer is brief . . . : He says that rage, born of grief, impels him to kill his fellow human beings.” During his stay among the Ilongots in the late 1960s, Rosaldo was troubled to discover that head-hunting was still surreptitiously practiced and that “every man in the settlement had taken a head.” A few months later, when the ethnographer was drafted for Vietnam, his Ilongot friends surprisingly urged him not to go and offered to conceal him in their homes: “They told me that soldiers are men who sell their bodies. Pointedly they interrogated me, ‘How can a man do as soldiers do and command his brothers to move into the line of fire?’”29 Set in the context of Rosaldo’s self-conscious analysis of available forms of ethnographic discourse (not a type of analysis Herodotus does), these exchanges provide a not too distant parallel for the strategy of representation by which Herodotus bridges the gap or evens the score between foreign and Greek customs by opening the possibility for what Rosaldo calls “reciprocal critical perceptions,”30 in which a native explains himself and/or criticizes aspects of the ethnographer’s culture. Ethnocentric remarks from the other at the very least communicate the idea that what one takes for granted as normal may also be perceived, with some justification, as undesirable and abnormal.31 When Herodotus’ foreigners evaluate Greek customs, they are almost always critical.32 Though they often display ignorance, exaggeration, or a 28. See Pelling 1997. The two functions often overlap. 29. Rosaldo 1989, 1 and 63. See Rosaldo’s entire chapter “After Objectivism” (46– 67). 30. Rosaldo 1989, 64. 31. In Herodotus, see, e.g., the Ethiopian king’s contempt for bread (3.22.3– 4) and the Egyptians’ ethnocentric criticism of the Greeks’ dependence on rainfall, the last both corroborated and countered by the narrator (2.13.2– 2.14.1). 32. Two exceptions (7.208.1– 3 and 8.26.3) occur in the highly celebratory narrative of Thermopylae, where we also find the similarly exceptional case of an entirely misguided criticism of Greek culture by a foreigner (7.103.3).

146

Telling Wonders

quaint partiality of their own (as Rosaldo’s Ilongots do when they define soldiers as “men who sell their bodies”), they also make valid points.33 Among observers of Greek nomoi who combine misunderstanding with brilliance is Mardonius. His main point to Xerxes that the Greek method of fighting is ineffectual and uneconomical constitutes “foolish words”  [µαταιοι λογοι], “slanderous” against the Greeks, as Artabanus will re34 tort. But when he implies that it is the peculiar nomos of the Greeks to fight one another incessantly, this marks one of those great polemic moments that leave no doubt as to what the Histories are about, whom they address, and what solution they prescribe: “And yet, since they speak the same language, Mardonius says, they should settle disputes with heralds and messengers and in any other way except by fighting” (7.9β1– 2). The reference to a common language, which Mardonius intends as a practical consideration, defines for the audience a conventional aspect of their Panhellenic cultural identity (8.144.2). The phrase “settle disputes with heralds and messengers and in any other way except by fighting” belongs to the political code of Herodotus’ time. Thus, in Thucydides, Pericles maintains that the Spartans “prefer to settle disagreements by war rather than by words”; Thucydides cites the suspension of parleys through heralds on both sides as the first decisive sign that their disputes have finally determined the outbreak of the Peloponnesian War.35 Cognitive Statements and Polemical Negations Since negative statements serve to contradict the audience’s expectations, they often enhance the representation of difference by underlining a culture’s lack of features typical of “normal” civilized living.36 At other times, however, negations reflect a people’s ideology and its polemics against foreign customs. The ethnographic gloss explaining that the Per33. See, e.g., 4.79.3 (Scythians’ polemic against Dionysus; see chap. 2, “Identification with the Other”; 4.142 (Scythian criticism of the Ionians as “slaves”), 2.160.2– 4 (Egyptian criticism of the rules of Olympic Games). 34. 7.10η1. For µαταιος applied to ethnic prejudice, see discussion under “Revising Greek Traditions” earlier in this chapter. On the economical nature of Greek warfare, see Hanson 1989, 1– 18. 35. Thuc. 1.140.2 (Pericles here echoes the exhortation of the ephor Sthenelaidas at 1.86.3), 1.146. 36. See Hartog 1988, 257. Among statements of true “lack,” the limiting case occurs at 4.106. On the portrayal of deficiencies and primitivity through “empty negatives” in modern ethnography, see the critique of E. Le Roy Ladurie’s Montaillou (1975) in Rosaldo 1986, 85.

Interpretation and Evaluation

147

sians “do not use marketplaces, and do not even have a marketplace to begin with” for example, is attached to Cyrus’ contemptuous definition of the Greek agora as “a designated place in the middle of the city where people gather and deceive one another under oath” (1.153.2). This cognitive use of negative statements underlines that foreign peoples are different in ways that are desirable to them. They do not have, do, or believe something that is normal for the Greeks, not because they are more limited in some sense, but because they do not consider that item useful, holy, just, or true.37 Negative statements, whether in the voice of characters, ethnographic subjects, sources, or the narrator, may also serve to preempt biased judgments in the listener (“It is not true, as you think, that . . .”). In book 4, Herodotus relates a remarkable system of commercial exchange that takes place between the Carthaginians (who are also the source for this information) and an unidentified Libyan people living beyond the Pillars of Heracles (4.196). After arriving to the shores of these men, the Carthaginians display their wares in orderly array along the beach, then return to their ships and make smoke signals. When the natives see the smoke, they go down to the beach, deposit the amount of gold they offer for the merchandise, and withdraw. The Carthaginians in turn disembark and look at the gold. If they judge it sufficient to pay for the merchandise, they take it and sail away; if not, they return to their ship and wait. The Libyan natives return and keep adding gold until the merchants are satis

 δ ε ουδετ

 fied, and “there is no foul play on either side [αδικε ειν ερους]: neither do the Carthaginians touch the gold before they judge that it has reached the price of the merchandise, nor do the others take the merchandise before the Carthaginians have taken the gold” (4.196.3). Despite peculiar circumstances and the stereotype about dishonest Phoenician merchants,38 the system works well. The reassuring negative statement here serves as a positive evaluation that defies common assumptions. There happens to be a thematic link among what Mardonius says 37. The statements that a certain people will not borrow foreign nomoi belong to this category (2.79.1, 4.76.1). For negative prescriptions, see, e.g., 2.37.3, 4– 5; 2.45.2; 2.46.1; 4.63. For negative belief implying criticism, see, e.g., 2.50.3. On the Greek side, cf. Sperthias and Boulis’ statement that it is not their custom to prostrate themselves in front of a human being (7.136.1). 38. Phoenician dishonesty, notorious in Homer, is, e.g., built into the tradition corroborated by Herodotus at 2.54, 56.

148

Telling Wonders

about the Greeks who speak the same language but prefer fighting to negotiations, Cyrus’ perception about the Greeks’ deceiving one another when they trade in the agora, and the Carthaginian account of their seashore transactions that overcome a language barrier and entail no wrongdoing. This thematic link is the issue of fair dealing within and between different societies. It demonstrates the web of meaning that runs through the logos and the consistency of Herodotus’ ideology of mutual exchange and cooperation. The “most beautiful” Babylonian custom of the market of brides and their custom “second in wisdom,” the exchange of medical advice in the marketplace (1.196– 97), present the same structure. Their prostitution ritual is “most shameful,” but the listener should not misunderstand its import and meaning: after a woman has fulfilled the requirement, “you are not going to get her no matter how much money you give” (1.199.4). In the episode of the marketplace on the beach, Herodotus’ epichoric sources report what they do. The boundary between informants and ethnographic subject is here ill defined, but the phrase “the Carthaginians say” counts as a gloss of source identifying the narrative as someone else’s logos. In other cases, ethnographic subjects state their own cultural beliefs. In most cases, these cognitive statements, like native criticisms of Greek customs, contribute to portraying a culture’s strong sense of itself. By presenting an alternative but legitimate way of thinking, they tend to convey positive evaluation.39 The most striking illustration of Herodotus’ use of cognitive statements is represented by the Persian ethnography. Here we also find several other phenomena I have mentioned: negative statements of rejection, native criticism of Greek customs, explicit (positive) evaluations, mention of customs that elicit a “Good” impression as well as implicit similarity between the barbarian culture and the Greek. The passage is coherent and illuminating.40 Since the Persians are the most prominent foreigners of the Histories, it deserves to be examined in detail. 39. See, e.g., 2.35.3, 4.65.2. See Redfield 1985, 99. Besides in cognitive statements with   λεγουσι/λ εγοντες, ␸ασι or νοµι ζουσι /νοµι ζοντες, and so on, customs are focalized by their owners in sentences of the type “For them it is most beautiful [καλλιστον] to gather at drinking parties in groups according to age and friendship, men, women, and children together” (1.172.1). See n. 44 in the present chapter. Unmarked implicitly “native” explanations also occur; see, e.g., 4.104: “They have their women in common so that they may all be brothers . . . ” 40. Pace Gould 1989, 98– 99. For a historical discussion of Herodotus’ ethnographic representation of the Persians, see Briant 1988.

Interpretation and Evaluation

149

Persian Ideology The Persian ethnography is a list of nomoi, pure and simple, just as the prospective introduction announces.41 But such predicates as “they do X,” “they have learned to do Y,” and “they have X” alternate with “they assert/believe Y.” The narrator’s first-person glosses guarantee accuracy and testify to the ethnographer’s involvement with his subject. As for the Persians, I know that they have the following customs, that   ποιευµενους]

ε ν νοµω  they do not consider it in their custom [ουκ to build statues, temples, and altars but even attribute foolishness  [µωρι ην ε πι␸ ερουσι] to those who do, because, it seems to me, they 

. . . ε νοµισαν] never believed [ουκ the gods to be of the same nature as men, as the Greeks do. (1.131.1) Here the Persians attack Greek beliefs on a weak point.42 The negative statements that follow either connote Persian rejection of nonsensical accessories in sacrificial ritual (1.132.1: no altars, fire, libation, flute, fillets, barley grain) or enhance the polemic by pointing out a feature that the Persians, unlike the Greeks, would consider unthinkable to do without. After the sacrificer has arranged the meat, a Magus stands by and chants a theogony; for they do not have the custom to make sacrifices without a Magus. (1.132.3) Interdictive negations abound, displaying intransigent correctness. The one who sacrifices is not allowed to pray for good things for himself privately but rather wishes for the prosperity of all Persians and of the king, for he himself is included among the Persians. (1.132.2) They do not make water into rivers, spit in them, or wash their hands in them, nor will they allow another to do so, for they revere rivers most of all things. (1.138.2) 41. There is no subordinate narrative of how institutions came into being, such as we find in the Lydian ethnography (1.92– 94), no mention of particular rulers or historical events, and no description of monuments. 42. See Burkert 1988, 20– 21. Cf. Xenophanes, DK 21 A14.

150

Telling Wonders

The Persians not only are different from the Greeks but also find the Greeks inadequate, with a negative of true lack. They say that the Greeks stop eating when they are still hungry, because after the main meal, nothing else to speak of is served to them. (1.133.2) Persian difference is “Good,” based on a sensible and rigorous value system. To their sons, from the age of five to twenty, they teach only three things: to ride, to shoot arrows, and to tell the truth. Before he is five, a boy does not come in the presence of his father but lives with the women. (1.136.2)  Here the narrator intervenes to approve (α ι ν εω) in the explicit evaluations we have seen. The clipped rhythm of the list of Persian customs contributes to creating a brisk impression of strength. We may compare the Lydian ethnography, which differs from this passage in both the type of information it contains and the form of discourse it employs.43 In relation to the marketoriented Lydian culture, Persian culture is masculine, characterized by rough physical activities, the value of andragathie (manly excellence, 136.1), anti-intellectualism, and the rearing of sons (on Lydian daughters, see 1.93.3– 1.94.1). In their practice of prostitution, the invention of games, and the migration to Tyrrhenia, the Lydians appear to cope rather than act by choice. By contrast, the cognitive approach of the Persian ethnography makes Persian action stem not from historical contingencies but from collective ideology, likes and dislikes. The Persians do not simply prepare a particularly large dinner on their birthday; “they think it right” to do so (1.133.1). Sixteen predicates indicate what the Persians consider, believe, assert, honor, or deem right, while the Lydians acquire a voice only to speak of their long-lost resourceful past.44 Correspond43. See chap. 2, “The Sameness of the Lydians.” 44. See also “this is [for them] a display of manliness” (136.1) and explanations of customs where the focalizer is not marked (1.132.2, 1.136.2). Herodotus’ attempt to capture Persian national ideology is emphasized by Wolff (1934, 158– 62); see also Cobet 1971, 117– 18. For the Lydians, three verbs of saying occur (at 1.94.2) in the narrative of their invention of games and migration to Tyrrhenia, which is all in indirect speech.

Interpretation and Evaluation

151

ingly, we find here a total of twenty-four negations, either in interdictions or in polemical representative statements. The Persians even appear ready to revise reality if it does not conform to their value system. They say that no one yet has ever killed his father or mother, and each time something of the sort happens, they say that when the 

 νγκην] it matter is investigated, of absolute necessity [πασαν ανα would be found that the children were adopted or bastards. For they say, it is not natural that a true parent may be killed by his own child. (1.137.2) Aggressively dogmatic, the Persians consider themselves the best of all cultures. To their internal hierarchicalism corresponds a view of the inhabited world based on ethnic rank. Most of all, after themselves, they honor those who live closest to them, secondly those who live second closest, and then the others in proportion to their distance. Their literally geometric notion of ethnocentrism is reproduced by the ring structure of the ethnographer’s discourse, which places “the Persians  at the center and the decreasingly inferior “oththemselves” (ε ωυτους) ers” at the periphery on either side. They hold in least esteem those who live the farthest from them, believing that they themselves are by far the best of men in everything, that other peoples live more or less far in proportion to their

 and that those who inhabit the lands farthest excellence [αρετ η], from themselves are the basest.45 By association with this Persian mental map, a gloss of similarity recalls a feature of the old Median Empire.  

οντων], During the rule of the Medes [ε π ι . . . Μεδων αρχ the various peoples even ruled [ ηρχε] one another—the Medes over all and over those who lived closest to themselves, these in turn over their neighbors, and these over those near them. The Persians give 45. 1.134.2. Cf. the gradation of barbarism from a Greek point of view in Xen. Anab. 5.4.34.

152

Telling Wonders

 αυτ  δ ε

ον respect according to the same criterion [κατα τον   λογον . . . τιµωσι]. [Among the Medes] each people farther and   δ η το father removed held rule and supervision [προεβαινε γαρ   εθνος αρχον τε κα ι ε πιτροπευον]. This imprecise analogy between the Persian system of “graduated respect” and the Median system of “graduated rule”46 connects ethnocentrism, the primary ethnographic feature of the Persians, to the notion

η),  which turns out to be their primary feature in the of empire (αρχ history. An imperialistic ideology also emerges from a different order of signs. The stark simplicity and prescriptiveness of some aspects of Persian culture combine with acquisitive tendencies. The Persians “believe that  δ η γεαται  ε ιναι] (1.136.1).  multitude is strength” [το πολλον ι σχυρ ον This taste for to pollon manifests itself in their insistence on having many trimmings to their meals (1.133.2), many legitimate wives, and an even greater number of concubines (1.135). Every year, the one who displays the most children receives a prize from the king (1.136.1). In the historical narrative, royal Persians delight in possessing large armies and many subject nations.47 Though multitude signifies masculine strength and though acquisitiveness is part of the Persian desire for domination, acquisitiveness also represents the most indulgent and feminine side of Persian culture. Despite their contempt for foreigners, “the Persians like to appropriate foreign customs more than does any other people” (1.135). To their austere nature religion, they have added the cult of Aphrodite, learned from Assyrians and Arabs (1.131.3), and they import from other cultures luxu

ries and “pleasures of all sorts” [ευπαθε ι ας . . . παντοδαπας]. They have adopted the Median national costume (“because they considered it more beautiful than their own”), Egyptian corselets, and the Greek practice of 46. This is the terminology of How and Wells (1928, 1:116), whose interpretation I have followed in my translation. The analogy misleadingly implies that just as the nation farthest from the Persians is lowest in their hierarchy of esteem, so the nation farthest from the Medes used to be lowest in the hierarchy of power. 47. Xerxes enjoys the sight of his army at 7.44 and counts it at 7.59– 60. An army is a typical Persian gift (see 9.109.3). Konstan (1987) examines many instances of the fondness of Persian kings for quantity and numbers in connection with imperialism, but he strangely does not cite any part of the ethnography. Here also the judicial rule of “calculating whether the offenses are more in number than the services” (1.137.1) is consistent with the habit of counting and measuring goods.

Interpretation and Evaluation

153

making love with boys (1.135). The Persians are accumulators of consumption goods. In the historical narrative, Atossa desires accomplished Greek maids, and Mardonius praises the orchards of Europe in the context of imperialistic schemes.48 We should underline the extent to which the Persian ethnography avoids mentioning orientalistic features, even some that regularly occur in the historical narrative. The description of Persian customs keeps the role of monarchical despotism to a minimum. In the Lydian ethnography, the Lydians en masse, in their triple socioeconomic subdivision, raise a monument to their king with the fruit of their labor (1.93). Here the Persians worship the gods (1.131– 32), celebrate birthdays (1.133.1– 2), engage in deliberations and social interaction (1.133.3– 1.134); they pursue pleasure and war and raise large families (1.135– 1.137.1). When they sacrifice, they include a prayer on behalf of the king (132.2). Honor to the king represents a portion of their busy lives, and monarchy itself is a part of a social structure, marked by a hierarchicalism that is stereotypically “oriental” but at least suggests a distribution of rights according to status (1.134.1). What the history represents as prerogatives of the Persian king (the prostration and the birthday banquet, elsewhere called tukta) are here cultural forms with broader applications.49 Conversely, the king is subject to rules of fair dealing with his inferiors, according to criteria that also apply to ordinary Persians.50 The equivalence between the relationship of master and servants, on the one hand, and that of king and subjects, on the other (1.137), implies the same metaphor of political enslavement that we find in the history. But the history tends to emphasize the overbearing and punitive actions of the king, his violation of established custom in various spheres,51 his subjects’ service to him, and the powerlessness of the Persians in general, regardless of their 48. For Atossa, see 3.134.5; cf. 5.12– 13 (Darius likes the prospect of skilled Paeonian servant-women). For Mardonius, see 7.5.3; cf. 7.8α2 (Xerxes), 9.122.2 (Artembares wishes for a land better than the Persian). 49. Cf. Egyptian salutations (see 2.80.2). For proskunesis to the Persian king, see 3.86.2, 7.136.1. On the tukta, see the ethnographic gloss at 9.110.2. 50. See Briant 1988, 85. 51. The Persian nomos not to inflict grave penalties at the first offense or before balancing offenses and benefactions is followed only once (7.194.2; see 3.127 for a reverse case) and broken several times by the king in the history. See Lateiner 1989, 153. See 3.35, 36; 4.84; 7.38– 39; 8.118 (though this story is refuted on different grounds). For royal violations of the Persians’ traditional reverence for rivers (1.138.2, 1.131.1), see especially 1.189, 7.35. Xerxes’ mutilation of dead Leonidas is explicitly glossed as a violation of  Persian custom (παρενοµησε, 7.238.2).

154

Telling Wonders

rank.52 This description, by contrast, represents the assertiveness of the ethnos as a whole and places it firmly in control of its own nomoi. As it downplays the monarchical code, the Persian ethnography also leaves out all references to nomoi that involve abusing the body.53 Impaling and castration are not mentioned. Cutting off the ears and nose of a wrongdoer, flaying, and other punitive practices are adumbrated by the antiseptic

ηκεστον  euphemism αν πα θος, “incurable harm,” in the section that describes the Persians’ judiciousness in disciplining their slaves or subjects.54 In the history, suffering mutilation even appears as a sign of valor, from the actions of Zopyrus and Boges and from the Persian soldier’s fascination with the horrible wounds of the Aeginetan Pytheas.55 In the ethnography, Persian andragathie simply means “fighting well” and producing many sons (1.136.1). Human sacrifice, attributed to the Persians on several occasions in the history, is out of the question.56 The narrative even omits the moment of the victim’s slaughter in the context of animal sacrifice. We find instead a description of the sacrificer’s cutting of the meat in tiny pieces, which are fastidiously arranged on a soft bed of grass.57 Among the vulgar stereotypes that are surprisingly transformed is the barbarian addiction to wine: Herodotus’ Persians put it to good use in deliberations.58 To the 52. See especially the episode of the anonymous Persian at Attaginus’ banquet (9.16), which has also the function of dissociating the community of the Persians from the actions of the king, as Corcella (1984, 181) remarks. 53. In his Persian ethnography, largely based on Herodotus, Strabo finds it necessary to add the following item: “They are governed by hereditary kings; he who disobeys has his head and arms cut off, and his body is thrown off” (15.3.17). The funeral custom of the magi, which includes a ritual mutilation of the corpse (Hdt. 1.140), will be discussed later in the present chapter. 54. 1.137.1. In the Histories, mutilations are first and foremost terms of the monarchical code, not merely Persian or barbarian. See Hartog 1988, 332– 34. Passages mentioning royal Persian mutilations include 3.69.5 (cutting off of ear); 5.25 (slaying and flaying); 7.35.3, 8.90.3 (decapitation); 7.39.3 (cutting in half); 7.194.1– 2 (crucifixion, stayed); 7.238 (decapitation of corpse and impaling); 9.112 (mutilations by queen); 3.16, 3.27.3, 3.29, 3.30, 3.31.1, 3.32, 3.35 (mutilations by Cambyses). Mutilations are also attributed to other Persians, not necessarily acting on behalf of the kings (3.79.1, 3.118.2, 3.125.3, 6.30.1, 6.32). 

ηκεστος).  55. For Zopyrus, see 3.154.2 (called a λωβη αν For Boges, see 7.107.2. For Pytheas, see 7.181.2– 3. Cf. the andreie of the medizer Hegesistratus (9.37.2). 56. See 7.114, with an ethnographic gloss inferring that burying people alive is a Persian custom. See also 7.180, 1.86.2. 57. See 1.131.3. The omission is exceptional. See Gould 1994, 99. 58. See 1.133.3– 4. In other texts, the stereotype of barbarian drunkenness applies especially to Thracians and Scythians (see Hall 1989, 133– 34; chap. 2, n. 208 in the present book), but it fits in with the notion of barbarian intemperance. Cf. Cambyses at 3.34.1– 3. The Persian custom in the ethnography recurs among Tacitus’ idealized Germans (Germ. 22). See How and Wells 1928, 1:114.

Interpretation and Evaluation

155

amazement of Plutarch, the notorious Eastern use of eunuchs is here replaced by pederasty in the Greek style.59 The near absence of expected indices of barbarity (torture, despotism, lack of restraint), the attribution to the Persians of a strong collective voice in defense of their nomoi, and the intrinsic righteousness of some of the nomoi contribute to create a “Good” impression of the culture as a whole. The negative side of the evaluation is then only conveyed through the symbols and signs of imperialism, acquisitiveness, and material abundance. Among these, the banquet represents a key event in Persian culture, and Herodotus’ description of Persian meals plays an important role in representing the Persians historically.60 The Persians are, as the present tense of the ethnography describes them from the time of their conquest of Lydia. Previously they used to be, according to Croesus, hubristai (arrogant/violent) by nature, but poor (1.89.2). In the words of Sandanis (1.71.2– 3), they used to dress in leather and had neither wine, nor figs, nor any other good thing (all negatives of true lack). The narrator’s ethnographic-historical gloss corroborates this speaker’s assessment: “before the conquest of Lydia, the Persians had nothing good and no luxury  ουτε  ουδ  αγαθ

 αβρ  ον

εν]”  at all [ουτε ον (1.71.4). In the ethnography, which records their definitive cultural forms, the Persians are elegantly dressed (1.135). On their birthday, wealthy men serve a cow or a horse or a camel or an ass, cooked whole in the stove; a variety of side dishes; and a great deal of wine (1.133.3) If in the positive sides of their culture, the Persians appear as good as or better than the Greeks, Herodotus’ representation of Persian wealth and acquisitiveness implicates the Greeks explicitly in two different contexts. 59. See Hdt. 1.135. Plutarch writes (De Malign. Herod. 13 ⫽ Mor. 857C): “How can

the Persians owe the learning of this intemperance [ακολασ ι ας] to the Greeks, when practically everyone recognizes that this people has practiced castration of young boys before even seeing the Greek sea?” See Hall 1989, 157, for this stereotype. In Herodotus, the existence of a market for eunuchs in Persian-dominated Asia Minor is acknowledged in the ethnographic gloss at 8.105.2, which does not, however, connect it with sexual practices. Another orientalistic feature that Herodotus fails to attribute to the Persians is the (apparently historical) practice of consanguineous marriage (see Dissoi Logoi, DK 90 2.15), which in the Histories only appears as one of the monarchic perversions of Cambyses (see 3.31). See Mora 1985, 165– 66. 60. In the historical narrative, banquets mark crucial moments at the beginning of Persian history (see 1.125– 126); at the end of their imperialistic dream, with the defeat of Plataea (see 9.82); and in the disastrous aftermath at Xerxes’ court (see 9.110.2). For other instances, see 1.207.6– 7 with 1.211, 3.79.3, 7.119, 7.135.1. In Thucydides (1.130.1), the orientalized Pausanias “keeps a Persian table.”

156

Telling Wonders

The Greek contribution of pederasty to the range of Persian pleasures appears in a list of importations from abroad that recalls the celebrated influx of foreign comforts to Athens after that city’s rise to imperial status.61 Though the Persian banquet is still an index of the discrepancy between Persian and Greek cultures, “the Persians say that the Greeks stop eating when they are still hungry, because after the main meal nothing else to speak of is served to them, and if it were served, they would not stop eating.”62 Just as the Persians, who now eat what they want, used to eat not what they wanted but what they had (1.71.3), so the Greeks of Herodotus’ present either had first, now have, or still want the material pleasures the Persians have acquired. Positioned precariously between contrast and similarity, the Greeks partially share in the strength and the vulnerability Herodotus’s ethnography attributes to the Persians from the time of their conquest of Asia to that of their defeat at Plataea. Dispassionate Narrative and the Limits of Relativism When the metanarrative does not intervene to express doubt or mention a source, ethnographic descriptions rely on a presumption of autopsy.63 The position of Herodotus with respect to the ethnographic material he presents is analogous therefore to that of the author of a modern ethnographic report. The facts are external to the text, and the narrator is not responsible for them. The problem is, then, how things are said more than what is said. How does one produce a culturally unbiased—that is, nonsensational and nonjudgmental—description of a foreign culture whose customs are inherently bound to produce a “Bad” impression? One solution is to avoid mentioning such features, and I have argued that the Persian ethnography is somewhat selective precisely in this sense. But this is not a practice an ethnographer can resort to throughout a work. The question of how things are said is raised by the argument of Hartog that the objective, dispassionate, nonevaluative style of discourse that Herodotus generally favors when describing foreign culture is in fact one of the many tricks of his trade, a part of his “rhetoric of otherness.”64 We should then consider other style options. In the context of a discus61. 1.135. Cf. Thuc. 2.38; [Xen.] Ath. Pol. 2.7. 62. 1.133.2. Cf. 9.82 with Thuc. 1.130.1– 2. 63. In ethnographic descriptions, hearsay plays only a very limited role. See, e.g., 1.183.1– 3 (description of Babylon); 4.176, 4.178, 4.183.4, 4.184.3, and 4.187.3 (ethnography of Libya). 64. Hartog 1988, 256.

Interpretation and Evaluation

157

sion about the possibility of our understanding cultural forms that are alien to ourselves, Clifford Geertz quotes L.V. Helms’ description of human sacrifice in Bali, which we may take as a modern term of comparison with Herodotean discourse.65 It is a striking and elegant passage, but since my specific purpose is to make a point about biased and unbiased narrative, I regrettably quote only a few excerpts, with not much regard for the continuity of the action described or the integrity of the account. 1. While I was in Bali one of these shocking sacrifices took place. The Rajah of the neighboring State died on the 20th of December 1847; his body was burned with great pomp, three of his concubines sacrificing themselves in the flames. It was a great day for the Balinese. It was some years since they had had the chance of witnessing one of these awful spectacles, a spectacle that meant for them a holiday with an odour of sanctity about it. . . . 2. They looked little enough like savages, but rather like a kindly festive crowd bent upon some pleasant excursion. The whole surroundings bore an impress of plenty, peace, and happiness, and, in a measure, of civilization. It was hard to believe that within a few miles of such a scene, three women, guiltless of any crime, were, for their affection’s sake, and in the name of religion, to suffer the most horrible of deaths, while thousands of their countrymen looked on. . . . 3. The victims of this cruel superstition showed no sign of fear at the terrible doom now so near. Dressed in white, their long black hair partly concealing them, with a mirror in one hand and a comb in the other, they appeared intent only upon adorning themselves as though for some gay festival. The courage which sustained them in a position so awful was indeed extraordinary, but it was born of the hope of happiness in a future world. From being bondswomen here, they believed they were to become the favourite wives and queens of their late master in another world. . . . Round the deluded women stood relatives and friends. Even these did not view the ghastly preparations with dismay, or try to save their unhappy daughters and sisters from the terrible death awaiting them. Their duty was not to 65. Helms 1882, 59– 66, quoted by Geertz (1983, 37– 39).

158

Telling Wonders

save but to act as executioners; for they were entrusted with the last horrible preparations, and finally sent the victims to their doom. . . . 4. The women were carried in procession three times round the place, and then lifted on to the fatal bridge. There, in the pavilion which has been already mentioned, they waited until the flames had consumed the image and its contents. Still they showed no fear. . . . Meanwhile their attendant friends prepared for the horrible climax. . . . The supreme moment had arrived. With firm and measured steps the victims trod the fatal plank; three times they brought their hands together over their heads, on each of which a small dove was placed, and then, with body erect, they leaped into the flaming sea below, while the doves flew up, symbolizing the escaping spirits. . . . This terrible spectacle did not appear to produce any emotion upon the vast crowd, and the scene closed with barbaric music and the firing of guns. This autobiographical narrative in the past tense describes a particular occurrence of a customary, if infrequent, event. The features that make Helms’ narrative the very antithesis of the altogether neutral fashion that Hartog rightly attributes to Herodotus, however, could just as well characterize third-person descriptions in the iterative ethnographic present. Even when we see no grammatical first person, the narrator, Helms, is pervasively there to direct the reader’s perception of it and to communicate his ideological stance.66 Evaluative modifiers occur at every turn, from “shocking sacrifices” in excerpt 1 to “barbaric music” in excerpt 4. What pretends to be the report of the native attitude is actually the narrator’s distanced interpretation of it. “A holiday with an odour of sanctity about it” (1) and “their duty was . . . to act as executioners” (3), for example, in no way reflect what “the Balinese say.” Actual native 66. In case there should be any doubt left, Helms’ narrative ends with an advertisement of narratability followed by a gloss of interpretation and evaluation: “It was a sight never to be forgotten by those who witnessed it, and brought to one’s heart a strange feeling of thankfulness that one belonged to a civilization which, with all its faults, is merciful, and tends more and more to emancipate women from deception and cruelty. To the British rule it is due that this foul plague of suttee is extirpated in India, and doubtless the Dutch have, ere now, done as much for Bali. Works like these are the credentials by which the Western civilization makes good its right to conquer and humanize barbarous races and to replace ancient civilizations.”

Interpretation and Evaluation

159

beliefs concerning the meaning of the event are judged to be absurd in the same breath as they are reported (3: “cruel superstition,” “deluded women”). Negative sentences signal amazement, especially with regard to the discrepancy between civilized appearances and the abnormal savagery of the ritual (2, 3, 4). Among the stylistic means that are available to Herodotus, the choice is between an evaluative and a neutral description.67 Herodotus prefers the latter form, which reproduces the style generally employed by the Hippocratics in their account of symptoms, but which eluded one of these authors, who undertook to discuss foreign peoples. It is more important to verify the extent to which a neutral ethnographic description fails to be truly neutral and why it does so in particular cases than automatically to rank objectification among the various devices for enhancing otherness. In Herodotus’ description of what the Scythians do with the skin of the enemies they kill in war (making napkins, cloaks, and the like), we find the following gloss.  ην αρα,  

  δερµα δ ε ανθρ ωπου κα ι παχυ κα ι λαµπρον σχεδον   δερµατων πα ντων λαµπροτατον λευκοτητι. [and, as a matter of fact, human skin turned out to be both thick and bright, almost the brightest in whiteness of all skins.] (4.64.3)  Two words here deserve attention. First, the particle αρα (as a matter of fact) expresses the sudden interest of an unexpected but revealing datum.68 Second, the imperfect ην (was, turned out to be) interrupts the string of ethnographic presents. To what time does ην refer? Evidently it  refers to the time of “fieldwork.” The construction ην αρα is a gloss of historie that places the narrator on the scene, inspecting the local crafts, as he brings his audience’s attention to the result of his observation. Somewhat analogous in tone to Herodotus’ implicit display of firsthand expertise with regard to garments made of hemp (4.74), this gloss is nevertheless bound to produce a different rhetorical effect, perhaps enhancing, rather than toning down, the sensational character of the description. It advertises the deliberately objective stance of an ethnographer who is 67. The type of discourse that puts this problematic on display, as in what Geertz (1988, 97) calls “author-saturated texts,” is of course not an option. 68. See especially Denniston 1934, 32– 43; Corcella 1993, 286. Herodotus’ nonevaluative description is equivalent to “minimally narrated narrative.” See chap. 1, “What Is Metanarrative?”

160

Telling Wonders

determined to eschew cultural bias and to apprehend a foreign universe on its own terms; but in the face of this data, he is bound to appear too determined. There are countless passages of this sort in Herodotus. In them, the absence of an ethical viewpoint in a detailed but compressed style of narration conveys an almost cheerful detachment of the narrator vis-a-vis ` his 69 subject. Because “no mode of composition is a neutral medium,” Herodotus’ professional disengagement, either voluntarily or involuntarily, sometimes produces both horror and humor.70 In other cases, however, the discourse attaches the factual description of a practice to its proper ideological motivations by means of cognitive statements, native reported speech, or the direct portrayal of native attitudes. This type of description is also likely to communicate a specific criterion for judgment. Exemplary in this respect is the account of a Thracian widow slaying (5.5), the ritual in Herodotus that most resembles the Balinese suttee described by Helms. Herodotus’ discourse builds up an animated momentum by sectioning the narrative by prospective sentences that increase expectation but withhold praise or blame. 1. And those above the Crestonaeans do this: Each one has many wives; when a man dies, there is a great competition among his wives, with intense pleading of friends and relatives on the following issue: which one of them was most loved by her husband. The report of the results of the competition continues to emphasize the enthusiastic adherence of all involved. 2. And the one who is chosen and awarded this honor, after receiving the praises of both men and women, is slain over the grave by her closest male relative and, having been slain, is buried with her husband. Finally, a cognitive statement reports the state of mind of the survivors. 69. See, e.g., the account of the Scythians’ human sacrifice to Ares (4.62.1– 4), with the last chilling sentence, “and there’s the arm, lying where it has fallen, and, somewhere else, the body.” 70. Rosaldo 1989, 46– 52, especially 49. For a genre of ethnographic writing that exploits humorous representation, see, e.g., Barley 1992, especially 74– 75.

Interpretation and Evaluation

161

3. The other wives consider it a great calamity; for this is for them the greatest shame.   µεγαλην ποιευνται᝽   [α ι δ ε αλλαι συµ␸ορ ην ονειδος γαρ σ␸ι  µεγιστον  τουτο γι νεται.] The sequence of joyful actions in passages 1 and 2, suddenly landing on the chilling predicate “is slain” [σ␸αζεται], strikes a sensational note (cf. 1.45.3). But all three sections convey the natives’ lack of ambivalence, without the incredulity Helms demonstrates at the apparent acceptance of the victims and their relatives. Although lack of pathos and even a comic element are here once again the price of detachment, Herodotus’ narrative implements two serious principles of ethnographic writing. It focalizes the custom through their owners, and though, unlike Helms’ description, it does not explain the native perception in light of existential beliefs, neither does it undermine or contest it. The case just examined communicates that a nomos that the ethnographer’s culture considers inhumane or unholy can be integrated into a relativistic view if it is internalized by all those whom it affects. Herodotus’ report about a custom of the Massagetae conveys the same message. No other limit of human life is set for them. But when one becomes very old, all his relatives assemble and sacrifice him and other animals together with him, and after boiling the meat, they feast upon  it [κατευωχεονται]. This is considered by them to be the most  σ␸ι νενοµισται],    τα ολβι ωτατα  blessed thing [ταυτα µ εν but if one dies of disease they do not feed upon him but rather bury him in the  ποιευµενοι]  ground, considering it a misfortune [συµ␸ορ ην that he did not reach the age of being sacrificed. (1.216.2– 3) The initial negative statement has the corrective function of dispelling the audience’s possible confusion between the Massagetan custom and more radical practices of other peoples.71 The combination of the code of   sacrifice (θυουσι, τυθηναι) with the secular term κατευωχεονται is jolting; but the statements about what this community regards as “most 71. It parallels the correction at 1.216.1. Asheri (1988, 386) cites the (explicitly evaluative) accounts by Strabo (11.11.3) and Aelian (VH 4.1) about cultures that fix the limit of human life at seventy or sixty years of age or kill the elderly or the sick in different ways (starving them, throwing them to the dogs, etc.).

162

Telling Wonders

  overcome the ideoblessed” [ολβι ωτατα] or a “misfortune” [συµ␸ορην] logical instability of the description by referring to a peculiar but deliberate and shared worldview of this society.72 The criterion of evaluation implicit in these passages is consistent with Herodotus’ explicit condemnation of ritual prostitution at Babylon (1.199.1) and with the implied disapproval of customs in which one portion of the society victimizes another. The Indian tribe of the Padaeans was particularly savage, according to ancient tradition, though predictably no statement to that effect appears in Herodotus.73 He merely calls them “eaters of raw meat” (3.99.1) and applies to their particular brand of cannibalism the same form of discourse employed for the Thracian and Massagetan funeral customs—a neutral description from the native point of view. When one of the townspeople gets sick, be it a man or a woman, if it is a man, the men who are his closest kin kill him, saying that as he wastes away with the disease, the flesh becomes spoiled for them. The other denies being sick; but hearing no reason, they kill him and   feast upon him [αποκτε ι ναντες κατευωχεονται]. And if a woman gets sick, the women who are closest to her do with her just as much as the men do with the men. For when one comes to old age, they  γαρ   δ ε ε ς γηρας απικ  sacrifice him and feast upon him [τον οµενον   θυσαντες κατευωχεονται]. But not many people reach this theoretical point, for before this, they kill anyone who falls sick. (3.99.1– 2) The final statement, with the narrator’s most intrusive evaluation, is the counterpart of the negative gloss introducing the Massagetan custom (1.216.1: “No other limit of human life is set for them”). In the immediately preceding sentence, the verbal correspondence establishes a parallel  . . . γερων   with the Massagetan ritual (cf. 1.216.2: ε πεαν γενηται     καρτα, . . . θυουσι µιν, . . . κατευωχεονται); this renders the differences between the two practices all the more conspicuous. The Massagetae sacrifice and eat the very old but do not eat those who die of disease, 72. They recall how unconventional Solon’s use of these terms appeared to Croesus (see especially 1.32.2– 9): see Flory 1987, 97. The phrase “to consider a misfortune” recurs at 5.5, quoted earlier, and at 4.79.5, in cognitive descriptions of native attitudes. 73. How and Wells (1928, 1:99) and Asheri (1990a, 326) suggest the derivation of the ethnic name from the Sanskrit padja, “bad,” and cite Tibullus 4.1.144– 45, testifying to the negative reputation of this tribe.

Interpretation and Evaluation

163

while the Padaeans kill and eat a person as soon as he or she gets sick, which means that few Padaeans reach old age. The killing of the elderly, both among the Massagetae and the Padaeans, is called a “sacrifice”   [θυειν], but the Padaeans’ killing of the sick is a “killing” [αποκτε ι νειν]; the secular verb κατευωχεισθαι occurs conjoined with both terms. Whereas for the Massagetae the dietary aspect of the custom is subordinate to its religious character, the reverse happens among the Padaeans. All the Massagetae agree that the sacrifice and the eating are a “blessed” thing, although as in the case of the Thracian widow slaying, we are not told on what grounds.74 The utilitarian/dietary ideology of the Padaean practice, by contrast, produces in their society a split between those who are killers and those whose turn it is to be victims, between those who benefit from the sacrifice and those who perceive they do not. The native voice is divided, as the typical dialogue embedded in the description testifies: the healthy say that the sick must be killed in a hurry, and the sick protest in vain that they are not sick. Herodotus’ cultural relativism is not ethical relativism in an extreme sense. Just as the moralistic historian objects to monarchical abuse, so the ethnographer signals disapproval of customs that oppress a society’s free members.75 This ethical principle is, however, supracultural and capable of accommodating a wide range of diverse practices and perceptions. If difference is not always “Good,” Herodotus’ logos is more concerned in establishing that it is not automatically “Bad.”76 This, the legitimacy of cultural difference as such, is his fundamental lesson as an ethnographer. One of the means through which he conveys it is his commitment to a nonevaluative mode of ethnographic description. Equal Knowledge: Cognitive Relativism Herodotus’ acceptance of the validity of foreign nomoi and native attitudes is subject to certain judgmental reservations in particular cases. In 74. Contrast the explanation given by the Dissoi Logoi (DK 90 2.14). See Mora 1985, 164. 75. See the evaluation implicit in the juxtaposition of Scythian royal funerals (4.71– 72) and ordinary Scythian funerals (4.73). 76. On Herodotus’ relativism, see Burkert 1988, 27– 28 and later discussion in the present chapter. I will apply the term relativism somewhat loosely to fifth-century thought (see Guthrie 1971, 164– 75; Kerferd 1981, 83– 110). See however the objections put forward by Bett (1989). On the modern controversy between relativism and antirelativism, and on the difference among different relativistic positions, see Geertz 1984.

164

Telling Wonders

the sphere of religious beliefs and beyond ethics, however, his cognitive relativism is more radical, theorized a priori, and often conveyed through the deliberate suspension of his normal activities of inquiring and narrating.77 As we have seen in the cases of the Thracian widow slaying and Massagetan cannibalism, Herodotus rarely describes the religious beliefs underlying customs. When he does, he is usually pursuing a polemic against the Greeks (e.g., at 1.131.1). His discussion of Salmoxis-Gebeleizis compensates for the ambivalence of his own description of the Getic ritual by exposing the more flagrant chauvinism of the Black Sea Greeks.78 After rejecting their version, which reduces Getic religion to an inferior byproduct of their own civilized culture, Herodotus quickly brings the section to a close: “Whether Salmoxis was a man or whether he is some local  Getic divinity, farewell to him [χαιρ ετω]” (4.96.2). Given the particular context, this rare form of programmatic conclusion recalls the formulaic hymnal farewell to the gods. The dismissal of the topic as such encodes a sort of agnostic acceptance of the religion.79 We have already seen how the first programmatic introduction of the Egyptian ethnography (2.3.2) connects Herodotus’ reticence to explain religious beliefs to his relativistic position.80 Here, his profession of belief (νοµι ζων) that “all men know equivalently about these things” [παντας   ισον περ ι αυτ  ων  ε πι στασθαι] goes beyond the evidence that ανθρ ωπους components of the different religions of the world can often be translated from culture to culture.81 Rather, it means that, even regardless of similarities or overlaps, the degree of accuracy possessed by the whole body of theological beliefs of one society is equal to the degree of accuracy possessed by the whole body of theological beliefs of another. When Herodotus attributes an equivalent value to the religious knowledge of different cultures, from the monotheistic Getae to the complicated Egyptians, this of course includes the idea that divine matters “cannot be   η and γνωµη  objects of enquiry ( ι στορι η) by οψις, ακο and, hence, can77. On cognitive relativism and related attitudes in modern thought, see Hanson 1979. 78. On Herodotus’ ambivalent evaluation of the Getae, see “Explicit Evaluation” earlier in this chapter.  79. Cf. Hom. H. 1.20, etc. The only other conclusion with χαιρετω occurs at 2.117 (see Lateiner 1989, 63), but the dismissal that is closest in function to 4.96.2 is at 1.140.3, quoted later in this chapter. On programmatic conclusions, see chap. 1, “What Is Metanarrative?” 80. See chap. 2, “The Texture of Nomos.” 81. This evidence exists, e.g., in the case of major divinities, worshiped by different peoples with different names. See Hartog 1988, 107 and n. 162, quoting Veyne 1971, 141.

Interpretation and Evaluation

165

not be objects of certain knowledge.”82 This is again too reductionist a statement, however, if by it one means that for Herodotus the real state of affairs is entirely out of reach. It is not Herodotus but Protagoras who denies the possibility of human knowledge concerning the gods. Consider  the first sentence of Protagoras’ Περ ι θεων. About the gods, I am not able to know that they exist or do not  θεων  ουκ  ε χω ε ι δ εναι,  exist or what shape they have [περ ι µεν  ε ι σ ι ν ουθ  ουκ  ως  ως  ε ι σ ι ν ουθο   ποιοι τινες ι δ εαν],  ουθ because many are the factors that prevent one from knowing, such as their invisibility and the shortness of human life. (DK 80 B4) The indirect questions in this passage bear resemblance to those Herodotus formulates in another interpretive gloss (2.53): here he asserts,  on his own authority (ε γω λεγω), that before the codification of knowledge recently effected by Homer and Hesiod, the Greeks “did not know whence each of the gods came into existence, whether they were forever, !  !  θεων,  and what kind of shape they had” [ οθεν δ ε ε γενοντο εκαστος των 83 "             ε ιτε α ιε ι ησαν πα ντες, οκοιο ι τε τινες τα ε ιδεα, ουκ ηπιστεατο]. This statement is designed to cut Greek theological representations down to size rather than to proclaim their validity as real knowledge.84 At 2.3.2, however, Herodotus generalizes in positive terms: all men know ison about divine things. They all “really know” something, and they all know an (indeterminably) equal amount.85 As an inquirer of the past, Herodotus is able to identify actions that appear attributable to something other  the than human planning or random contingencies—το θει ον, ο θεος, divine in purely theological terms—with no need or possibility to be more specific.86 At the same time, however, he also verifies the existence of a line of communication between human beings of all nations and god (as an ontological principle transcending culture) that passes through cultural channels, such as prayer, sacrifices, and oracles. A space that is 82. Lloyd 1976, 17. 83. 2.53.1. Burkert (1988, 26) cites the fragment by Protagoras as a parallel to this passage; Lloyd (1976, 17) cites it as part of the intellectual background of 2.3.2. Both times,  Herodotus’ ε πι στασθαι, in lieu of Protagoras’ ε ι δεναι, places less emphasis on the sensible origin of the knowledge. See Untersteiner 1967, 1:67 n. 37. 84. Cf. 7.129.4. See Burkert 1988, 20– 22. 85. In positive terms, they know not “equally little” (Stein 1883, I.2.6) but “equally much” (see, e.g., Grene 1987, 666). For a different rendering, see Mora 1985, 136– 39. 86. See Linforth 1928.

166

Telling Wonders

sacred by nomos is sacred in absolute terms; the god whom “they” worship in the shape of a bull is objectively divine.87 The objectivist component of Herodotus’ theological relativism is a paradox that preserves traditional Greek polytheism, reinterprets its openendedness, and places foreign religions on a par with it. It also determines the narrator’s reluctance to take on τα θεια as a part of his ethnographic subject. To inquire and explain is to question and therefore, potentially, to negate. The limiting case of perverted theological inquiry is Cambyses’ brutal practical test of the divinity of the Apis bull (3.27– 29). In matters of religion, however, even verbal inquiry, which goes hand in hand with a sort of arbitration, is at least dangerous. When the narrator displays his own historie in the secular sphere, contradictory logoi that are pitted against each other result, implicitly or explicitly, undermined; indeed, in such cases, to expose their subjectivity and the instability of truth is for Herodotus an important goal. In a somewhat parallel fashion, to discuss the hieroi logoi (sacred history) of different peoples will raise questions concerning their correctness, absurdity, or impropriety, as well as the intellectual and cognitive skills of the people who hold them as true. Though at very different levels, both the vulgar account of the Black Sea Greeks concerning Salmoxis and the philosophical polemic of a Xenophanes of Colophon are deconstructive criticisms of culturally determined religious beliefs. Per of Protagoras had put accounts of different theologies haps the Περ ι θεων 88 to a similar use. As the promoter of the validity of difference, Herodotus sometimes needs to shake his audience’s self-assurance in their own religious traditions or justify those of others; precisely these motivations, more often than not, provide the “overwhelming compulsion” (see  ε ξαναγκαζοµενος in 2.3.2) to discuss sacred knowledge.89 But to subject foreign hieroi logoi to that kind of scrutiny, to expose them to ethnocentric debate, sophistic skepticism, and perhaps ridicule—this, as a rule, he is   . . . προθυµος] “not eager” [ουκ to do (2.3.2). 87. For sanctuaries, see, e.g., 9.65.2, with a cautionary gloss of opinion. A gloss of comparison (at 3.64.3) implicitly presents Cambyses’ death as evidence of the divinity of Apis. 88. Philostratus (DK 80 A2) makes Protagoras’ statement derive from a conversation with the Persian Magi; see Untersteiner 1967, 1:56. For a denial of the possibility of knowledge about the gods, cf. Xenophanes, DK 21 A34. 89. See 1.131.1, 2.43– 44 (concluded with an apology to gods and heroes at 2.45.3), 2.143– 46. For the different case of 2.156.4– 5, on the myth of Leto, see chap. 4, “Herodotus and the Conventional Code.” For the myth at 2.42.3– 4, see chap. 2, “The Texture of Nomos.”

Interpretation and Evaluation

167

Funeral Customs and Other Nomoi: Cultural Relativism The only mutilation tentatively attributed to the Persians in the ethnography is the practice of exposing the dead to be torn by a bird or a dog. After this treatment, which recalls the Homeric indignity against the bodies of one’s enemy, the Persians cover the body in wax and bury it.90 Clearly marked off by a retrospective/prospective system (1.140.1), this custom is deliberately not integrated into the rest of the description, and it is not allowed to contribute to an understanding of Persian ideology or beliefs. The ritual simply exists, either among the Magi or among both Persians and Magi, and it is different, just as the Magi are unaccountably different from the rest of mankind also in other respects (1.140.2). One sign of the custom’s unfathomable arbitrariness is that the narrator, some what as he does for Salmoxis, bids it farewell: “let it be [ε χ ετω], just as it has been established as a custom to begin with” (1.140.3). This ritual appears as the first of the funeral customs in the Histories; it constitutes the limiting and representative case of them all—opaque, arbitrary, of unknown origin, and a symbol of difference, yet not entirely unique. Only one other ethnos in the Histories abandons the dead to corruption (3.100), but many embalm and bury. Similarly, only one ethnos celebrates joyfully when someone dies (5.4), and only one (5.8) holds athletic contests, though that is the old heroic manner. Several eat or cremate the dead or make their skulls into objects; several sing dirges, sacrifice, or mark tombs.91 Individual features recur again and again in different combinations, revealing both people’s mysterious opportunity for difference and the limit of that difference. And since each ethnos is bound to do something—which is always the same ritually prescribed 90. Mutilation of corpses by “dogs and all the birds” in Homer (Il. 4– 5; cf. Il. 2.393, 4.237) is not value-free. See Segal 1971, 9– 17. Cf. Soph. Antig. 204– 6, 258– 59. Boedeker (forthcoming) contrasts Hdt. 1.140 with Artabanus’ “Homeric” threat to Mardonius at Hdt. 7.10θ3. In this case, the discrepancy between Herodotus’ Persian ethnography and history (see discussion earlier in the present chapter) follows an inverted pattern. 91. Embalming (with different techniques) is performed by the Babylonians (1.198), the Egyptians, the Ethiopians (3.24), and the Magi. Burial is practiced by the Magi, the Persians, and the Massagetae (1.216.2), the Libyan nomads, the Nasamones (in a sitting position, 4.190), and the Trausians (5.4). Eating the dead occurs among the Massagetae (1.216.2), the Callatians (3.38.4), the Padaeans (3.99), and the Issedones (4.26.1). Cremation occurs among the Greeks and the Lydians. Skulls are made into objects by the Issedones (skulls of loved ones, 4.26.2) and the Scythians (skulls of enemies, 4.65.2). Dirges are sung by the Babylonians, the Egyptians (1.198), and the Thracians (5.8). Tombs are used by the Greeks and the Nasamones (4.176).

168

Telling Wonders

act—with their dead, funeral customs are inevitable signs of culture. They belong to every society’s body of most compulsive norms—the ι ρα   in the proper τε κα ι νοµαια—and mediate between what is sacred ( ι ρον) sense, connected with the cult of the gods, and what we would more loosely call “sacred.”92 A society’s ritual disposal of dead bodies has no bearing on relations with others, and it does not involve issues of justice or injustice toward its own members in the same way as do customs that have to do with the treatment of the living. In Herodotus, therefore, funerals are an important cultural symbol. They illustrate the sense of alienation that foreign cultures inspire simply as they go about their business of being different from ourselves; consequently, they allow the ethnographer to theorize on cultural relativism in the broadest possible terms. This happens in a far-ranging interpretive gloss that, standing at the intersection of ethnography and history, generalizes on the basis of the narrative about Cambyses’ behavior toward custom (3.38). Cambyses is the extreme representative of Herodotus’ negative paradigm for historical and ethical action, the monarchical model. He is also a scientist of sorts, a researcher of nomoi, and in this capacity the foil of the histor of the Histories.93 Their shared field of observation is principally Egypt, a fundamental source of theoretical and anthropological learning for Herodotus, and a land Cambyses oppresses as well as observes. Through his brutal testing in the sphere of custom—from religion, to funeral procedures, to marriage laws (3.16, 31, 35.5)—Cambyses attempts to find out whether practices, cultural beliefs, statutes, public opinion, and common morality will hold out under skeptical critique, objective scrutiny, and external force. Herodotus’ own respectful inquiry on the value of nomos here takes advantage of the historical case of Cambyses himself. To Herodotus, Cambyses represents especially useful evidence for two reasons. First, his violations of the nomoi of others are strictly connected with his destruction and deconstruction of less culture-specific ethical rules and of the laws of his own society. In symbolic terms, the fulfillment of his early promise of putting everything upside down in Egypt corresponds to the unprecedented upside-down burial of his own Persian sub92. Heinimann (1945, 79) gives nomos as applied to these practices the meaning of fas. In the taxonomy of ethnographic descriptions, funeral customs are sometimes adjacent to religion, other times not. 93. See Munson 1991, 59– 60; Christ 1994, 186– 87.

Interpretation and Evaluation

169

jects.94 Second, Cambyses’ madness is not a hyperbole for criminal behavior but a clinical (i.e., physical) illness.95 This means that Cambyses’ dysfunction with respect to all sorts of customs—foreign and native, religious and secular—coincides with an impairment of the mind that clearly belongs in the sphere of phusis (nature). By Herodotus’ time, phusis, nomos, and the divine, once blurred in archaic Greek thought, have grown increasingly distinct.96 As far as Herodotus is concerned, however, Cambyses’ triple abnormality demonstrates e contrario the existence of a close connection between divine will, nature, and culture as a universal phenomenon.97 Herodotus’ conclusion at 3.38.1 begins by interpreting Cambyses’ deri sion of religion and serious customs ( ι ρα τε κα ι νοµαια [things sacred and pertaining to custom]) as evidence that confirms his madness  is here an understatement that in(ε µα νη). But “derision” (καταγελαν) vites vertical analogy between the crimes of Cambyses (accompanied by laughter: 3.29.1– 2, 3.35.3, 3.37.2) and more commonplace manifestations of contempt. In fact, the case of Cambyses will be paradigmatic for  narrator and audience: any one who makes fun (γελωτα . . . τι θεσθαι) of such things is mad (3.38.2). The positive counterpart of this insane individual is normal people in general, who simply scrutinize all sorts of customs (διασκεψαµενοι) and are attached to their own. Protagoras may ultimately be the source for the interpretive hypothesis at 3.38.1 that if “all men” were able to choose from a display of nomoi, they would choose their own, believing them to be the most beautiful/  καλλι στους). In a very similar passage of the sophistic honorable (τους Dissoi Logoi, the idea has been put to the service of a “different strokes for different folks” brand of relativism that denies the absolute validity of  and the shameful (το the ethical concepts of the honorable (το καλον) 94. 3.3.3, 3.35.5. See also the mutilation of the body of Amasis, by which Cambyses violates both Persian and Egyptian customs (3.16). 95. Explicit references to Cambyses’ medical madness occur at 3.30.1, 3.33, 3.34.1, 3.37.1, and 3.38.1. Only Cleomenes shares this distinction. See chap. 2, n. 71 and corresponding text. 96. See especially Heinimann 1945, 110– 62; Guthrie 1971, 55– 147. Cf. chap. 2, “The Texture of Nomos.” 97. Individual nomoi have of course human origin in Herodotus. See Evans 1965, 145– 46. However, the idea of Nomos as an overriding impulse to culture recalls Heraclitus DK 22 B114 (“all human laws are nourished by a single law, which is divine; for it has as much power as it wishes and is sufficient for all and is still left over”). See Heinimann 1945, 65– 66.

170

Telling Wonders

 98 Herodotus, however, is already traveling in a different direcα ι σχρον). tion: the lack of objective validity in people’s perception that their own nomoi are καλοι only goes to show that all nomoi are equally καλοι . To deride them is therefore madness. The subjectivity of all men, illustrated by the hypothetical scenario of  the display of nomoi, is next confirmed by the evidence (τηκµηριον) of the actual experiment of another royal histor (3.38.2). Darius once proposed to the Greeks that they eat their dead parents and to the Callatian Indians, who customarily eat their parents, that they burn them instead; both groups refused to practice the other’s nomos (3.38.3– 4). Other than showing again that each likes his own customs best, the exchange also dramatizes the repulsion that the harmless other is likely to inspire. The subjectivity of each party’s reaction is enhanced by the fact that the competing nomoi are here no longer the unidentified assortment of the hypothetical scenario envisioned earlier. They are specifically funeral customs, the most suitable for conveying the principle that diverging cultural norms are not α ι σχροι and, since they are all equally compelling, are inherently καλοι . Here the implied equivalence of all men’s nomoi from the point of view of their moral goodness corresponds to the equivalence of all men’s religious beliefs from the point of view of their truth-value at 2.3.3. On the basis of this equivalence, Herodotus’ next generalized statement no longer follows the distributive pattern of 3.38.1—“all men would each  !  ε ωυτων  νοµους] [εκαστοι] believe that their own customs [τους are the most beautiful.” The unified formulation borrowed from Pindar, rather, posits a single human community and, over it, a single rule.99 Thus are these things determined by nomos, and it seems to me that Pindar was right when he said in poetry that nomos is king of all.  !  νυν ταυτα  νενοµισται,  µοι δοκεει  Πι νδαρος [ουτω µεν κα ι ορθως    ␸ ησας  ποιη σαι νοµον παντων βασιλεα ειναι.] (3.38.4) In this definitive maxim, nomos comes to mean custom/law/culture as something abstracted from this or that nomos or set of nomoi. It encompasses all humankind (Androphagoi excepted: 4.106). It represents the 98. DK 90 2.18 (cf. 2.26). On the connection between the Dissoi Logoi and Protagoras, see Lasserre 1976, 73– 74; Robinson 1979, 51– 59. 99. For the meaning of Pindar frag. 169 SM in its original context, see especially Gigante 1956, 72– 102; Schroeder 1917; Stier 1928; Ostwald 1965; Humphreys 1987. Another intriguing fragment of Pindar (215 Bergk) expresses the idea of cultural differentiation.

Interpretation and Evaluation

171

universal fact of having nomoi, whatever these may be, and of behaving according to them, in culturally determined ways. At 3.38, Herodotus takes the opportunity offered by Cambyses to supplement in a crucial way the message that the ethnographies communicate to the audience. The inquiry into the customs of foreign peoples and the presentation of such inquiry to an audience navigate between the Scylla and Charybdis of ethnocentric absolutism and ethical relativism, respectively. Our ethnological observation may reinforce the sense of our  in comparison with worse communities who do monopoly over το καλον things differently than “we” do. As we have seen, the Histories include more or less automatic negative evaluations (“they couple like animals”) as well as ultimately presupposing an absolutist—and deliberate—view of what is kalon (beautiful/moral) or aischron (shameful/immoral). But though certain particular customs may indeed be aischroi (see 1.199.1), the ethnographic discourse of Herodotus attempts to steer the audience clear of unilateral chauvinism toward barbarian customs as such. A relativism according to which different nomoi are on principle equivalent in function and worth is fundamental to this lesson. For someone like Cambyses, however, the realization that different peoples have different nomoi with roughly the same validity leads to denying the validity of them all. This is the Charybdis, the second danger that Herodotus addresses in this gloss. The monarchical position of Cambyses, who obeys his own personal law and “does what he wants” (3.31.4), is in fact similar to the ethical monarchism and extreme relativism of a theoretician of the ilk of Plato’s Callicles.100 To Cambyses/Callicles, nothing is “sacred,” while Herodotus’ brand of relativism teaches that every  [Custom/ thing is. Callicles attributes the saying νοµος παντων βασιλευς Law king of all] to the law of nature that lets the stronger prevail, which is, according to him, the only law that counts (Plato Gorg. 483B). We should not doubt that the phrase served to validate sophistic thought long before the time of the Platonic dialogue. Herodotus, at any rate, uses the same phrase to support a position that is the opposite of that of Callicles and to replace Cambyses as the king- despotes (and 100. Plato Gorg. 482E– 484D. The parallelism between Herodotus’ Cambyses and Plato’s Callicles is enhanced by the continuous intrusions of the monarchical code in Callicles’ speech. Callicles’ models are the Persian kings Xerxes and Darius (483E); he speaks of strong men coming of age as “lions” (483E) and as kings’ sons (492B); he claims (491E) that they should give free rein to their desires (ε πιθυµι αι, a monarchic word in Herodotus: see chap. 2, n. 232) and that they should not impose on themselves conventional nomos as a despotes (492B).

172

Telling Wonders

researcher and destroyer of custom) with the kingship of custom and conventional law. The exchanges Herodotus describes throughout his work show to what extent both Greeks and non-Greeks—“all men,” in other words— are assiduous observers and critics of each other’s nomoi. For each ethnos of histores, the sense of allegiance to their own nomoi cannot prescind from the realization of the allegiance of others to theirs. Couched in these terms, Herodotus’ ongoing polemic against cultural chauvinism, cultural imperialism, and racism—all of which allegedly preserve a person’s attachment and obedience to his own nomoi while allowing for his contempt toward others and theirs—takes the remarkable form of an ideology that squeezes this double standard out of existence. As in the case of Cambyses, imperialistic contempt for others is madness that overrides all laws. Ethnocentrism is universal because, to paraphrase Herodotus with the words of Geertz, we “cannot escape preferring our own preferences.”101 But for the ethnographer of the Histories, morality, sanity, and piety at home as well as abroad are contingent on the acceptance of the universal and absolute rule of culture. This entails respect for the subjectivity of others on a par with ours and the belief, from an objective point of view, in the essential moral equivalence between burning and eating the dead. Thus, the notion of the abnormality of foreigners is replaced with the idea that abnormal is the one who derides their nomoi. Interpretation in the History This section focuses on those metanarrative passages where the narrator directly intervenes to explain in his own voice what an event of the past “means.” The meaning of a specific historical action or event is in most cases connected to its motives, causes, and results; an interpretation of these factors may in turn indicate the action’s worth.102 I begin by examining two fundamental glosses, both the narrative of Xerxes’ invasion, where the interpretive and evaluative functions are strictly combined. One is the famous judgment about the consequences of Athenian policy of naval resistance (7.139); the other is the explanation of Leonidas’ decision to remain at Thermopylae (7.220). These two interpretations are almost symmetrical, one about Athens and the other about Sparta, each 101. Geertz 1986, 261; cf. 257. 102. So Plutarch (De Malign. Herod.) understands Herodotus in terms of praise or blame.

Interpretation and Evaluation

173

vis-a-vis ` Persia and in contrast to other Greeks. Both statements are clearly marked by self-referential signs of opinion as interventions in the voice of the narrator. Specific Glosses of Interpretation: Sparta and Athens The first passage anticipates at long range the Salamis narrative and introduces at short range the account of Athenian deliberations before Xerxes’ invasion (7.140– 43). It is in turn programmatically introduced as a gnome, or opinion, that will cause resentment. At this point, I am compelled by necessity publicly to display an  odious interpretation [γνωµην ε πι ␸θονον] for most people, but   I shall not still, since it appears to me to be something true (αληθ ες), refrain. (7.139.1) The interpretive gloss itself rushes on in a rhetorical fugue of contraryto-fact past conditions sketching the scenario of what would have happened if the Athenians had not opposed Xerxes on the sea. It ends by explaining the meaning of the Athenians’ initiative in terms of both mo tives (“since they chose [ε λοµενοι] that Greece remain free”) and results (“one saying that the Athenians were the saviors of Greece would not 103   miss the mark of the truth [ταληθ εος]”). In his study of the concept of aletheia (truth) in Greek literary discourse, Detienne has traced the early use of the word to connote the uncontested truth that emanates from the gods and finds expression in prophecy and poetry. In the context of the polis, aletheia enters in competition with doxa (opinion), which informs the secular and more provisional discourse of public debates among peers.104 Herodotus rarely invokes aletheia as the foundation of his logos or as a realistic goal of his inquiry.105 Though accompanied by the narrator’s more normal vocabulary of opinion and evidence, this is the Histories’ most unambiguous proclamation of “truth” in the sphere of human knowledge. It emphasizes not only the maximum certainty of the gnome in terms of evidence (␸αι νεται) but also its general validity, its nonlocal and nonrelative status. We can perhaps transfer to historical reports the connotation, 103. 7.139.1– 5. See Demand 1987 for the rhetorical aspects of this passage. 104. See Detienne 1973, 25– 47, 81– 110. 105. On the predominance of opinion over truth in Herodotus, see Darbo-Peschanski 1987, 164– 89, especially 165– 67.

174

Telling Wonders

which Nagy applies to the realm of poetic traditions, of aletheia as the feature that characterizes the canonical version and excludes all other local variants.106 The phrase “one saying that the Athenians were the   saviors of Greece would not miss the mark of the αλη␽ ες” (7.139.5) canonizes what happens to coincide, on the whole, with the Athenian version of the war.107 It validates it as the Panhellenic version, since one (τις) can refer to any member of the audience, whether Theban, Corinthian, or Argive, or any Greek, including the floating histor. Herodotus here publicly performs a gnome that alone must be and must remain accepted by all Hellenes. It is, as he says, compulsory. The true gnome is compulsory but unpleasant. It is ε πι ␸θονος, likely to make those who express it the objects of phthonos (envy) for most men, because so is Athens, the tyrant city whose self-glorifying claims it confirms.108 A contemporary argument, surely a commonplace response to those Athenian claims in the 430s, was to point out that the Athenians perhaps behaved well in the war against the Persians but now were behaving badly toward the Greeks.109 Herodotus here reverses the terms and places the Panhellenic gnome about the past in the foreground while alluding to the accusations concerning the present: if Athens is ε πι ␸θονος to the Greeks now, it has been, at the time of Xerxes’ invasion, their savior. This clear-cut judgment frees Herodotus in the subsequent narrative of Salamis and beyond to continue indicating to his audience the signs of trouble to come within the available record of past events.110 Herodotus’ sense of 106. Nagy 1990, 59– 68. 107. See Loraux 1986, 58. See the Athenian speakers in Thuc. 1.73.4– 74.4, especially the contrary-to-fact condition at 1.74.4; cf. Lysias 2.45. The Peloponnesian version would have maintained the primacy of the battle of Plataea (see Immerwahr 1966, 240 n. 8) and minimized the merit of the Athenians by emphasizing the strategical errors of the Persians (see the Corinthians in Thuc. 1.69.5).  108. See Thuc. 2.8.5, and 2.64.5 (το ε πι ␸θονον). A typical recipient of ␸θονος is the tyrant: see Hdt. 3.52.4– 5, 7.236.1. See McGlew 1993, 31– 33. For the notion of the tyrant city, see chap. 2 n. 49 and corresponding text. 109. See Sthenelaidas in Thuc. 1.86.1. Whatever one thinks about the historicity of  Thucydides’ speeches (see Hornblower 1987, 45– 72, for a discussion), τα δεοντα (1.22.1) guarantees the historicity of the political code—types of arguments, word combinations, and so on. 110. On the qualified nature of Herodotus’ praise of Athens at 7.139, see Payen 1997, 189– 93. I have examined this aspect of Herodotus’ account of the battle of Salamis in Munson 1988, and see further “Disputes, Arbitration, and the Subjectivity of Opinions” later in the present chapter and chap. 4, “Vertical Analogy.” The evidence throughout the Histories does not support Evans’ view (1979, 117) that 7.139 demonstrates Herodotus’ acceptance of the moral justification of Athenian imperialism.

Interpretation and Evaluation

175

continuity between past and present often produces a representation that is nonidealized or, as Plutarch qualifies it, “malicious,” because it interprets the past in light of the present.111 This form of revisionism is nevertheless conjoined with the determination to keep the record straight and not to revise history in a way that detracts from past achievements. The result is a discourse that tends to swing back and forth between explicit praise and more covert blame. The hypothetical history of the war minus the Athenians, which supports the interpretation at 7.139 that “the Athenians were the saviors of Greece,” involves an ambivalent portrayal of the efforts of everyone else—the strategic futility of the wall across the Isthmus, the role of the nonmedizing Greeks as mere followers, and the practical uselessness of the isolated valor of the Spartans. With the realistic hypothesis that even the Spartans, when all hope would be lost, might have come to terms with the Persians, Herodotus slightly corrects the impression just conveyed by the narrative about the intransigent courage of Sperthias and Boulis.112 This instability of evaluation is typical of Herodotus’ interpretive technique in his account of Greek city-states in the Persian Wars. Thus, with the narrative of Thermopylae, we are back to the full recognition of Spartan achievements. Precisely in the narrative of Thermoplyae, we find the Spartan counterpart of the praise of Athens just considered. It similarly highlights a moment of choice in the face of the invader and is again squeezed in between a Greek council and an oracle. The Greeks at Thermopylae learn that the Persians are surrounding them; from sacrifices, they receive omens that death is about to overtake them together with the dawn. They meet to decide what to do, and their opinions are divided. Finally, part of the army leaves the pass, scattering “each to his own city” (7.219.2, index of divisiveness). At this point, the narrator steps in to interpret the action in terms of both motive and results. He corroborates a received logos (certainly Spartan) and expands on it. His praise of one party, somewhat as in the Athenian gloss, goes hand in hand with a certain ambivalence toward the others (7.220.1– 2). 111. See Fornara 1981, 155. 112. 7.134– 36. The hypothesis is a recognition of Spartan pragmatism (see Loraux 1977, 113), but it contradicts the Spartan “image” (cf. Thuc. 4.40.1, 4.36.3). Plutarch (De Malign. Herod. 29 ⫽ Mor. 864A– B) comments that at 7.139.3, Herodotus “obviously praises the Athenians not to praise the Athenians but to speak ill of everyone else.”

176

Telling Wonders

They say that it was Leonidas himself who dismissed them, concerned that they should not die; as for himself and the Spartans who were with him, [he thought] it would not have been seemly for them to leave the post they had come to guard to begin with.  I also am very much of this opinion [γνωµην], that Leonidas, after realizing that the allies lacked eagerness and did not want to share the danger, ordered them to depart, whereas for him it was not honorable to leave. By remaining there, he in fact left behind great  glory, and the good fortune of Sparta was not obliterated [κλεος    ε ξηλει ␸ετο]. µεγα ε λει πετο κα ι η Σπα ρτης ευδαιµον ι η ουκ (7.220.1– 2) Herodotus then explains why he attributes to Leonidas the role of saving Sparta, by reporting an oracle that earlier on had predicted either the destruction of the city by the Persians or the death of one of its kings (7.220.3– 4). The interpretation then resumes in similar terms as before. And [I am of the opinion that] it was because he considered this and  wished to establish glory [κλεος] for the Spartiates alone that Leonidas dismissed the allies, rather than them leaving in disorderly fashion because their opinion differed. (7.220.4) Just as the Athenians “became the saviors of Greece,” so Leonidas “left behind great glory, and the good fortune of Sparta was not obliterated”: this result encapsulates the ultimate significance of Leonidas’  choice in the perception of the narrator. In the poetic tradition, κλεος (glory) is almost a technical term for the glory of heroes, especially in death. With a connection that is again traditional in epic, the two occurrences of kleos in this gloss frame the mention of penthos in the oracle’s  prophecy that the city will mourn (πενθησει) the death of its king descended from Heracles.113 Herodotus, in other words, directs the audience to interpret Thermopylae in terms of the epic code of heroic achievement and commemoration, thereby reinforcing elements of that code that are scattered throughout the narrative.114 Previously Herodotus had   113. 7.220.4, lines 3– 4. On κλεος compensating πενθος in several passages of the Iliad, the Odyssey, and the Theogony, see Nagy 1979, 94—117. 114. In this narrative, what starts out as a hoplitic battle becomes a heroic battle (7.223– 25); see Loraux 1977, 116; Dillery 1996, 235– 42, 245– 49. The notice of the fight over Leonidas’ body (7.225.1) evokes the Homeric fight over the body of Patroclus (Il.

Interpretation and Evaluation

177

stepped in with a true gnome (opinion/interpretation) to mark the essence of Athenian merit vis-a-vis ` the rest of Greece. Here, again on the authority of his gnome, he marks the moment at which Sparta, through Leonidas, fulfills its potential as a Greek city-state in the Panhellenic tradition of heroic valor.  The uniqueness of this gnome is confirmed by the use of κλεος and derivatives in the rest of the Histories. The stem κλε- appears in the narrator’s code of celebration within the program of the first sentence, where he  promises not to let human achievement become ακλε α (inglorious). Other  and αποδε  terms we find there (the verbs ι στορεω ι κνυµαι, the nouns εργον and θωµα)  recur repeatedly in the work, both in metanarrative and in the narrative. Words of the κλε- family, by contrast, appear only three other times besides in the Leonidas gloss, exclusively in connection with Spartans, as if the notion of heroic glory represented a standard of measure   “ingloriously,” evaluates appropriate only to them. The adverb ακλε ως, the abortive expedition of Cleomenes to Attica (5.77.1). At Plataea, the Persians complain, with some justification, that the Spartans are not living  up to their renown for valor (κατα κλεος, 9.48.3). After the battle, the Aeginetan Lampon utters what the narrator evaluates as “the most impi and ous speech,” when he praises Pausanias for having achieved κλεος encourages him to finish the work by impaling the corpse of Mardonius (9.78.1– 3). These are all tainted uses, leaving the battle of Thermopylae to shine forth not only as the superior achievement with respect to the “fairest victory” of Plataea (9.64.1) but also as the only perfect fulfillment the Histories have to offer of their author’s promise in the proem to celebrate heroic glory. Leonidas leaves behind a great kleos and causes the eudaimonie  ε ξηλει ␸ετο). (happiness/good fortune) of Sparta not to be blotted out (ουκ His achievement is obviously complementary to Herodotus’ task not to let the events of men become faded with time (ε ξι τηλα) and not to let the great and wonderful deeds of Greeks and barbarians become aklea. Just as Leonidas acquires kleos and preserves the eudaimonie of the city, so Herodotus preserves the kleos of Leonidas.115 The long gloss of identification that introduces Leonidas at the beginning of the Thermopylae narrative 17.274– 87). See How and Wells 1928, 2.230. The heroic ancestry of Leonidas is noted three times, first with full genealogy (7.204, 7.208.1, 7.220.4). See also the Homeric way of indicating time at 7.215 and 7.223.1 (cf. Il. 11.86 and Od. 12.439, on which see Lloyd 1966, 186). 115. See Nagy 1990, 221– 27.

178

Telling Wonders

already marks him as a prime target for Herodotus’ commemoration: he is a descendent of Heracles and at the same time a private citizen who has  become king (7.204– 205.1). He is ο θωµαζοµενος µα λιστα, the highest object of wonder (7.204.1), and consequently belongs to the category of θωµαστα that the narrator has singled out for attention in the first sentence. As Leonidas, therefore, realizes his full potential at Thermopylae as citizen-hero and king– not king, so Herodotus fulfills the potential of his Homeric role as celebrator of deeds through the narration of Leonidas. The praise of Athens I have considered earlier stresses the collective body of Athenian citizens, while the latter magnifies Sparta through its first citizen. The two passages are, however, parallel in other respects, not least because also in the Athenian gloss, the narrator represents himself as personally achieving something exceptional and hard to do, as we have seen: there he performs a gnome that is alethes, absolutely true, both in the epistemic and in the Panhellenic sense. Generalized Glosses of Interpretation Though important and far-ranging, the two glosses I have just examined are among those that confine themselves to discussing events within a specific historical context. In a smaller number of cases, by contrast, the interpretation makes a shift from the past tense of historical narrative to the timeless present and interprets the particular by predicating what is valid in general for all, regardless of time and space. These sentences are our most precious indicators of what Herodotus’ speech act ultimately “means.” Aristotle calls a generalization of this sort a gnome and defines it as “an assertion—not, however, about particulars, such as what kind of person Iphicrates is, but of a general sort, and not about everything (for example, not that the straight is the opposite of the crooked), but about things that involve actions and are to be chosen or avoided with regard to actions.”116 A statement “such as what kind of person Iphicrates is,” to use Aristotle’s example, represents in the terminology of this book a specific gloss of interpretation (or evaluation) like the two I have just discussed. Since all sorts of interpretive statements, specific and general, 116. Arist. Rhet. 2.21.2 (1394a), in the translation by Kennedy (1991). For an analysis of Aristotle’s treatment of the subject, see Lardinois 1995, 7– 13. Generalizations in Homer are also discussed by Richardson (1990, 144– 45).

Interpretation and Evaluation

179

are identified by Herodotus as his gnomai, or opinions, this is the way I will here use the word gnome; I will call Aristotle’s gnomai “gnomic sayings,” “gnomic glosses,” “maxims,” or simply “generalizations.” These are not necessarily equivalent to our proverbs, ready-made, anonymous, and handed down by tradition. Like “the wise words of the Western Apache,” they can be “the property of a particular speaker and created on the spot.”117 Aristotle calls them gnomai because they represent, or purport to represent, the speaker’s opinion. The generalizations that I will consider directly communicate the opinion of the histor Herodotus. Whether or not they reproduce traditional modes of thinking, they are always based on his interpretation of specific events or on a synthesis of his historical experience. Aristotle does not consider all possible generalizations as gnomic. By excluding such a statement as “the straight is the opposite of the crooked,” he attempts to draw the line between moralistic and scientific discourse. Herodotus’ generalizations are “scientific” in the sense that they are always worded as statements of fact, never as instructions on what would be best or what one should do.118 In terms of speech-act theory, they are “representatives,” that is, sentences that make the words fit the world, not the other way around.119 But if we compare, for example, the statement “Diseases among men derive from changes of seasons” (2.77.3) with the statement “Good fortune never stays in the same place” (1.5.4), we see that only the first is purely a description. The second pointedly makes the historical experience of its primary referent, the cities that have risen and fallen in the past, relevant to “us all.” Its gnomic character has to do with the implicit prescription to the listeners not to disregard the “law” or “rule” that the generalization formulates but rather to regulate their behavior according to it. Most generalizations in Herodotus are of this type, statements of fact with the indirect force of advice or warnings. As they provide a reading of the evidence, they also, as Aristotle says, convey a message “about things that involve actions and are to be chosen and avoided with regard to actions.” 117. Lardinois 1995, 5, quoting Basso 1976, 98. For the “coined” character of archaic Greek gnomai (in the Aristotelian sense), in which traditional recurrent themes are constantly reworded and reshaped, see Lardinois 1995, 22– 26. 118. The same is true for some of Aristotle’s own examples of gnomic sayings, as Lardinois observes (1995, 11). See Rhet. 2.21.2 (1394a– 1395b). 119. At least, this is their primary “illocutionary point.” I am relying on the taxonomy of Searle (1976). See, in the present book, my introduction, n. 41 and corresponding text.

180

Telling Wonders

Through these interventions, more conspicuously than through narrative alone, the scientific researcher appropriates in relation to his audience the stance of the sage of archaic tradition vis-a-vis ` the citizens of the polis. In particular, the narrator comes to resemble somewhat the wise advisers who populate his narratives. But wise or not, most speakers in the Histories generalize relentlessly, on all sorts of topics.120 Whether their gnomic sayings corroborate or contradict those of the narrator, they inevitably— as far as the general message of the work is concerned—tend to complicate the issues: alongside with events narrated, they are all parts of the factual evidence the Histories presents. At the same time, the utterances of speakers are never automatically reflections of Herodotus’ interpretation of that evidence’s meaning. The overt deliberative rhetoric in which Herodotus’ speakers mostly engage affects the form of their maxims. Occasionally characters generalize in a string of gnomic sayings, use maxims in the mode of “should,” posit general truths as the basic assumption or as supporting evidence for their argument, phrase them in the grammatical second person (7.50.1), or deliver them in hortatory form to their listener (µαθε  . . . , 1.207.2, 7.49.3). The narrator, for whom the deliberative aim is ως hidden behind the judicial and scientific/representative modes, never does any of these things. He tends to be a brief, spare, and cautious generalizer of what certain particulars reveal to be the case. The abundance of maxims in the reported speeches of the Histories has caused some confusion between what the narrator says and what his characters say. Moreover, generalizations tend to be strictly connected to their respective narrative contexts. These factors have led some critics to devalue Herodotus’ generalizations as merely the self-contradictory and strictly occasional by-product of the storyteller’s impulse to narrate.121 The impulse to narrate, however, is for Herodotus the impulse to accumulate evidence. A narrative may already function as the paradigmatic account of a particular action that implicitly conveys certain truths applicable to the recipient of the narrative.122 But if the proliferation of narratives and utterances creates a picture that is crowded, nuanced, diverse, 120. The seventy-five maxims of the list compiled by Lang (1984, 58– 66 and notes) occur mostly in speeches. See also Shapiro 2000. 121. See, e.g., Gould 1989, 81. 122. See Lardinois 1995, 110– 16, on the functional equivalence between gnomic sayings and paradigmatic narrative (ainoi) in the utterances of Homeric characters.

Interpretation and Evaluation

181

hard to know on the factual level, and unstable from the point of view of how and why things happen, the narrator’s generalizations highlight certain moments when the meaning of particulars appears especially clear to him. Herodotus does not pretend with his audience that the evidence always goes his way or only one way; this is an important aspect of his persona as inquirer and of the interrogative nature of the text. But from the multiplicity of experience, a moral can be derived, and occasionally the narrator is so bold as to express it. The moral is then in turn both a tool for capturing the diachronic and synchronic patterns of human experience and a guide to human action. Few in number and remarkably consistent, Herodotus’ maxims identify some of the major ethical concerns of the historie. Because they are fundamental reference points for understanding his message, we have already encountered some of them in preceding discussions. I will now survey them as a group, subdividing them primarily according to what they generalize about, and I will use them as a guide for examining some specific glosses of interpretation that are thematically related. We can isolate a category of generalizations of a rather philosophical and metaphysical sort, concerning three major issues: the instability of human fortune; the nature of divine action in the visible world, especially divine retribution; and the phenomenon of divine communication. Within another, political category, we first of all find a subgroup that wants to be considered from the point of view of the topic to which the statements are attached, because here the primary referent and the representative/ evaluative character of the maxims overshadow other aspects. This subgroup consists of two generalizations on government that are inspired by the historical case of Athens. A second subgroup of political generalization concerns war. A third perhaps holds the ideological key to the paucity of generalizations in Herodotus: it has to do with the subjectivity and relativity of opinions. The Instability of Happiness The first maxim in the work is a preliminary generalization that does not stem from the narrative but rather motivates the narrator’s program with regard to the contents and structure of his logos. It appears here underlined, quoted in its context. The self-referential metanarrative appears in bold face.

182

Telling Wonders

. . . after placing in relief the one I myself know to have been the first to initiate the wrongdoing against the Greeks, I will proceed with my logos, going through the cities of men, small and great alike. For those that were once great have for the most part become small, and those that were great in my time were previously small. Therefore, since I know [ε πιστα µενος] that human good fortune  ανθρωπη   never stays in the same place [τ ην ι ην . . . ευδαιµον ι ην   ω  ουδαµ α ε ν τωυτ &  µενουσαν], I shall mention them both alike. (1.5.3– 4) The first part of the program establishes that the first narrative of the logos will be about Croesus, whom the highly interpretive gloss identifies as “the first . . . to initiate the wrongdoing against the Greeks.” He is, in other words, that original aitios (guilty one) for the East-West conflict about whom the historie had started out to inquire (α ι τι ην, first sentence; α ι τ ι ους, 1.1). The second programmatic movement (“I will proceed . . .”) identifies the logos as a whole, describing it as a sort of odyssey (cf. Od. 1.3) in which the wanderer-narrator can choose where he wants to go: he will narratively visit all sorts of astea because most of them have an interesting history of becoming large or becoming small.123 So there are two competing topics in the logos. One centers around aggressions against the Greeks; the other is represented by the histories of “cities,” how they have grown and how they have declined. Implicitly, Croesus is part of both: he is the first aggressor of the Greeks, and the narrative about him as it turns out, relates his becoming great and becoming small. The gnomic saying itself is formulated by the narrator on his own authority (ε πισταµενος), as is typical of the generalizing element in the metanarrative of the Histories.124 The self-referential signs marking Herodotus’ gnomai contribute to displaying the researcher who does not take traditional wisdom for granted but derives a teaching from a careful evaluation of the evidence he has obtained. Here the maxim “Human good fortune never stays in the same place,” though already a result of Herodotus’ historie (research) is preliminary to the apodexis (presentation). Its purpose is to broaden the scope of the historical synthesis about 123. On the metaphor of the logos as a journey, see chap. 1, “Self-Referential Glosses.” 124. This is infrequent in other authors (see Lardinois 1995, 157– 61, 61– 63, 195– 225, 231– 72) and in Herodotus’ reported speeches; exceptions occur, for example, at 1.32.1 (see n. 127 in the present chapter), 3.40.2, and 7.18.2.

Interpretation and Evaluation

183

what has happened to cities, projecting it beyond the boundaries of the historical past. Large cities have become small, and small cities have become great; history has been so consistent in this respect that Herodotus can surmise that this sort of thing will continue to happen, not only to cities, but to nations and states as well as individuals.125 The transition to the present tense signals that the logos of Herodotus, although it will narrate things about the past, is relevant to listeners at the moment of the narration and that it communicates a universal historical norm about their future. Croesus will call this norm of instability the “cycle of human things” (1.207.2), but the narrator’s formulation bears no suggestion of cyclic regularity, no determination of time, and no declaration of historical necessity. Under which circumstances, then, and for which causes does eudaimonie migrate from one place to another? Divine Retribution In the course of the logos, the narrator eventually offers two generalized interpretations that answer this question (2.120.5, 4.205). Both passages connect the decline of cities and loss of eudaimonie mentioned at 1.5.3 with human injustice punished by the divine. The discontinuity between the immediate and long-range topics in the global program I have just considered are thereby to some extent filled in and resolved. Herodotus’ historie inquires about guilt (starting with Croesus) and about rise and fall (also starting with Croesus) because it inquires about the extent to which a rise in power entails guilt and a fall is caused by divine punishment.126 Before examining the crucial statements in which the narrator generalizes on this causality, I will briefly examine how he interacts with interpretations of other speakers on the role of the divine in the instability of human fortune. In the narrative about Croesus, the text engages with the evidence through a process of verification that betrays its own uncertainties. Two oracular utterances connect the demise of Croesus with the tisis (vengeance) for Gyges’ overthrow of the Heraclids and attribute to Croesus his 125. Eudaimonie is a political term (see, e.g., 7.220.2), but it also denotes personal happiness. In this case, as usual, the historical/political and the biographical/personal codes are strictly conjoined. 126. On the guiltiness and loss of eudaimonie of Croesus, see Croesus’ self-exculpation at 1.87.3, contradicted at 1.91.4. See Nagy 1990, 240– 42. On the principle of divine retribution in Herodotus see Fornara 1990 and, most recently, Harrison 2000, 102– 21.

184

Telling Wonders

own share of guilt (1.13.2, 1.91). These constitute a privileged sort of evidence. The narrator, for his part, especially seeks signs of Croesus’ personal culpability. We see it in the introduction to the narrative of Croesus’ misfortunes in the descending phase of his reign, where the verb ε ι κα ζω is a mark of the most speculative degree of opinion: “After Solon  left, great anger from the god overtook Croesus because, one imagines [ως   ε ικα σαι], he believed himself to be most blessed among men” (1.34.1). The interpretive gloss just quoted adapts the description of human experience just provided by Solon to the story of Croesus’ downfall in a moralistic way. This might seem strange. Though the Athenian sage has confirmed the narrator’s generalization on the instability of good fortune, his maxims appear to emphasize either the random aspect of the process of reversal or the amorality of divine envy, which targets not wrongdoers but especially those who have everything.127 But Solon’s words are cryptic and deliver a mixed message. In his opposition between the rich man, who is likely to experience disaster, and the fortunate man, who is not (1.32.6), the disaster is called ate, a term that traditionally denotes both the misfortune and the moral folly of one who has brought the misfortune on oneself as a result of surfeit and transgression.128 Similarly, ε πιθυµι η refers to the desire that the rich man is in a better position than others to fulfill (and from which the merely fortunate man is exempt): the word suggests an irrational impulse leading to self-detrimental action, which in the Histories is typical of monarchical rulers.129 An allusion to the personal guilt connected with the opportunities that power affords therefore intrudes surreptitiously in Solon’s interpretation of the instability of fortune. The narrator’s interpretation at 1.34.1 takes the evidence of Solon’s speech into account by attributing the cause of Croesus’ loss of eudai monie to a superior force: ε κ θεου νεµεσις [anger from god]. The word nemesis (rightful indignation) is both emotional and moralistic, thereby striking a compromise between Solon’s ethically problematic phthonos (envy) and the judicial terms timorie and tisis, which Herodotus uses  µε το θειον παν  ␸θονερον  τε  ε ον 127. See the words of Solon at 1.32.1: ε πιστα µενον  κα ι ταραχωδες [“since I know that the divine is utterly invidious and troublesome”]. Plutarch (De Malign. Herod. 15 ⫽ Mor. 857F– 858A) finds this statement insulting to the  ε στι ανθρωπος  gods. Cf. 1.32.4: παν συµ␸ορη [“man is entirely a thing of chance”] and  the reference to τυχη [chance] at 1.32.5. 128. Cf. discussion of 8.77 under “Divine Communication” later in this chapter. 129. See chap. 2, n. 232 and corresponding text.

Interpretation and Evaluation

185

elsewhere to denote retribution.130 Thus, Croesus, in Herodotus’ interpretation, is neither the embodiment of mysterious human chance (Solon’s συµ␸ορη at 1.32.4) nor the victim of a divine power that prevents men from having too many good things. In the world of nature, the divinity that prunes excessive growth is a rational principle of balance: “divine providence is, as one would expect, wise.”131 But in the sphere of human history, Herodotus looks for evidence of a divine participation that makes sense also in ethical terms. Solon and other characters who are speaking to kings or are themselves kings establish a connection between calamity and greatness rather than between calamity and wrongdoing.132 The last speaker who refers to divine envy, however, will place it in an ethical context. The Greek victory against the Persians, Themistocles says, was the work of “the gods  and heroes, who begrudged [literally “envied,” ε ␸θονησαν] that one man reign over both Asia and Europe, a man who was impious and unbearably reckless, who treated in the same way temples and private buildings, burning and overthrowing the statues of the gods; a man who even flogged the sea and lay fetters on it” (8.109.3). With respect to the historical case of Xerxes, at any rate, the divinity emerges as an ethical and rational force. Divine “envy” turns out to be the response of the divine to culpable human attempts to rival, antagonize, and replace it. Herodotus first formulates a general rule concerning divine justice in relation to the loss of human eudaimonie in his interpretation of what Greek tradition regards as the paradigm of all fallen cities. In Egypt, where issues concerning the history of humankind and the most remote past of the Greeks themselves attain special clarity, Herodotus has also learned the real meaning of the destruction of Troy. By disclosing that Helen spent the entire duration of the Trojan War in Egypt, the Egyptian priests in Memphis have suddenly provided for the fall of Troy an explanation that makes sense in both historical and ethico-theological terms (2.113– 20). The interpretive gloss by which the narrator corroborates the logos argues that if Helen had reached Troy together with Alexander, 130. See Macan 1895, 1: cxiv. Contra Chiasson 1986. On the root nemes- used to denote a reaction to a wrongdoing, see especially Il. 24.53 and the survey in Giraudeau 1984, 67– 73. !  ε στι, ε ουσα  σο␸ η.  See especially 131. 3.108.2: του θει ου η προνοι η, ωσπερ κα ι ο ι κος Immerwahr 1966, 312– 13. 132. Amasis does so at 3.40.2 (cf. 3.43.1), Artabanus at 7.46.3 and 7.10ε. So does the epinician poet in his praise of prominent individuals and tyrants. See McGlew 1993, 41; Nagy 1990, 274– 313. But Artabanus reveals his moralism at 7.18.2.

186

Telling Wonders

as Homer says she did, the Trojans would have ended up returning her to the Greeks, and Troy would not have been destroyed (2.120). Thus, if the chance arrival of Helen in Egypt was the single factor that deprived the Trojans of the means of reparation and therefore survival, it was not a chance arrival at all. The simplicity of this discovery is so compelling that it proves to the narrator that for great injustices, great are also the punishments from the gods των µεγα λαι ε ι σ ι κα ι α ι τιµωρι αι παρα  µεγα λων αδικηµα  [των  θεων].  των (2.120.5) We should note, as in the case of the maxim on the instability of human happiness, all the subsidiary issues that this generalization still leaves up in the air. If the divine always punishes human injustice, is the loss of human eudaimonie always the manifestation of divine punishment? Do cities always “become small” because of human injustice? Herodotus’ evidence in the narrative does not support the notion that divine action can be intelligible to men in every case.133 The narrator even acknowledges the often unexplained mixture of fortune and misfortune that is inherent to the human state.134 But the general rule that grave offenses will meet with great punishments represents Herodotus’ minimal interpretation of what the gods’ action at Troy intended to make manifest to men, an interpretation that Herodotus, in his turn, under(κατα␸ανες)  takes to make manifest to his audience (ε γω γνωµην απο␸α ι νοµαι). An especially important problem remains unresolved on the paradigmatic battlefield of Troy: the ethical status of the human avenger. Because the war the Greeks waged against Troy represented an instrument of divine retribution against Trojan wrongdoings, was it a just action? Menelaus’ “impious” deed in Egypt, clearly evoking the human sacrifice at Aulis in one of the Greek traditions, gives a hint that not all may be morally right with the aggressor against Troy.135 Herodotus’ second gener133. See especially the logos of Mycerinus (2.129, 133). 134. See Harrison 2000, 112– 13. Thus, though the ultimate end of Cambyses does verify the rule of divine retribution (see 3.64.3), the cause of his madness is uncertain. See 3.33 (analyzed in Munson 1991, 51– 53), where the generalization “many are the evils that are wont to happen to men” stands to the causes of misfortune, as the generalization at 5.9.3, “in the vast length of time, everything is possible,” stands to the vicissitudes of history. 135. See 2.119.2– 3 and discussion under “Revising Greek Traditions” earlier in this chapter. On Greek responsibility for the war against Troy, see also 1.4.1. Outside of Herodotus, see especially Aesch. Agam. 40– 72, 104– 39, 183– 226. See also the discussions in Nussbaum 1986 (32– 38) and Williams 1993 (132– 36).

Interpretation and Evaluation

187

alization about divine retribution precisely evaluates the justice of human revenge in the eyes of the god. This generalization occurs in the narrative about the antecedents of the Persian expedition against Cyrene during the reign of Darius. The sixth king of Cyrene, Arcesilaus, is killed in Barca. His mother, Pheretime, eager to make war against the city, obtains the help of the Persians, who use this revenge as a false pretext for the conquering of Libya (4.165– 67). Once they capture Barca, they enslave most of the citizens. Pheretime impales around the bastions of the city those most responsible for her son’s death and cuts off the breasts of their women, planting them all around the walls (4.202– 203.1). Like the excess of her revenge, the horror of Pheretime’s subsequent end gives her story the clarity of limiting cases. But Pheretime did not end her life well either. For as soon as, having obtained her revenge from the Barcaeans, she returned to Egypt from Libya, she died a bad death: still alive, she started breeding worms  out of herself, because, truly then [αρα], vengeful acts that are excessively violent are for the gods a cause of resentment/envy against men  θεων   ε πι ␸θονοι [ανθρ ωποισι α ι λι ην ι σχυρα ι τιµωρι αι προς γινονται]. Of such a kind and so great indeed was the revenge of Pheretime, the daughter of Battus, against the Barcaeans. (4.205)  The particle αρα marks the maxim as an inference based on the events narrated. As in the Trojan logos, these involve the “becoming small” of a city—in fact, its fall as a result of a siege in turn motivated by revenge. Here, however, the narrator’s interpretation focuses on the culpability of the avenger. In each case, the generalization drawn from a specific historical event involves a shift to the grammatical plural. Just as great adikiai (injustices) attract great timoriai (vengeful actions) from the gods—and these divine timoriai may be carried out by human agents—so when the human timoriai are excessive they become injustices, which will in turn deserve divine punishment because they are ε πι ␸θονοι (cause for resentment/envy) to the gods. Only in this passage does the narrator participate in the theological code of his speakers by using the phthon-stem in connection with the calamities that god sends to man. Like Themistocles in the speech we have already seen (8.109.3), he here joins the Solonian notion of divine envy of human power to that of divine anger against wrongdoing. While the timoriai of the gods are never “excessively violent” (as devastating as they may seem), the

188

Telling Wonders

excessively violent timoriai of men, here represented by Pheretime’s mutilations, constitute “monarchic” usurpations of a divine prerogative— hence invoking the phthonos of the gods.136 Punishment from the gods is a fundamental historical cause of human reversal in the Histories and is the only historical cause at any level that the narrator proclaims in general terms. This goes a long way toward explaining why Herodotus’ inquiry into human causes is largely concerned with assessing the aitiai (guilts or accusations that become causes of actions for the accuser) and who is aitios (responsible, guilty).137 From beginning to end, the logos provides evidence for the recurring role of human vengeance as the only possible motive that has a theoretical claim of legitimizing a range of aggressive actions (dispossessing, murdering, making war on others) that are otherwise adika, “unjust.”138 The narrative of the destruction of Barca, where vengeance is a pretext for the Persians (4.167.3) and a criminal impulse for Pheretime, encapsulates the criteria by which Herodotus’ logos undermines such a claim. Many timoriai of the Histories, with or without ulterior motives, are equivalent to adikiai, and as such they are vulnerable to divine punishment. From ancient history and foreign settings, Herodotus’ investigation of the workings of divine vengeance eventually reaches the context of the struggles between and within the cities of Greece. In an astounding compositional move, Herodotus gives the Spartan king Leotychides, himself a paradigm of crime and punishment, the role of asserting the ineluctability of divine retribution through his own narrative: this is the story of Glaucus, a man who presumed to obtain Delphic permission to commit perjury and whose house was subsequently obliterated.139 Much as do Herodotus’ own narratives, Leotichides’ story functions at once as an ainos and as the presentation of a piece of historical evidence. It is ad136. The monarchic connotation of ε πι ␸θονος suits the present context also because Herodotus’ history of Cyrene illustrates the way in which a royal dynasty of founders becomes transformed into a tyrannical dynasty (see especially 4.161.3– 4.164). See McGlew 1993, 172– 73. 137. The fundamental study of the various aspects of causality in Herodotus is still Immerwahr 1956. See also Pagel 1927; de Romilly 1971; Lateiner 1989, 189– 210. 138. Cf., e.g., Croesus’ aitiai against cities of Ionia (1.26.3); the role of revenge in Croesus’ decision to attack Cyrus (1.46.1, 1.71.1, 1.73, 1.75.1), in Darius’ Scythian expedition (4.1.1; cf. 4.118.3– 5, 4.119.2– 4), and in Xerxes’ Greek expedition (7.5– 9, 7.11.2– 4). See Immerwahr 1956, 253, 254– 64. 139. See 6.86α– δ. On Leotychides himself as victim of divine tisis, see 6.72.1 (where the negation reproduces the pattern of the gloss about Pheretime at 4.205).

Interpretation and Evaluation

189

dressed to the Athenians who, just as Glaucus tried to keep a deposit of money, now refuse to return certain Aeginetan hostages. Its context is the account of the hostilities between Aegina and Athens, which Herodotus pursues piecemeal, starting from the time when the Aeginetans, exalted by their prosperity and mindful of their “ancient enmity” against Athens, help the Thebans to avenge themselves against the Athenians (see 5.79.1: τει σασθαι, τι σις) by initiating a “heraldless war.” The Athenians then prepare their timorie against Aegina despite an oracle that advised them to wait for thirty years to attack; at that moment, however, the threatened Spartan invasion of Attica intervenes to suspend their plan (5.89.2– 90.1). As a result of this conflict with Athens, Aegina medizes, which induces the Spartans to seize Aeginetan hostages and confine them in Athens (6.49– 50, 6.73.1– 2). Leotychides’ parable about Glaucus has no effect: the Athenians keep the hostages, with the excuse (προ␸α σιας) that they do not “consider  to return to only one Spartan king what has been enjust” [δικαιουν] trusted to them by two (6.86). The Aeginetans feel injured by the Athenian refusal, even though, for their part, they still have not “paid the των)   penalty” [δουναι δι κας] for their own previous injustices (αδικηµα against the Athenians when they “acted with hubris” to please the Thebans (6.87: the reference is to the heraldless war). So the Aeginetans retaliate by capturing an Athenian theoric ship during a festival at Sounion (6.87); the Athenians in turn respond by conspiring with a cer tain Nicodromus, an Aeginetan who had a grudge (µεµ␸οµενος) against his fellow citizens and who now agrees to betray the city to the Athenians. The plan does not succeed, but the Athenians eventually inflict a major defeat by sea on the Aeginetans (6.88– 90, 6.92.1). In the whole intermittent narrative that centers around the hostility between Aegina and Athens (including a flashback at 5.82– 88 about the beginning of their “ancient enmity,” a unified code emphasizes a tangle of abuses and retaliations. The objective ethical status of all these timoriai and tiseis, which in turn elicit retaliation, remains remarkably ambivalent.140 Mutual conflicts pitting city against city and citizen against citizen  140. This is confirmed by the insight of Lateiner (1980) that the phrase δουναι δι κας (which appears at 6.87 in reference to the Aeginetan failure to pay the penalty to the Athenians) always reflects the claims of various parties, not the position of the narrator. For timoria words emphasizing “the key motive for heroic action” in fifth-century Athenian political discourse, see Rose 1995, 77– 79, on the Ajax and on Pericles’ statements in Thuc. 2.42.

190

Telling Wonders

leave no one free of adikemata and impiety, even though everyone acts on the basis of alleged criteria of justice, justified vengeance, procedural issues, and the like.141 In the midst of all this, the parable of Glaucus reintroduces the notion that in contrast with men, the divine operates with surefooted simplicity and unerring autonomy, cuts across legalistic pretenses, identifies each wrongful action, and correctly assesses intentions. Nothing is said about the consequences of the Athenians’ violation of a binding agreement parallel to that of Glaucus (i.e., their failure to return the hostages). But given the context, that omission does not necessarily imply that they have avoided divine punishment, nor does having the message on divine justice conveyed by an individual who will himself be a historical exemplum for it impugn its validity. In the same narrative, in fact, Herodotus identifies direct evidence for divine retribution. In Aegina, the wealthy faction executed the democrats led by Nicodromus; they even cut off the hands of one of them who was hanging on to the temple of Demeter Lawgiver. Herodotus interprets,  “From this action, even a curse [αγος] resulted for them, from which they were not able to free themselves by means of sacrifices, although they kept trying, but rather they were banished from the island before the goddess would become appeased toward them” (6.91.1). Through the connection this gloss imposes on mutually distant and causally unrelated events, Aegina is the most recent instance presented in the Histories in which a city’s loss of eudaimonie is a historical process related to guilt. Precisely the word eudaimonie is used to denote the prosperity that led the people of Aegina to initiate their heraldless war (5.81.2; the war is called an act of hubris at 6.87). The Aeginetans’ history of violations, which begins with their theft of the statues of Damias and Auxesias from Epidaurus (5.83.2), culminates in their execution of the suppliant and ends, well within the recent memory of Herodotus’ audience, when Aegina ceases to exist as a polis.142 Divine justice has been as radical in this case as for Glaucus and Troy.143 Herodotus’ interpretation of the fall of Aegina as the result of divine 141. See Immerwahr 1966, 211– 15; Lachenaud 1978, 51– 55. 142. 431 b.c. See Thuc. 2.27.1, where the statement that “the Athenians expelled from the island men, women, and children, reproaching them for being not the least responsible [α ι τι ους] toward themselves for the outbreak of the war” shows that claims of justified vengeance have not gone out of style. 143. Herodotus’ unmarked narrative, through the oracle at 5.89.2, also connects the Aeginetans’ “injustice,” represented by the heraldless war, to their loss of independence in 457 b.c. (see Thuc. 1.108.4).

Interpretation and Evaluation

191

punishment for its guilt against the gods may well stem from contemporary Athenian propaganda. Even if this is the case, those scholars who have maintained that therefore Herodotus justifies the Athenian action against the island are far off the mark.144 The ethical status of those who serve as instruments of divine tisis is for Herodotus a separate issue that he tends not to explore when he presents evidence for the tisis. The gloss at 6.91.1 does not even mention the Athenians as the agents of the destruction of Aegina. But the entire narrative of the hostility between Athens and Aegina in the Histories accumulates instances of Athenian culpability in other respects and attributes to both parties largely symmetrical guilts.145 Herodotus’ Aegina is both the double of Athens in the same narrative and the antecedent of the Athens of later times.146 The city whose cycle of rise and fall has just been completed by the time of narration does not provide a comforting analogue for one that seems bound on a similar course. With the parable of Glaucus, explicitly introduced to underline the Athenians’ bad-faith refusal to return the hostages, the issue of Athenian vulnerability in the face of the divine first comes to the surface of the text. Both Athens and Aegina survive to contribute to the subsequent resistance against Persia; by some mysterious compensation, their mutual hostility will produce the fleet that will save all the Greeks (7.144.2). But further on in the logos, Herodotus alludes to the possible consequences of Athenian injustice when he wonders what happened to Athens in retribution for throwing into the well the heralds whom Darius had sent to demand submission. What undesirable thing happened to the Athenians as a result of this action, I cannot tell, except that their land and city were 144. See How and Wells 1928, 2:100. 145. For signs of Athenian culpability in the religious sphere as well as in the secular sphere in the account of early hostilities with Aegina, see 5.85– 86, 5.87.1– 2, 5.89.3, 5.86.1, 5.87. 146. See the allusions to contemporary circumstances within the narrative of their early hostilities (5.82– 88), discussed by Figuera (1985, 66). At 5.83.1– 2, the Aeginetans are described as θαλασσακρα τορες (masters of the sea), a term that normally refers to fifthcentury Athenian sea power ([Xen. Ath. Pol. 2.2.1, 14; see Figuera 1985, 91– 92 and nn. 28– 30). The analogy between Athens and Aegina is reinforced by the way in which each of them mistakenly relies on the support of the Aeacidae, who stand for divine protection for the purposes of defense (8.83.2), not of waging or helping others to wage aggressive wars (5.79– 81; cf. 5.89).

192

Telling Wonders

devastated, but I do not think that this happened for this cause/guilt [α ι τι η]. (7.133.2) The question is particularly cogent because the narrator asks it while again in the process of verifying the rule of divine tisis. To the guilt of the Athenians corresponds an analogous guilt of the Spartans, who threw Darius’ heralds into a pit (7.131, 133.1). But while the punishment of the Athenians remains undetermined, that which befell the Spartans on account of the anger of Talthybius, the heroic patron of heralds, is a matter of record. Just as the Spartans, through Sperthias and Boulis, offered compensation to the Persian king, whose heralds were sacrilegiously killed, so the gloss just quoted tentatively mentions the Persian king’s devastation of Athens as a punishment for the Athenian crime. Herodotus immediately discards the interpretation as unsatisfactory, however, just as he shows how the artificial reparation of the Spartans succeeded only in part. The divine works at its own pace, and in both cases, it postpones tisis to a later time. When Sperthias and Boulis offered their lives to Xerxes, the king declined to break in his turn the “laws of all men” and to free the Spartans from the aitie (guilt) by making himself aitios (guilty) through vengeance (7.136.2). Divine anger, however, “flared up again much later during the war between Athenians and Peloponnesians, as the Spartans say.” The Spartans make the causal connection between the wrath of Talthybius and the later occurrence (7.137.1); but as he introduces the proleptic report, Herodotus corroborates their interpretation in his own voice. Under the circumstances especially, this event appears to me very much the work of the divine [µοι . . . θειοτατον ␸αι νεται]. The fact that the anger of Talthybius was discharged against heralds and did not abate before it found its expression—it was justice that brought  ε␸ερε]; but the fact that it should fall on this about [το δι καιον ουτω the children of these men who had gone to the king on account of [Talthybius’] anger, Nicolas the son of Sperthias and Aneristus the son of Boulis . . . well, it is evident to me that this was the work of the  µοι οτι  θειον ε γενετο divine [δηλον ων το πρηγµα]. (7.137.1– 2) This gloss, which brings about the latest datable reference in the Histories, also constitutes the most spectacular confirmation of the maxim

Interpretation and Evaluation

193

attached to the Egyptian logos about Troy that the gods punish human injustices (2.120.5). Opinion in that passage (γνωµη) is here replaced by glosses of evidence (“this event appears to me,” “it is evident to me”). Herodotus’ inquiry on the morality of the divine has found another landmark that precludes randomness (“it was justice that brought this about”). As in the case of Helen’s stay in Egypt, the lack of factual connection between human planning and the achievement of ethically intelligible results proves the effect of a higher causality that cooperates with and exploits, for autonomous ends, the political motives of men and the natural course of events. These events are finally recounted: “many years after the expedition of the king,” Peloponnesian ambassadors bound “for Asia,” including the son of Sperthias and the son of Boulis, were captured at the Hellespont and put to death in Athens (7.137.3– 138.1). According to Thucydides (2.67), this happened in the second year of the Peloponnesian War, and the ambassadors in question were on their way to seek money and military cooperation from the Persian king. The narrative of Herodotus avoids emphasizing the painful discrepancy between the mission of Sperthias and Boulis to Persia and that of their sons fifty years later. Yet the bare mention of this second set of Spartan envoys to Persia would have been enough to bring to the consciousness of the listeners the changed circumstances of their own time.147 More striking is the uncomfortable asymmetry the narrative creates between Sparta and Athens in respect to their parallel crimes. Not only does the Athenian killing of the Persian heralds remain unrequited in the presence of the Spartan evidence for the inevitability of divine retribution, but the very same action that frees the Spartans from their debt serves also to compound the guilt of Athens.148 The depopulation of Aegina in 431 b.c. and the killing of the Peloponnesian heralds in 430 owe their place in Herodotus’ history to their value as the most recent demonstrations of the rule of divine tisis.149 In both cases, the human agents of tisis are the Athenians; but in the second case, the Athenians also figure as the perpetrators of an action that is analogous to the crime for which the gods hold the Spartans accountable. By 147. Would they have known that the Athenians threw the Peloponnesian envoys into a pit [ε ς ␸α ραγγα] and that their motive was revenge (see Thuc. 2.67.4)? 148. This point is made by Georges (1994, 161– 63). 149. See also 7.233, which again suggests the idea of punishment and further guilt, this time of the Thebans. See Cobet 1971, 71.

194

Telling Wonders

430 a series of setbacks, some entirely unforeseeable, made the future of Athens more uncertain than ever before.150 Herodotus’ silence on the topics of the Peloponnesian invasions of Attica and of the outbreak of the plague is in keeping with his usual reticence. But his accumulation of historical evidence for the moral action of the gods leads him to identify Athens as an anticipated, not yet consummated paradigm at the end of a series of complete and closed cases starting with Troy. What will happen to Athens—what has perhaps already started to happen—for her past and more recent aitiai, symbolized by the old and the new heralds? This question remains implicit and suspended. At the inception of the account of Xerxes’ invasion that immediately follows, moreover, Herodotus deliberately sets it aside and counterbalances his reference to the guilt of Athens with his famous “true opinion,” a forceful assessment of Athens’ merit in the Persian War (7.139). Divine Communication The participation of the divine is constant. Only exceptionally, however, does it fail to coincide with the natural course of things, and only exceptionally can men detect it. In the affair of the heralds, the particular coincidence of events was “evidently divine” (7.137.2). In the moments before the battle of Mycale, the combination of the epiphany of a heralds’ staff on the beach and a rumor in the Greek camp about successful fighting at Plataea is for the narrator evidence of the divine nature of things in general.151 Evident from many signs is the divinity of events [δηλα δ ε πολλοι σι  πρηγµα των], if also on that occasion τεκµηρι οισ ι ε στι τα θεια των it happened that . . . (9.100.2) The miracle of Mycale stands at the intersection between divine communication and divine interference. By affecting human events, the gods, 150. See Thuc. 7.28.3 and, for the strangeness of the plague, especially Thuc. 2.50, 2.51.1. 151. See 9.101.2. Another coincidence is that the battles of Plataea and Mycale were both fought in the neighborhood of a precinct of Eleusinian Demeter (see 9.101.1). On other occasions , even one coincidence occurring by itself is noted (see 7.166, 8.15.1; cf. 6.116). See Immerwahr 1966, 254 and n. 52, 258. On divine communication, see Harrison 2000, 122– 57.

Interpretation and Evaluation

195

as in the destruction of Troy, make their criteria of regulation manifest to 2.120.5). Here they affect an event humankind in general (κατα␸ανες, through communication, because their revealing the success at Plataea increases the energy of the Greek army at Mycale (9.100.2). In most cases, however, divine communication does not entail effective action. Herodotus’ exemplary case of divine communication tout court is what happened to the Chians at the end of the Ionian revolt. In the battle of Lade, where they performed splendid deeds of valor, the Chians suffered the greatest number of losses (6.15– 16), and in the aftermath, they became the victims of a seemingly gratuitous attack by Histiaeus (6.26). Factually unrelated but symbolically analogous natural disasters preceded this series of political and military misfortunes that “brought the city to its knees.”152 To Herodotus, this is proof of a mysterious link (κως [somehow], 6.27.1; cf. 3.106.1, 3.108.2) and of a general rule of divine communication. There are wont to be signs somehow, when great evils are about to  αν δε κως προσηµαι νειν, ευτ  happen to a city or to a people [␸ιλ εει η εθνει εσεσθαι]; for before these events, η µεγα λα κακα η πολι µελλ µεγα λα] came also to the Chians. (6.27.1) great signs [σηµηια It becomes evident that the gods send great semeia (signs) in the imminence of great kaka (misfortunes), just as they punish great adikiai (injustices) by means of great timoriai (punishments). Between these two areas of divine intervention—retribution or any type of effective action, on the one hand, and prediction, on the other—it is not possible to establish a necessary or clear correlation. The text does not direct us to interpret Chian misfortunes as divine punishments. Their higher causes are perhaps imbricated with other guilts from other agents of the Ionian revolt, but they remain, and Herodotus allows them to remain, unknown.153 Here, empirical historical facts only allow him to perceive the divine as the sender of signs. To the broader sphere of prophecies, Herodotus seems intent on 152. See 6.27.2 for two separate incidents that cause the death of Chian children. The analogical link between these event signifiers and the event signifieds is that the victims here represent the future of the city, which will be more seriously compromised as a result of the war. See Immerwahr 1954, 16– 17. 153. See the oracle quoted at 6.19.2 with an unexplained reference to “evil deeds” of  ε πιµηχανε Miletus [Μι λητε, κακων εργων].

196

Telling Wonders

applying the method he attributes to the Egyptians, who write down each teras (prodigy) and what it turns out to have predicted; in this way, if a similar sign occurs, they will know what it means (2.82.2). Capable of performing a great variety of speech acts, sometimes within the same utterance, dreams and oracles are more likely than are event signifiers to indicate not only that something will happen but also, retrospectively at least, why, in terms of divine criteria of regulation, something was going to happen.154 Thus, a dream vision, not a human speaker, formulates the only clear-cut maxim concerning divine punishment in a voice other than that of the narrator. This provides some of Herodotus’ best evidence that divine verbal communication with men can be especially useful for the interpretation of history. On the eve of his murder at the hands of Harmodius and Aristogeiton, Hipparchus saw in his sleep a man of imposing and splendid appearance who “uttered the following riddling lines” [α ι ν ι σσεται τα δε τα επεα]: “Bear, o lion, unbearable things, suffering with forbearing heart. / No one among men who commits injustice will avoid vengeance  ι ς ανθρ    τι σιν ουκ  αποτ  [ουδε ωπων αδικ ων ι σει]” (5.56.1). These words exemplify both the value of verbal signs from the gods and the problems they create for the inquirer. An exhortation in the first line implies a prediction or warning, reinforced by a gnomic saying. What subsequently happens to Hipparchus renders the prediction, at least, entirely clear, ε ναργεστα την (5.55), turning the vision into yet another item of proof of the reliability of divine communication. But aspects of the dream utterance, first and foremost the specific applicability of the maxim, remain in the realm of riddle. What was the guilt that called for tisis (retribution)? Does the address “lion” point to Hipparchus’ monarchical status, to tyranny as a field of adikie (injustice), or just to the fortitude of one who must prepare for death?155 Words and facts combined do not always lend themselves to reconstructing a clear overall picture. The interjection of a great number of oracles in the logos contributes to its meaning but also turns it into a series of riddles. 154. Crahay (1956, 40) counts ninety-six oracles in Herodotus. Their function in the narrative is examined by Kirchberg (1965); see also Lachenaud 1978, 244– 305. Dreams in Herodotus are examined by Frisch (1968). See also Harrison 2000, 122– 57. On non-Greek mantic prophecy, see Klees 1965; Lachenaud 1978, 229– 44. 155. For the lion as a term of both the monarchic and the heroic codes, see chap. 4, “Wondering Why.”

Interpretation and Evaluation

197

Herodotus’ collection includes oracles impossible to decode except retrospectively (3.57.4) and “double” or ambiguous predictions that are bound to come true—in some sense—either way. There are dreams that are perhaps not divine (7.12– 18), responses obtained through corrupt means or falsified by dishonest professionals (6.66, 7.6.3– 5), oracles that have allegedly failed a test of veracity (1.46– 49), and prophecies that are part of self-serving traditions in the discourse of international disputes (7.148.3). Among the most problematic are the two oracles the Athenians receive from Delphi before the arrival of Xerxes to Greece. Both appear to discourage resistance against the Persians.156 The narrative emphasizes the stubborn determination of the Athenians and their refusal to accept the first oracle. Their ingenuity in making the second mean what they want it to mean might in other circumstances appear an exercise in false hope (7.142– 43; cf. 1.53.3). At Mycale, an omen of victory helps the Greeks defeat the enemy by offering encouragement; here, the Delphic responses are terrifying (7.139.6) and therefore potentially self-fulfilling in a negative way. If the Athenians became the “saviors of Greece,” it is because they chose to fight at Salamis despite the oracles. Against the background of this aporia-generating complex of evidence, the narrator steps out to proclaim the principle of the veracity of divine predictions. The defensive quality of the discourse has here impaired the normal form of what I have so far been able to call a “maxim.” Nevertheless, a generalization it is, firmly attached, according to the usual procedure, to the display of a particular exhibit.  1. As for oracles, I cannot contradict/object [αντιλ εγειν] that they  are not truthful [αληθ εες], not wishing to attempt to devalue them when they speak clearly [ε ναργεως], looking at facts like these: When they bridge with their ships the sacred shore of golden-sword Artemis and marine Cynosoura, having sacked lovely Athens in their mad hope [ε λπι δι η], µαινοµεν 156. See 7.140– 41. For what both of these oracles “really” mean, see Elayi 1979, 227– 30.

198

Telling Wonders

divine Dike shall quench strong Koros, son of Hubris, dreadfully furious, who thinks he rules all things.157 2. For bronze shall clash with bronze, and Ares will redden the sea with blood. Then the all-seeing son of Cronos and the lady Victory will bring over the day of liberty for Hellas. In the face of such things and against Bacis speaking so clearly [ε ναργεως], neither do I dare myself to utter contradictions  [αντιλογ ι ας] about oracles, nor will I accept them from others. (8.77.1– 2) Through the self-referential and double negative form of the statements framing the quotation of the oracle, like the reference to the “opinion that will cause resentment” at 7.139.1, the text encodes an unreceptive attitude among the audience.158 Moreover, just as at 7.139.1 the narrator represents himself as overcoming also his own resistance and compelled publicly to display what “appears [to him] to be a true thing”  so here he comes across as an objec[τ η γε µοι ␸αι νεται ειναι αληθ ες], tive investigator who must yield to the evidence of truthfulness (ε ι σ ι  αληθ εες, 1). This time, the results of historie reassign aletheia (truth) where it properly belongs—the sphere of divine utterances. Herodotus’ statement “Looking at facts like these, I cannot utter contradictions” implies a readiness to contradict, if the empirical evidence so required. Herodotus’ position with respect to divine matters is, as we have seen, in certain respects related to the agnosticism of the Sophists. What distances Herodotus from the Sophists is the recurring message that his research keeps confirming the relevancy of the divine in both history 157. I am preserving the reading of the manuscripts, though How and Wells (1928, 2:262) say it is meaningless, and though several emendations have been proposed. All of them, at any rate, refer to the relentless will to power of Koros. 158. Skepticism in the late fifth century with regard to oracles is exemplified by the attitude of Thucydides (at, e.g., 2.17.2, 2.54.2– 3, 3.96.1, 5.26.3; see Gomme 1956, 160– 61). Even in Athens, the general public would have held more traditional views (see, e.g., Thuc. 2.52– 54, 2.8.2– 3), but modern ideas, the war, the alleged Delphic oracle favorable to Sparta (see Thuc. 1.118.3; cf. 2.54.4), and the plague contributed to the erosion of religious beliefs. Jordan 1986; Forrest 1984, 7. Ideologically, Herodotus’ assertion of the veracity of oracles is the counterpart of Sophocles’ Oedipus Tyrannus, produced in a similar climate. See Dodds 1966, 47; see especially Knox 1971, 159– 84.

Interpretation and Evaluation

199

and culture.159 Here, therefore, his profession of belief takes the form of an emphatic refusal to counter with sophistic speech the speech of god. He cannot antilegein (both “utter a rebuttal” and “make antilogies,” the last a technical term of sophistic argumentation), and he will not accept antilogiai from others.160 Daring to do so would risk offending the gods (cf. 2.45.3). But Herodotus most especially wants to protect his oversecularized audience from another danger: a disregard of divine speech that is bound to entail the misunderstanding of history. Herodotus’ generalization on the veracity of oracles is in fact attached to a special item of proof, a response that both predicts intelligibly and interprets accurately. This prophecy of Bacis corresponds point by point to the historical action of the battle of Salamis, whose narrative, as if in a movement of spontaneous recognition from the part of the narrator, it suddenly interrupts. Unlike the Delphic oracles to the Athenians concerning the same event, an oracle that speaks with unmistakable and vivid clarity corroborates a view of the divine that rehabilitates all its utterances. This oracle comes from a relatively minor source rather than from the politically involved Delphic shrine, and it is found in an ancient collection rather than being solicited for the occasion. On the strength of the evidence it provides, the ambiguous and the obscure also find their place in the scheme of things. This last point leads us to the interpretive aspect of the oracle as a part of that enargeie (clarity: 8.77.1– 2) that makes Herodotus identify it as decisive proof of the aletheie (truthfulness) of the divine word. In the case of numerous prophecies recorded in the logos, the historical context does not call for a moral judgment; if it does, the god may express it unclearly or even appear to go out of his way to avoid providing moral guidance. When people inquire about what they should do, the god likes to put the ball back in their court, leaving them to decide on the basis of their judgment of circumstances, their values, and their moral sense.161 By contrast with these cases, the oracle of Bacis concerning Salamis clearly formulates divine criteria of justice and even amounts to a direction for reading history in ethical terms. The Persian invader here becomes the embodiment of koros (surfeit/insatiability/excess), the son of hubris (violence/arrogance/transgression), which δια ∆ ι κη, “divine justice” or 159. See “Equal Knowledge” earlier in this chapter. 160. On the sophistic uses of the term antilegein, see Kerferd 1981, 60– 67, 84– 107. 161. Consider, e.g., the oracles to the Cymaeans (1.158– 59); to the Thebans and the Athenians concerning Aegina (5.79.1, 5.89.2), and about the fall of Miletus (6.19.2).

200

Telling Wonders

the justice of Zeus, finally suppresses.162 In the archaic poetic tradition,   (Pind. the word κορος, strictly associated with υβρις, as here, or with ατη Ol. 1.55– 57), denotes the negative repercussion of olbos in the sense of “material prosperity.” Koros is an undesirable excess of good things (food, wealth, power), the state of being glutted with them. It causes blind action and its disastrous consequences.163 The tyrant in Otanes’  formulation of the monarchical model is υβρι κεκορηµενος [glutted with  arrogance] (3.80.4). The word υβρις is applied twice more to the typical monarch in the Constitutional Debate, and the same stem recurs to describe either the criminal acts of kings or individuals with monarchical leanings or a people’s military aggressions.164 In the expression ε λπι δι η (8.77.1, line 3), elpis denotes the unfounded expectation of a µαινοµεν king in Herodotus’ logos, (1.80.5), and its metaphorical qualifier mainomene (crazed) recalls Herodotus’ attribution of literal madness to the most extreme of his monarchical paradigms.165 The oracle of Bacis interprets the Persian defeat of Salamis in a way that agrees with the generalization of divine retribution formulated by the narrator (2.120.5) and by the dream of Hipparchus (5.56.1). The transgressor is here the monarch in a literal and metaphorical sense, and the transgression is an imperialistic attack. Within the limited geography of Salamis, the oracle represents the battle in the same terms by which Herodotus himself throughout the logos visualizes the whole idea of transgression. The phrase “When they bridge with their ships the sacred shore of golden-sword Artemis and marine Cynosoura” (8.77.1, lines 1– 2) represents an image equivalent to the bridging of the Hellespont by Xerxes and all the other violations of physical boundaries, not as real geographical subdividers, but as the symbols of ethical laws. When the Persians bridge the strait of Salamis with the encircling maneuver, this constitutes, in the words of the oracle, an act of hubris that has as its goal 162. 8.77.1, line 4. For an early testimony of retributive justice, connected with Zeus, see Il. 16.384– 92. The formulation of the concept is discussed by Lloyd-Jones (1971). In Herodotus, cf. Hermotimus’ mention of the νοµος δι καιος [law of justice] of the gods at 8.106.3. 163. The intimate association of koros and hubris is confirmed by Solon frag. 5 Diehl3, 9, and Theog. 153– 54, where the second is the offspring of the first rather than, as in Herodotus’ oracle and in Pind. Ol. 13.10, the other way around. Cf. Aesch. Agam. 766. See Nagy 1990, 131, 281– 82, 291– 92; for the Bacis oracle in Herodotus, see 327. 164. See 3.80.2, 3.81.1, 1.114.5, 3.118.1, 3.126.2, 3.127.3, 6.127.3, 8.3.2, 9.27.2,  τυραννον 9.73.2, 5.77.4, 6.87, 7.16α2. Cf. Soph. O.T. 872: υβρις ␸υτευει (see end of n. 158 in the present chapter). 165. See chap. 2, “The Monarchical Model in Sparta,” text and n. 71.

Interpretation and Evaluation

201

to enslave the Greeks. The Persian defeat by Dike on that occasion, therefore, brings “the day of freedom” as a gift to the Greeks (8.77.2, line 8). This corresponds to Herodotus’ own account of the naval battle of Salamis as the crucial event that, as he says, allowed Greece to “survive in freedom” [περιειναι ε λευθερην] (7.139.5). It also corresponds to the idea advocated in the Histories that the ultimate act of hubris is the attempt to expand one’s rule by enslaving others. The oracle of Bacis, then, not only proves the veracity of divine utterances in view of the precision with which it predicts the subsequent factual course of events— what happened, when and where, and for whom. It also agrees point by point with Herodotus’ own understanding of the events’ meaning and encapsulates in a short utterance all the most important principles of his historical logos and of the message it wants to convey. From the point of view of the apodexis (performance), this oracle serves to validate Herodotus’ interpretation, just as from the point of view of the historie (inquiry), it may well constitute its source.166 Herodotus’ understanding of history and his interpretation of divine signs affect one another in the case of one natural phenomenon, less eloquent than an oracle, but unique and specially timed: the earthquake of Delos. Thucydides must reflect a widespread contemporary belief in mainland Greece when he says that the event, unprecedented in the history of the Hellenes, occurred “shortly before” the outbreak of the Peloponnesian War and was interpreted as a semeion (sign) of that imminent trouble.167 But Herodotus, perhaps correcting this notion, cites the authority of the Delians to assign the earthquake to 490 b.c., exactly when the Persian general Datis and his force, on their way against Greece, pass the island at the middle point in the Aegean and thereby cross the ideal boundary between Asia and Europe.168 On that occasion, he specifies, “Delos shook . . . for the first time and the last down to my day” (6.98.1). Herodotus does not contradict the public perception that the prodigy predicted the Peloponnesian War, but he rather uses it to broaden the significance that it would be more natural to attribute to the earthquake if it occurred when the Delians say it did. He interprets it, in other words, 166. For the view that the story of the Persians’ encircling maneuver at Salamis derives from the oracle, see Immerwahr 1966, 278– 79. 167. Thuc. 2.8.3. For discussion of the discrepancy between Herodotus and Thucydides, see How and Wells 1928, 2:104; French 1972, 21; Stadter 1992, 788– 99. 168. 6.98.1. See Stadter 1992, 785– 95, especially 787.

202

Telling Wonders

as a sign not merely of the misfortunes of the Persian Wars but also of the subsequent turmoil from the wars of Greeks against Greeks down to the moment of narration. The earthquake of Delos indicates that the campaign of Datis and Artaphrenes in 490 b.c. was opening for Greece a new epoch of about one hundred years, scanned by three generations of Persian kings. κου] this was a prodigy [τερας] 1. And no doubt [µεν that the  that were about god manifested to men for the evils [κακων] to happen. 2. For in the time of Darius, the son of Hystaspes, Xerxes, the son of Darius, and Artaxerxes, the son of Xerxes, during these three consecutive generations, more evils happened to Greece than during the twenty other generations that came before, some [of these evils] deriving to Greece from the Persians, and some from the leaders themselves fighting over the rule/empire/hegemony  αυτ  ων  των  κορυ␸αι ων περ ι της αρχη   ς πολεµεοντων]. [απ  εν  . . . αεικ  that Delos 3. So, it was not at all out of order [ουδ ης] be shaken, having previously been unshaken. (6.98.1– 3) The gloss verifies the maxim I considered earlier that the gods send “great signs” in anticipation of “great evils” (6.27.1). Such an extraordinary phenomenon is justified, according to the narrator, by the unprecedented dose of kaka (evils) that followed, brought about first by external and then by internal war. As we have seen happen in the course of Herodotus’ verification of cases of retribution, so also here the chronological progress of the events in the logos, coupled with the special timing of the divine, causes the narrative to rejoin the time of the narration, thereby establishing both analogy and continuity between the past and present. Several fifth-century sources look at the Persian Wars and the subsequent wars among Greeks as parallel entities with respect to size, strategy, and the ethical status of the participants. According to the official Athenian position, the achievements of Athens after the Persian Wars reproduce Athens’ achievements against the barbarians.169 On the other side of 169. Pericles compared Agamemnon’s Trojan War, as the archetypal war against barbarians, and his own Samian War (Plut. Per. 28.7, quoting Ion). The paintings of the Stoa Poikile may have set battles of the Greeks against Amazons, Trojans, and Persians side by side with one between Athenians and Spartans (Paus. 1.15). See Holscher ¨ 1998, 173– 76.

Interpretation and Evaluation

203

the ideological struggle, the “tyrant city” tradition of political discourse casts Athens, the onetime liberator of the Greeks, in the role that had formerly been played by Persia.170 For Herodotus in the passage I am considering, the analogy between the two sets of wars is based on the kaka that they both have brought to Greece. Just as the Persians bring evils on Greece for the sake of increasing their arche (empire) or of turning Greece itself into their arche (satrapy), so the leading Greek states bring evils on Greece by fighting with one another for the sake of the arche.171 Athens is not here the keystone of the analogy, except insofar as it was the only city of Greece that, after the Persian Wars, had fought for and acquired what was in fact called an arche (empire). Many interGreek wars, including the Peloponnesian War, had started as fights over Athens’ arche. Herodotus’ special understanding of the divine sign establishes continuity between the two sets of kaka by bracketing off as a unit the onehundred-year period of the reigns of the three kings Darius, Xerxes, and Artaxerxes in contrast with the previous twenty generations.172 As additional proof that this time constitutes a unity and is all therefore covered by the omen, he ends by pointing out the ominous significance of the kings’ names (6.98.3). The epoch of kaka begins with the earthquake of Delos at the time of the Marathon campaign (490 b.c.) and, implicitly, with the advent to the throne of Darius, the first of the three kings (521 b.c.). In For the continuity of Athenian achievements in the tradition of the epitaphios, see Loraux 1986, 132– 71. 170. See Elpinice’s reproach to Pericles in Plut. Per. 28 and the parallel made by the Corinthians in Thuc. 1.69.5. Implicit comparison between the Persian Wars and the Peloponnesian War is pervasive in Thucydides. See especially Connor 1984, 155– 57, 175– 76, 198– 200. Herodotus’ narrative describes Persian actions against the Greeks that suggest future Athenian actions: the plan to occupy Cythera, an item of Athenian strategy from 454 (implemented in 424), is discussed by the Persians (7.235); Persian imperialistic speeches contain terms of a code that at the time of narration were applied to Athens (7.5– 9, 7.45– 53); the report of the Persians’ establishing democracies in Ionia after the Ionian revolt (6.43.3) may allude to the Athenian practice to establish democratic governments in the states of its league. See Raaflaub 1987, 227– 28. 171. See Nagy 1990, 308, on this passage. 172. Going back twenty generations before Darius brings us to the time of the Dorian settlement in the Peloponnese, which marks the beginning of the Spartan king list. For a general scheme of Herodotus’ chronology, see Lloyd 1975, 171– 94, especially 177– 82. It would be attractive to be able to interpret the twenty-generation period as starting right after the archetypal event of the fall of Troy; but since the three subsequent generations here mentioned cannot amount to more than about one hundred years, we cannot make twenty generations reach back to ca. 1330– 1250 b.c., the approximate time of the Trojan War according to Herodotus’ chronology.

204

Telling Wonders

between these two events, the narrator elsewhere identifies another beginning, the Athenian expedition in support of the Ionian revolt (499 b.c.): Herodotus calls the Athenians’ twenty ships with a pointed allusion to the  η κακων]  for both paradigm of all wars, the “beginning of evils [αρχ Greeks and barbarians.”173 The three-king “century” includes the wars against Persia, the fighting that went on during the Pentecontaetia, and the Peloponnesian War down to the time of performance. The events of the Pentecontaetia are telescoped together with those of the Peloponnesian War. Under the heading of kaka deriving from the leaders of the Greeks fighting for the arche, both are presented as a natural continuation and extension of the kaka imported by the Persians.174 To Herodotus, in other words, the wars of the past, the Persian Wars, have perpetuated themselves. And in fact, according to his narrative, the ships built for the war against Aegina, ships that “saved Greece” (7.144.2) and allowed the Athenians to be “saviors of Greece” (7.139.5), become available again very soon for use against other Greeks. Herodotus puts no interval between the battle of Salamis and Themistocles’ attack on the islanders from whom he exacts monetary contribution by force.175 The defensive war of the Greeks against the Persians is still going on, and the tables have already turned. Immediately after Salamis, the Greeks also raise the issue about their pursuing the war against Persia offensively, turning it into a war “about the king’s own country,” as the  expression goes (8.108– 9, especially 8.108.4: το ε ν␽ευτεν δ ε περ ι της   ποιεεσθαι  ωνα;  ε κει νου ηδε τον αγ cf. 8.3.2). Similarly, after the godgiven victory of Mycale, the Greeks reject the option to move the Ionians out of Asia, the Athenians choose to preserve the “Ionian arche,” a league is founded with the Athenians in charge, and the Greek fleet departs for the Hellespont.176 These clear elements of continuity with the future, 173. 5.97.3 (see also 5.28.1, 5.30.1). Cf. Il. 5.62– 63. See Cobet 1971, 63. 174. For the Pentecontaetia as an intermediate period and a fighting prelude to the Peloponnesian War, see Thuc. 1.89– 117. Herodotus mentions the Peloponnesian War as a separate entity only at 9.73.3. 175. See 8.111– 12, a passage affected by contemporary rhetoric and referring to fifthcentury oppressive Athenian diplomacy toward smaller states. The resemblance to Thucydides’ Melian Dialogue has been noted by many (see, e.g., Aly 1929, 99; Strasburger 1955, 21; Gigante 1956, 136 and n. 1). See Stadter 1992, 795– 98. The idea that, historically, Athenian policy became very aggressive soon after 478 is argued by Fornara and Samons (1991, 76– 113). Herodotus makes the change start even earlier. 176. See 9.106.2– 4. The divine signs at Mycale (which I discuss earlier in this section) prevent interpreting the Greek crossing of the Aegean beyond Delos to liberate Ionia as a violation, as Stadter does (1992, 892– 94; see rather Immerwahr 1966, 288); on the other

Interpretation and Evaluation

205

especially the smooth transition to the offensive stage of the war and to the already ambivalent leadership of Athens, rob the Histories of a clearcut triumphant closure: the narrative simply stops shortly afterward, precariously balanced at the Hellespont and full of negative signals.177 The open-endedness of Herodotus’ logos, the suggestion that the outcome of the second rebellion of Ionia is only preliminary to further fighting, and the proleptic references to the later wars of Greeks against Greeks scattered through the last part of the history of the war against Persia178 agree with the gloss interpreting the earthquake of Delos as the announcement of a continuous chain of evils, with no relief or divide between the war he is narrating and subsequent internal struggles. The “beginning” of the three-king period is staggered over a number of years and marked by several events (the advent of Darius, the Ionian revolt, the expedition of Datis with earthquake). Its completion must be similar. The narrator positions himself still within that period, though at a point when the trend of events in Artaxerxes’ reign could already be evaluated. By then, the outbreak of the Archidamian War and all the accompanying catastrophes we learn about from Thucydides must have appeared to contemporaries to signal the beginning of the end of something.179 If we accept this as the extratextual context of Herodotus’ performance, his interpretation of the earthquake of Delos has a prophetic side: it contains an implicit prediction that a change is imminent and that a new epoch is about to begin in the next generation of Persian kings. What this change will be, what the wars of the koruphaioi of the Greeks will finally produce, remains unknown. In search for guidance for reading the future, Herodotus has canvassed the past. From the mass of sometimes uncertain evidence, he is able to derive three maxims: that hand, the aggressive operations against Andros, Paros, and Carystus, on this side of Delos, are obviously represented as unjust acts. This is another demonstration that continental boundaries as such should be taken not too rigidly but merely as symbols of ethical boundaries in specific transgressive situations. The Greek actions after Mycale are, however, presented as worrisome; the very existence of Greek cities in Asia emerges as problematic. Compare 9.106.2 with 1.164– 68 and 1.170: at 1.170, if they had followed the advice of Bias the Ionians, according to a gloss of the narrator, “could have been most prosperous among the Greeks.” 177. See chap. 2, “The Monarchical Model of Athens.” 178. See 9.35.2, 9.73, 9.75, 9.105 (Athenian siege of Carystus; cf. Thuc. 1.98.3); Schmid and Stahlin ¨ 1934, 1/2.590 n. 9; Cobet 1971, 59– 82. 179. Fornara (1971b, 32– 33; 1981, 150) argues that this passage was composed after the death of Artaxerxes (424 b.c.) and even after the end of the Archidamian war. But see Cobet 1977, 5.

206

Telling Wonders

human eudaimonie is unstable, that the gods punish wrongdoings, and that divine messages are true. Of only five explicit references to what we may broadly regard as the narrative now of the logos (i.e., the time after the outbreak of the Peloponnesian War), at least three are brought in for the purpose of proving with facts the validity of these general rules.180 Athenian Antilogies The Histories are a collection of worldwide historical evidence that ultimately seeks and offers guidelines for an objective assessment of the contemporary kaka of the Greek city-states at the time of performance and their future prospects. The histor takes an outsider’s position, and the work is directed not against any particular polis but to all the Greeks. The interpretations I have examined make clear, however, that the image of Athens as an unfinished paradigm of rise and fall and the issues of Athens’ power, merit, and guilt loom large in the second part of Herodotus’ logos. The sure sign of Athens’ prominence first occurs at the point in the narrative when that city acquires freedom from tyranny. By virtue of its new internal order, Athens “becomes great” and is the only state in the Histories that prompts the narrator to generalize in his own voice on the issue of government. While the metaphysical maxims attempt to derive absolute cosmic, ethical, and divine laws from the data provided by the visible world, when the narrator observes political realities, he acknowledges them to be inherently mixed. His two maxims on democracy occur twenty chapters apart and in the same narrative framework. They are structurally parallel and antithetical in substance. 1. And it is clear [δηλοι ], not only in one single respect but in general, that equal right to address the assembly is a serious asset [η ι σηγορι η . . . ε στ ι χρη µα σπουδαιον], if also the Athenians while they were being ruled by a tyrant were no better in war than any of their neighbors, but once they were rid of the tyrants they became by far the first. This then demonstrates that when they were being held down, they fought badly on purpose as working for a master, but once they became free,  εκαστος  ος  each man was eager to achieve for his own sake [αυτ ε ωυτω  προεθυµεετο κατεργα ζεσθαι]. (5.78) 180. See 6.91.1, 7.137, 6.98.2– 3 (discussed earlier). For 7.233.1, see n. 149 in the present chapter. The last occurs at 9.73.

Interpretation and Evaluation

207

 2. It seems that it is easier to deceive many men than one [πολλους  οικε ειναι ευπετ   γαρ εστερον διαβα λλειν η ενα], if Aristagoras could not deceive Cleomenes of Lacedaemaon, who was alone, but managed to do so with thirty thousand Athenians. (5.97.2) This antilogical set is related to the contrasting arguments of the Constitutional Debate, in which three characters analyze the advantages and disadvantages of the three basic forms of government: democracy, oligarchy, and monarchy (3.80– 83). There, each speaker holds a fixed subjective position in favor of one form of government and opposes one or two of the remaining ones. In the overall picture from the point of view of the text, however, none of the three forms appears immune from dangers for the commonwealth. The derivation of the debate from a relativistic discourse of Protagorean stamp is evident.181 The most conspicuous part of the contest is played out between monarchy and democracy, which appear mutually antithetical and parallel at the same time. In a monarchical regime, one man is permanently in charge and does what he wants without being accountable (3.80.3), while democracy is characterized by accountability, the rotation of offices, and common deliberations (3.80.6).182 Both the monarch and the demos, however, are susceptible to hubris (3.80.3; cf. 3.81.1– 2) and wrongdoing (κακοτης, 3.82.4; cf. 3.82.4). As the narrator takes stock of the contradictory character of the historical evidence, the phenomenon of the Athenian democracy prompts him to transfer to himself the antilogic functions that the Constitutional Debate distributes among different characters. The duality of Athens that emerges from other late-fifth-century texts (pure and impure, innocent and guilty, liberator and enslaver) is a central feature of that city’s portrayal in the Histories.183 Because Athens is double, it can only be described, as here, antilogically, in a form of discourse that is particularly 181. Lasserre (1976, especially 81) argues that Protagoras was the author of a constitutional debate also in a Persian setting. On the partial affinity of Herodotus’ thought with that of the Sophists, see discussions in earlier sections of this chapter. 182. The democracy of the debate (called isonomie at 3.80.6 and 3.83.1, democratie at 6.43.3) is described in terms suitable to Cleisthenes’ democracy (called democratie at 6.131.1 and 5.78). See Ostwald 1969, 96– 160; Fornara and Samons 1991, 41– 56. 183. E.g., the Spartan ambassadors at 8.142.3 contrast Athenian responsibility for the enslavement of the Greeks (α ι τι ους δουλοσυνης τοισι Ελλησι) and their role as liberators   (πολλους ε λευθερωσαντες ανθρ ωπων). Herodotus offers contradictory evaluations of Pisistratus’ reign (1.59.1, 6.). See Strasburger 1955, 10– 15. For Themistocles as the embodiment of Athens, see his antilogic arguments at 8.108– 9; see also Munson 1988, 103– 4.

208

Telling Wonders

congenial to that city and its perception of things. These glosses point in fact to another Athenian trend. Unlike most of the narrator’s generalizations we have seen so far, and unlike the discourses on government in the Constitutional Debate, they employ the vocabulary of strategy and the  useful (χρη µα σπουδαι ον, ευπετ εστερον διαβα λλειν), in preference to the ethical code used to evaluate something being good or bad.184 Both generalizations assess the advantage of the opportunities for deliberation enjoyed by the Athenian demos in the assembly, both evaluate this advantage by its results in the field of war, and both contrast it with the concentration of power in the hands of one man. In the first generalization (5.78), the term isegorie (equal speech) underlines the psychological effect on the ordinary citizen of his being able to stand in the midst of his peers and participate in the debate about policies. This effect extends beyond the field of deliberation (“not in one thing only, but in a general way”) and is here measured by performance on the battlefield. In the case of the Spartans also, Herodotus’ narrative conveys the notion that freedom makes men more eager to fight than does subjection to a ruler (7.135.3, 7.101– 4). But whereas Spartan arete, according to Demaratus, partially derives from an unquestioning obedience to the city’s nomos-despotes, stronger than any human despot (7.104.4– 5), the Athenian brand of freedom replaces the human despotes not with “the law” but with the individual η ε ργαζοµενοι). citizen (ε ωυτω   . . . κατεργα ζεσθαι vs. δεσποτ Since the Athenian citizen fights in wars that he has personally deliberated on through his right to speak in the assembly (isegorie), or, in other words, since he makes the laws, he “works” in his own interest, “each one for  εκαστος  ος  himself” [αυτ ε ωυτω   ]. Individual self-interest is the ideological foundation of popular sovereignty in the democratic state.185 The positive evaluation of isegorie is based on the specific evidence of a success in battle: the Athenian defeat of an aggressive attack by Boeotians and Chalcidians (5.77; cf. 5.74.2). The narrative that illustrates the 184. The predominance of the useful over the just in political deliberations is often mentioned by Athenian speakers in Thucydides (see, e.g., Thuc. 1.75). The theme of selfinterest recurs obsessively in the evaluation of the Athenian democracy by the “Old Oligarch” ([Xen.] Ath. Pol. 1.6– 9, 2.20, and passim). In Herodotus, the Athenians make the transition from an ethical to a utilitarian discourse in their speeches at 8.143– 44 and 9.7, 11; see particularly 9.7α1– 2.  185. See Munson 1988. The term εκαστος (each) is an index here of individualism, elsewhere of the particularism of states (1.169.1, 7.219.2, 8.57.2). The distributive form distinguishes Herodotus’ description of democracy from that of the Hippocratic author of  . . . κινδυνους  περι κινδυνεουσι Airs, Waters, Places: τους ε ωυτων (16; cf. 23).

Interpretation and Evaluation

209

successful implementation of isegorie itself is rather the report of the assembly deliberations on military policy after the oracle of the wooden wall (7.142– 43): here, the self-interest of the Athenians, empowered by freedom from tyranny and right of speech, leads to the “Good” decision to resist the Persians. The system proves again to be a “serious asset” for the commonwealth as a whole.186 The antithetical counterpart of this assembly scene is another assembly scene, which leads to a “Bad” war decision and provides the textual context and the historical evidence for Herodotus’ second, and negative, generalization on the democratic state, at 5.97.2. This evaluation occurs in the narrative of Aristagoras’ journey to mainland Greece for the purpose of obtaining support for the Ionian revolt. The narrative itself in turn provides the frame for the analeptic sequence on Athens’ liberation from tyranny and first successes, concluded by the narrator’s praise of isegorie. Thus, the coming of Aristagoras looms on the horizon, both historically and narratively, during the decisive moment of Athens’ democratic beginning and consequent rise to prominence. When he arrives holding out hopes of plunder, Athens is powerful (ε δυνα στευεν µεγιστον) and ready for an imperialistic war.187 In Herodotus’ narrative of Aristagoras’ visit, however, the representation of the Athenians’ new democratic energy, restlessness, and self-interest remains subsidiary to a less grandiose portrayal. In Athens, Aristagoras repeats the tendentious ethnographic information he has given at Sparta (5.49.3– 8), with the aid of a map such as the histor Herodotus has elsewhere judged oversimplified and inadequate.188 He adds the idealistic argument that the Milesians, as colonists of the Athenians, should be able to count on their help.189 Whereas he has failed with Cleomenes at Sparta, Aristagoras persuades the Athenians by talking of the wealth of Asia and the Persian method of warfare, in which “they use neither shield nor spear  . . . χειρωθη ναι].”190 In Herodotus’ so as to be easy to defeat [ευπετ εες 186. Similarly, the report that follows, on the earlier deliberation concerning the building of a fleet, shows that the Athenians are able to understand when private and public interests coincide (see 7.144). 187. See 5.97.1. For the connotations of dunasteia, see chap. 2, n. 32 and corresponding text. 188. See 5.49.1. Cf. 4.36.2 (see chap. 2, “The Texture of the Earth”). 189. 5.97.2. This replaces the argument of Aristagoras in Sparta on the “shame and grief” that the enslavement of the Ionians brings on the Spartans as the leaders of Greece (5.49.2– 3). That he makes this point before the representation of Eastern wealth reflects his assessment of what would be persuasive to each audience.   χειρωθη ναι). 190. 5.97.1. Cf. 5.49.3 (ευπετ εως), 4 (ευπετ εες

210

Telling Wonders

explicit interpretation at 5.97.2 (passage 2 quoted earlier), the people’s decision to support the revolt in Ionia—in turn interpreted with the famous “beginning of evils” gloss (5.97.3)—is first and foremost a demonstration of the people’s collective ignorance. As one of the speakers in the Constitutional Debate observes in stronger terms (3.81.2), this is bound to put a democratic state at a disadvantage. Throughout the Histories, Herodotus thematizes together with Athenian sophie (cleverness) the paradox of Athenian euethie (simpleminded της, ness). After Aristagoras, Miltiades deceives the Athenians (απα 6.136.1), and then Themistocles does too (διεβαλλε, 8.110.1). In their earlier history, the Athenians are deceived twice into putting a tyrant in ε ξαπατηθει ς, 1.59.5– 6; µηχανωνται   place (µηχανη, . . . πρηγµα ευηθεστα τον, 1.60.3). They take a local maiden for a goddess (1.60.3), are blind to tyrannical schemes, and do not know the world. They are  persuaded that grandiose enterprises will be easy (ευπετ εως, 6.132; cf.  χειρωθη ναι 5.97.1), because they are themselves, like all multiευπετ εες  tudes, easy to deceive (ευπετ εστερον διαβα λλειν, 5.97.2). To the ambivalence of Athens in the ethical sphere corresponds a contradiction at the level of knowledge and intelligence. This factor cuts the image of Athens down to size. We are reminded of the besotted Demos in Aristophanes’ Knights or, more strikingly, of the assembly that in Thucydides deliberates on the Sicilian expedition—sovereign, vociferous, and ready to go, but not competent or truly in charge.191 At the same time, Herodotus’ general evaluations serve a broader purpose than describing their primary referents. Just as he proclaims that freedom from tyranny and the citizens’ right to make policies through speech is a great advantage for all, so he notes that vulnerability to speech makes democracies blunder. This is due less to monarchical hubris than to the simplemindedness of the ordinary Greeks who run the state. When it comes to euethie, the people of Athens in the logos has much in common with the audiences Herodotus’ logos addresses, both Athenian and not (cf. 2.45.1). Other than communicating a more abstract moral message, the histor takes it on himself to display and to cure through his own, non-Aristagorean brand of speech this shared na¨ıvet´e about the reality of foreign peoples and lands, the shape of the world, the motives of leaders, 191. See Thuc. 6.8– 26. Athenian ignorance of the realities of Sicily (see Thuc. 6.1.1) prompts an exceptionally lengthy ethnographic and geographical passage in Thucydides (6.2– 6.5.3). See also Eur. Suppl. 410– 25, with its positive counterpart in the praise of isegoria at 432– 41.

Interpretation and Evaluation

211

and the correct and falsified signs of divine support. Ignorance in these matters affects public decisions and brings about the “evil” of unnecessary wars. The Evils of War The two interpretive glosses I have just considered assess the repercussions of a political regime on war performance and war deliberation, respectively, in a way that is consistent with Herodotus’ overarching message about these topics. Among the “achievements great and wonderful” that the apodexis histories promises to save from oblivion, acts of courage and prowess are, as in Homeric poetry, among the most conspicuous objects of the narrator’s praise.192 In the experience of the narrator, those qualities that ensure a high level of performance in battle are tied to the accomplishment of important actions that affect whole communities and represent the morally neutral “great deeds” of history.193 Equally conspicuous, however, is the idea of the intrinsic moral value of arete (excellence) in war, as the performance of a public duty and as a display of personal and national merit. A people’s struggle in war on behalf of freedom is especially “worthy of description” (see, e.g., 1.177, 2.157). But armies, contingents, and individuals, be they Greek or barbarian and whatever their cause, are regularly evaluated for the extent to which they rise to the occasion in the moment of battle. If arete is an absolute moral value, however, war is just as absolutely an evil, as the glosses on the three generations of kaka and on the beginning of kaka indicate (6.98.2, 5.97.3). This is again a Homeric idea, as is the attribution of kleos to war deeds. In Herodotus, however, the evaluation of arete competes with the assessment of responsibility and dike (justice), so that since war is an evil, making war is a culpable action, not at the level of fighting, but at the level of deliberation. Thus, Croesus, the only character in the Histories to utter a gnomic saying on the evils of war, also raises the question of who is responsible for bringing it about: “It was the god of the Greeks who, having incited me to make the expedition, was responsible [αιτιος] for these things. For no one is so out of his mind as to 192. See “Specific Glosses of Interpretation” in this chapter; chap. 1, “Referential Glosses.” See also Immerwahr 1966, 262– 65, 308– 9; Sa¨ıd 1980. Dillery (1996) examines the pattern of the duel as a contest of worth. 193. See the generalization at 7.153.4. The case that contradicts it suggests divine intervention.

212

Telling Wonders

choose war over peace. For in peace, children bury their fathers, and in war, fathers bury their children” (1.87.3– 4). Of course, Croesus is wrong on two counts. Many rulers choose war over peace in the Histories, and
 ρξαντα like the rest of them, Croesus is aitios for the wars he initiates (υπα

ι κων εργων, 1.5.3; cf. the Dephic oracle at 1.91). αδ In the distinction between providing for national security and waging a war of aggression, Herodotus’ didactic message makes one of its main points by countering the contemporary tendency to view aggression as a function of defense.194 Embedded in the evaluation of the special strategy that makes the Scythians immune from external enemies is the generalized statement that to possess means of defense is essential for any nation:  το µεγιστον   “The single most important of all human things [ε ν µ εν των

ανθρωπη ι ων πρηγµατων] has been devised by the Scythian nation in the most intelligent way we know about” (4.46.2). By the same token, Herodotus’ narrative indirectly criticizes the Ionians because, for one reason or another (lack of unity, organization, discipline, or mobility), they are unable to provide for themselves and safeguard their freedom (see, e.g., 1.169.1, 6.11– 12, 1.170). The symbolic significance of the Scythians in the Histories is to a great extent connected with their exemplary ideology of war. Though warlike by custom and unconquerable, the Scythians manage to make war as little as possible. In the episode of the Scythians and Amazons, the crossing of the Tanais signifies, as we have seen, a renunciation of aggression.195 The Scythians’ imperialistic exploit, in which they “started a war” and “began the injustice,” and for which they have received divine retribution, lies far in their past.196 They are henceforth firmly autarchic, with no impulse to expand, and are even able to avoid battle if their land is attacked. If reduced to an extreme position of defense, they are ready to fight (4.127.3). Among the Greeks of the Histories, a similar attitude toward war is attributed to the Spartans, whose imperialism has also mostly spent itself at an earlier stage of their history.197 For the benefit of Xerxes, who has 194. See the Athenians in Thuc. 1.75 and the Spartans in Thuc. 1.23.6, as well as the preemptive aggressions of Croesus and Xerxes in Herodotus (1.46, 7.11.2). See Payen 1997, especially 79– 93, 247– 319. 195. See chap. 2, “The Other Is Same.” 

  196. Cf. the words of the Scythians’ neighbors: προτερον αδικ ησαντες Περσας κα ι   αρξαντες πολεµου, etc. (4.119.2– 4). On the Scythian invasion of Asia, see 1.103.3– 106.1, 4.1.1– 2. 197. See 1.66– 68, especially 68.6; Immerwahr 1966, 200– 206.

Interpretation and Evaluation

213

challenged the Spartan claim to superior arete, Demaratus differentiates between readiness and initiative in warfare: “I do not promise to be able to fight against ten men at once or against two; if it were a matter of  τε ειναι], I would not even fight against one. But if there choice [ε
κων

were necessity [αναγκα ι η] and if some great trial urged me on, I would like nothing better than to fight against one of those men who say that they are each a match for three Greeks” (7.104.3). The anankaie Demaratus is here talking about compels men to defend their freedom with spears and even with axes and drives the Spartans at Thermopylae to fight with their hands and teeth when they lose their swords (7.135.3, 7.225.3). If a war imposes anankaie on others, by contrast, though it may be undertaken for a variety of more or less subjectively valid motives, it lies outside of the realm of necessity. In objective ethical terms, it is an unjust war, and in utilitarian terms, it is unwise because, as Artabanus says to Xerxes, it exposes one to danger “without necessity” [µηδεµιη ς 198

 γκης ε ουσης].  ανα Sparta’s reluctance to make expeditions almost leads that city to underestimate the Persian danger. At least twice in the sixth century, Sparta plans or undertakes hostile invasions of Attica but does not carry them out in the end.199 Herodotus’ portrayal would have reminded his audience of the striking change in policy that determined the outbreak of the Peloponnesian War. Previously, says Thucydides, the Spartans would avoid getting  ζωνται, Thuc.  µ η αναγκα

into wars unless compelled by necessity (ην 1.118.2). But in 431, everyone was eager to fight. With a much lower

 γκη, Thucydides attributes threshold than Herodotus for invoking ανα compulsion to both sides.200 For Herodotus’ Athenians, perceiving the difference between defense and aggression has been difficult since the beginning of democracy. Their newly acquired excellence is described in terms that suggest an agonistic effort vis-a-vis ` their fellow Greeks.201 “Extinguishing the hubris” of 198. 7.10δ1. Artabanus reiterates the thought in positive terms at 7.18.3. On Herodotus’ notion of a necessary war as a defensive war, see Munson forthcoming. 199. These Spartan missions include the abortive invasion of Cleomenes, both “unjust” (5.75.1) and “inglorious” (5.77.1); and the projected reestablishment of Hippias (see 5.91– 93, especially 5.93.1, with allusions to later events). The list of 5.76 recalls three other interventions, one mythical and two beneficial. For Spartan reluctance to go out against the Persians, see 1.82– 83, 152; 5.50; 6.106; 9.6– 8. 200. Thuc. 2.8.1; cf. 1.80.1 with 1.23.6, and see Ostwald 1988, especially 1– 5. 201. Herodotus says at 5.78, “they became by far the first.” Cf. Miltiades’ words at 6.109.3. See Stadter 1992, 801.

214

Telling Wonders

Boeotians and Chalcidians, who had attacked them first, entails “crossing over” [δια βαντες] to Euboea and establishing a cleruchy in Chalcis, an expansionistic move.202 The subsequent expedition of the Athenians in support of their Ionian kin, who in turn were fighting for their own freedom (5.97), blurs the distinction between defense and offense; according to the previously examined gloss on the three generations of kaka (6.98), it also leads from war against Persia to war among the Greeks. Herodotus’ generalized statement that war is an evil occurs precisely in a gloss that refers proleptically to the transition from the defensive to the offensive stage of the Persian Wars, after 479. The passage also represents the Athenians as causing a concomitant transition from war against a foreign enemy (polemos) to struggle within Greece (stasis). A. For there had been a discussion at the beginning, even before sending the embassy to Sicily concerning the alliance, that the Athenians should be in charge of the naval force. But when the allies objected, the Athenians yielded [εικον], because they were extremely concerned that Greece should survive and knew that  if they quarreled about the command [ε ι στασιασουσι περ ι τ ην η
γεµονι ης], Greece would perish,  and right they were [ο ρθα νοευντες]: A1. for internal struggle is a greater evil than a war fought in agree ment by as much as war is a greater evil than peace [στα σις γαρ  ε στι οσω    ! κα κιον " ! εµ␸υλος πολεµου ο
µο␸ρονεοντος τοσουτω   πολεµος ε ι ρηνης]. Being aware precisely of this fact, they did not make opposition, but yielded [εικον], B. so long as they very much needed the others, as they [subse
διεδε " 

ων,  ως  quently] showed [µεχρι οσου κα ρτα ε δεοντο αυτ ξαν]: for when, having repelled the Persian, they began to fight  for his territory, putting forward as a pretext [προ␸ασιν] the

ι λhubris of Pausanias, they snatched away the leadership [απε  η
γεµονι ην] from the Lacedaemonians. οντο τ ην But this happened later. (8.3.1– 2)  202. 5.77.1– 4. On the verb διαβαι νω, see chap. 2, n. 257. The term κληρουχους may be an anachronism (see French 1972, 18), but it evokes contemporary Athenian imperial policy.

Interpretation and Evaluation

215

Herodotus’ discourse here appropriates and modifies the contemporary anti-Athenian topos, such as we find in the speech of the Thucydidean Sthenelaidas, comparing the “good Athenians before” the Greek victory against the Persians and the “bad Athenians after.”203 Section A complements the praise at 7.139.5, where the Athenians “decided that Greece should survive and be free,” with the added element of Athenian compliance for the purpose of avoiding stasis (εικον, repeated in section A1). The transition between “good” and “bad” Athenians begins in section B,  " 

ων  [so long as they very much where µεχρι οσου κα ρτα ε δεοντο αυτ needed them] isolates within the altruistic motive just attributed to the Athenians (the safety of Greece) the motive of self-interest (the safety of
δι εδεξαν,  Greece and therefore their own). ως [as they showed], is a gloss of evidence marking the retrospective interpretation that the Athenians’ Panhellenism at the time was strategic and that their behavior after the Persian danger had passed proves their loss of concern in avoiding stasis. This behavior is connoted as a high-handed usurpation of the command

ι λοντο]; cf. 1.14.1), achieved by exploiting a situa(“snatched away” [απε tion that provided the Athenians with a legitimate official claim  (προ␸ασιν). But the passage is less clear-cut than this exegesis makes it sound. First, “the hubris of Pausanias,” though mainly focalized by the Athenians through their insincere claim, is presented as an uncontested fact. It brings the ethical code to bear to the detriment of the Spartans. Second, the fuzziness of the text is evident in the controversies of modern critics, who (not unlike Herodotus’ listeners, one presumes) have not all been equally inclined to read anti-Athenian sentiments into this passage. Thus, Pohlenz  " and Immerwahr translate µεχρι οσου at the beginning of section B not with “for so long as” but with “until”; they argue that the subject of  ε δεοντο and all the other verbs that follow must be “the allies” (implied

ων,  the object of from “the allies” in section A); and they regard αυτ

ε δεοντο,  as referring, not to the allies in 480, but to the Athenians. In this rendering, it is the allies (i.e., the members of the alliance later on, even though evidently not the same states as the allies in section A) who at one point “very much needed” the Athenians because the behavior of Pausanias had caused a crisis in the leadership. The allies therefore publicly 203. Thuc. 1.86.1. Because of its comparative element, this topos is an important subtext in Herodotus. See, in the present chapter, n. 109 and corresponding text, as well as discussion under “Disputes, Arbitration, and the Subjectivity of Opinions.”

216

Telling Wonders

 cited the hubris of Pausanias (προ␸ασιν, then, would merely point to the public nature of the protest); they took the leadership away from the Spartans and gave it to the Athenians instead.204 This is not the most natural interpretation of the passage.205 One would be almost certain that a contemporary audience would have understood it in the other way, if it were not for the words Thucydides uses when he reports how the Athenians justified the origin of their imperial rule on the basis of this change of leadership in the war against Persia. In a passage cited by Pohlenz in support of his interpretation of Herodotus’ gloss, the Athenians in Thucydides say that they obtained their empire (arche) not by

ων  force but because “the allies came to us and asked us themselves [αυτ 206   δεηθεντων] to become the leaders [η
γεµονας]. In light of a familiar contemporary pro-Athenian argument that the Ionians had asked/needed/  wanted (verb δεοµαι) the Athenians to replace the Spartans in the leader  ship, Herodotus’ κα ρτα ε δεοντο, with implied subject and object, does not express all that clearly who needed whom, and the obscurity tempers the directness of the anti-Athenian tone of the passage.207 The ambiguity is very much in keeping with the discourse of Herodotus, who, when it comes to certain issues and cities, tends to praise openly (as he here praises the Athenians for yielding the command of the fleet in the war of resistance) and to blame allusively. This gloss, at any rate, stresses not so much the issue of subsequent Athenian wrongdoing as the generalization on war and stasis, the value of which the Athenians might have continued to recognize but evidently did not. The narrator proffers it in his own voice to establish a bridge between the past of the narrative and the present of the narration, between past war and present stasis. The generalized statement is not here based on specific evidence presented in the Histories, as in other cases; rather, it derives from the current ex204. See Pohlenz 1937, 170– 72; Immerwahr 1966, 220– 21 and n. 87. 205. See especially Strasburger 1955, 20 and n. 4. 206. Thuc. 1.75.1. Here the story of the hubris of Pausanias is a subtext; cf. 1.77.6 in the same speech. Badian (1993, 130) calls this “the foundation myth of the Athenian empire.” Fornara and Samons (1991, 84) follow the regular translation of Hdt. 8.3.3 and remark on its points of contact with Thucydides’ account of the foundation of the Delian League (1.96.1).  207. Immerwahr (1966, 220– 21 and n. 87) maintains that καρτα ε δεοντο in Herodotus means “[the allies] needed very much” and not “[the allies] asked eagerly,” but I do not think that the shift in meaning between Herodotus and Thucydides is crucial here. The point is that, according to Immerwahr’s interpretation, Hdt. 8.3.2, like Thuc. 1.75.1, suggests that the idea of Athenian leadership came from the allies, not the Athenians, and that the Athenians did the allies a favor at the time.

Interpretation and Evaluation

217

tratextual experience of narrator and audience. In “if they quarreled about the leadership, Greece would perish” in section A, the verb στασια σουσι refers to the quarrel that would cause the allies to withdraw from the resistance, as they had threatened to do (8.2.2); the η
γεµονι η (leadership) that causes the quarrel is the military command in the war against Persia. The ruin of Greece will come at the hands of the Persians if the Greeks do not keep a united front. In the generalization, however, στασις (quarrel) heightens its impact because the comparison places it  within the semantic field of πολεµος (war): stasis emphulos is worse than polemos by as much as polemos is worse than peace. The ruin of Greece that derives from stasis is therefore brought about not by the Persians on account of Greek divisiveness but by the Greeks themselves fighting with one another and fighting not simply with words and withdrawals. The term η
γεµονι η similarly assumes a larger meaning: the divisiveness over a military command stands for an internal war for arche or supreme power. The transition from the defensive war of the Greeks “of one mind” against Persia in the past to the war of the Greeks against each other in the present passes through the offensive stage of the war against Persia, when the Greeks [turned the conflict into one about [the king’s] terri αγ 

ωνα   tory” [περ ι της ε κει νου ηδη τον ε ποιευντο] (8.3.2, in B). This transition is characterized by an escalating degree of kakon (evil) as well as by a decreasing degree of necessity in the Herodotean (or “Spartan”) sense of the term. As at 6.98.2, the kaka deriving from the Persians lead to the kaka caused by the Greek leading states fighting for primacy. Disputes, Arbitration, and the Subjectivity of Opinions The gloss on the evil of war and internal struggle shows that the notion of stasis encompasses verbal and diplomatic infighting as well as armed internecine conflict. Herodotus’ text rarely mentions the contemporary ideological or violent struggles of the Greeks at the time of performance, but it either symbolically represents or literally echoes them by reporting, at different narrative levels, a number of verbal controversies. A final group of gnomic statements in the Histories requires that we examine the way in which the narrator positions himself in the midst of different claims and how he is able to apply his experience as histor/researcher to a role of histor/arbitrator and composer of quarrels.208 208. Herodotus’ stance as judicial arbitrator has been examined especially by Nagy (1990, 250– 73, 314– 22).

218

Telling Wonders

Instances of stasis in the Histories fall into two main categories, narrative and metanarrative. The first include, besides war among or within Greek city-states (e.g., that between Athens and Aegina), several episodes of Greek divisiveness and disputation. These primarily occurred during the Persian Wars and concerned four related topics: Medism, strategy, behavior in battle, and command or other privileges.209 The various areas of friction overlap. Moreover, since in some cases, nonbelligerent stasis in the past either results in military conflict or raises the specter of future armed conflicts, it serves as the overarching symbol of stasis in the belligerent sense at the time of narration.210 A second category of stasis is represented by the conflicting traditions at the time of narration concerning actions of the time of the narrated. These metanarrative quarrels connect past and present in a literal way, because they constitute the verbal backdrop of active contemporary struggles. As invisible narrator of the past, Herodotus often paints an uncomfortable picture of the highhandedness, self-interest, and agonism of the Greeks and even implicitly dispenses blame. But when he appears in the text as researcher of current traditions or as judge of past actions, he records the quarrels, presides over them, absolves the actors involved, and dispenses praise. The passage I last considered establishes the continuity between diplomatic stasis of the Greek states at the time of the wars against Persia and belligerent stasis in the subsequent period; it also makes an implicit reference back to the episode of the Greek attempt to enlist Gelon of Syracuse in the cause of Greek freedom (see 8.3.1, section A: “even before sending the embassy to Sicily”). Among the narratives of quarrels, this one contains the most striking signal that in the symbolic code of the logos, the divisiveness of the Greeks in the world of the narrated stands for their subsequent state of war against one another. When the Spartan and Athenian envoys both refuse to give Gelon a share of the command of the Greek forces, he declines to participate in the resistance altogether. Just as Spartans and Athenians have argued their respective rights to leadership going back to the heroic age on the basis of ancient texts,211 209. On the theme of divisiveness among the Greeks, see Immerwahr 1966, 225– 35. 210. When the Athenians prospect an alliance with Xerxes, they allude to attacks on the other Greeks (see 9.11.2). The pact of the confederate Greeks included the imposition on the Medizing states of a tithe for Delphic Apollo to be collected after the war (see 7.132.2). Collections of war indemnities tend to lead to military aggression (see, e.g., 8.111– 12). 211. See 7.159 (reproducing Il. 7.125; see How and Wells 1928, 2:197), 7.161.3 (cf. Il. 2.552).

Interpretation and Evaluation

219

so Gelon replies with an unself-conscious quotation to a (for him) future text. Apparently the Greeks have only people who want to be 

commanders/rulers (αρχοντας) and no one who will be ruled (αρξ οµενους), he says; very well, then, the envoys should go back and report to Greece that “the spring has been taken out of the year” (7.162). The narrator interprets “the spring” as referring to Gelon’s army, which the Greeks have forfeited by their inflexibility, but his gloss merely draws attention to the oddity of the metaphor in this context (7.162.2). More appropriately, Pericles used the phrase, according to Aristotle, to eulogize the young men of Athens fallen in war and snatched from the city like the spring from the year.212 We do not know what was the occasion of this funeral oration—the Samian War of 440 has been suggested as a possibility. By appropriating the metaphor, at any rate, Herodotus guides the listener to establish vertical analogy with an extratextual context, where Gelon, who is here at the center of the dispute and plays his thematic role within the monarchical model, is no longer relevant. The point of the narrative, at the level of meaning I am talking about, is that the mythical and rhetorical themes of hegemonic propaganda, here mobilized by the various contendants to establish their primacy over the others, are an ominous symptom of discord; that they will recur in the political discourse, fueling later inter-Greek conflicts; and that these conflicts, in a much more serious sense than political speeches can express, will “take the spring out of the year” for Greece.213 Herodotus’ unmarked allusions to the public discourse of his time are likely to be more extensive than we are able to realize, given the incompleteness of our evidence. Thucydides’ narrative and speeches represent, nevertheless, a valuable source for capturing the cultural codes shared by the narrator of Herodotus’ history and his audience. Thus, for example, in one of the metanarrative disputes, the way in which the Athenians tell the story of their expulsion of the Pelasgians from Attica seems to reproduce their self-justifying argument that their empire deserves praise for not being as oppressive as it could be.214 In another narrative, a “great 212. Arist. Rhet. 3.10.7 (1411a) and 1.7.34 (1365a). How and Wells (1928, 2:198), following Wesserling, consider the gloss an interpolation on the grounds of its inappropriateness to the Herodotean context. 213. See Fornara 1971a, 84. 214. See 6.137.4; cf. the Athenians in Thuc. 1.76.3– 4, applicable to specific cases of the type of Thuc. 2.70.3 or 3.50.

220

Telling Wonders

wrangle of words”215 breaks out at Plataea concerning whose contingent has the right to occupy the left wing of the army, Athenians or Tegeans. The two contendants enumerate “deeds old and new” (9.26.1; cf. 9.27.1) that make them worthy of being deemed the best. The Athenian speech mentions past services to Greece as a whole and actions taken in defense of particular groups of Greeks against the alleged hubris of other Greeks. The list conforms to the manner of fifth-century hegemonic rhetoric.216 Also, the rhetorical topoi of this speech reproduce current conventions. The Athenians begin (I am paraphrasing), “We are gathered here to fight, not to make speeches, but since the Tegeans have challenged our primacy, we must answer” (9.27.1). Similarly, on the eve of the outbreak of the Peloponnesian War, the Athenian ambassadors at Sparta say, according to Thucydides: “We are not here to argue with your allies but on other business. However, since they have raised an outcry, we wish to show that we have a right to our possessions” (Thuc. 1.73.1). The dismissal of remote deeds as secondary to recent merits in the Persian Wars is also typical (cf. the Athenians at 9.27.5 and at Thuc. 1.73.2). In the speech of Herodotus’ Athenians, moreover, the transition between the mythical and the historical exploits ironically suggests the possibility of an ethical change: “But it does no good to recall these things [i.e., the defeat of the Amazons and so on]: for men who were excellent [χρηστοι ] then could  be no good [␸λαυροτεροι] now, and men who were no good then could be better now” (9.27.4). These words parallel those of the Thucydidean Sthenelaidas that, as I have already conjectured, seem to be based on a current anti-Athenian argument: “Yet, if they were good/courageous

[αγαθο ι ] then against the Medes and are bad [κακοι ] toward us now, they deserve double punishment because from good they have become bad” (Thuc. 1.86.1). By combining different themes of later political rhetoric in his representation of hegemonic quarrels among Greeks during the battle of Plataea, Herodotus’ logos makes reference both to the continuity and to the reversal of circumstances in the historical transition from the polemos of the narrative to the contemporary stasis. In other cases, the evocation of later events emerges from the way in which the narrative describes the actions of the specific protagonists. On  is also applied to the verbal quarrels of the Greeks at

215. 9.26.1. The word ωθισµ ος 8.78, but it denotes the melee of combatants in a physical battle at 7.225.1 and 9.62. See Immerwahr 1966, 274. 216. See 9.27.2– 5; Loraux 1986, 67– 75. One of the deeds is the Athenians’ defeat of the Amazons, on which see chap. 2, “The Other Is Same.”

Interpretation and Evaluation

221

the eve of Xerxes’ invasion, the Corcyreans contrive to remain neutral even though, as they themselves claim, they possess “a not inconsiderable number of ships, in fact the greatest number after the Athenians” (7.168.3). The emphasis on Corcyra’s selfish neutrality and its sea power here recalls how, according to Thucydides, Corcyra became a cause of discord in Greece later on: when the city saw fit to abandon its traditional policy of isolation, the size of its fleet gave it the bargaining power for obtaining an Athenian alliance.217 The point of Herodotus’ narrative, in other words, is that Corcyra’s fleet, which recently did so much to exacerbate the internal conflicts in Greece, could have helped the Greeks at the time of their defensive war against Persia but deliberately failed to do so, just as the city that had been fence-sitting in the hour of need for Greece later came out of its famous neutrality only to contribute to the outbreak of the Peloponnesian War. In the stasis narratives I have examined, the invisible narrator links instances of past controversies among the Greeks with the extratextual developments close to the time of narration.218 In a “metanarrative stasis,” by contrast, the dispute that historically and symbolically corresponds to other manifestations of enmity does not or does not only belong to the world of the narrated but has already spilled over into the world of the performance. These cases are brought to the surface of the text as the histor/researcher visibly attempts to find out and report what happened: “From this point on [in the narrative], I cannot exactly describe [present tense] which of the Ionians were [past tense] cowardly or brave men in this naval battle; for they accuse one another [present tense:

ηλους   καταιτιωνται]”  αλλ γαρ (6.14.1). The narrator’s lack of knowledge with respect to distant events reveals the interconnectedness between the characters of the logos and its sources, both apparently the  subject of καταιτιωνται, “they accuse.” The disunity of the Ionians at Lade continues in the mutual accusations of their descendants. The display of historie (inquiry) throughout Herodotus’ text communicates the important teaching that a variety of subjective factors motivates 217. For the Corcyrean affair, see Thuc. 1.24– 55, especially 1.32.4 (policy of isolation); 1.33.2, 1.36.3, 1.44.2 (Corcyrean fleet). 218. See also 8.56– 64, 75 (Themistocles vs. the Greeks: see Munson 1988, 101); 8.11– 12 (see n. 175 in the present chapter); 8.141– 9.11 (Athenians and Spartans: see Fornara 1971a, 84– 86; Raaflaub 1987, 240). Also, 6.108 should be read against the background provided by Thucydides’ Plataea narrative (Thuc. 2.71– 78, 3.20– 24, and especially 3.52– 68).

222

Telling Wonders

people’s logoi. Characteristic of the last tract of history treated by Herodotus, however, is the fact that the controversies are internal to Greece, and the problem of uncertain or contradictory reports is connected with the contemporary ideological struggle that accompanies the military struggle of the Greek city-states. Herodotus’ role as investigator and recorder of the past coincides more than ever, therefore, with a role of political arbitrator and judge. His choice between competing versions is almost bound to be equivalent to a verdict on what are the aitiai (causes/accusations/guilts) and who is aitios (responsible) in a judicial and moral sense. The defendants in these cases are not paradigmatic figures of the past, like Croesus, but Herodotus’ fellow Greeks at the time of performance.219 This perhaps explains why a negative verdict so rarely occurs. When Herodotus displays to his audience which logoi are available concerning events of contemporary political significance, he does one of two things: either he reports discrepant versions with glosses of sources and lets them stand or fall on their own merit, or he rejects or avoids corroborating those logoi that amount to indictments.220 Acquit, praise, or suspend judgment, but never convict: we may call his explicit and official judicial practice. The categorical conviction of Ephialtes at Thermopylae is the exception that proves the “no explicit conviction” rule and helps to illuminate its meaning. The judicial character of the indictment illustrates the distinction between utterances that “make the words match the world” by representing what is the case and those that “make the world match the words”: “It was Ephialtes who led the Persians around the mountain by  the path, and this man I write as the guilty one [τουτον αιτιον γρα␸ω]” (7.214.3). Herodotus’ interpretation of culpability does not represent facts as much as it establishes the record, whose concreteness and permanence he conveys by the performative verb γρα␸ω.221 As Herodotus’ own report of a variant version reveals (7.214.1), and as we can infer from the historical situation at Thermopylae, the identification of Ephialtes as the 219. Nagy (1990, 250– 73, 314– 21) examines ancient evidence concerning interpolis arbitrations and shows the juridical vocabulary and correspondence with Herodotus’ vocabulary of historie. 220. See, e.g., 5.85– 86 (logoi of Athenians and Aeginetans about their early hostilities); 6.137 (logoi of Hecataeus and Athenians about expulsion of the Pelasgians from Attica); 6.121– 31, especially 6.121.1 and 6.123.1 (accusation against the Alcmaeonids rejected: see chap. 4, “Wonder and Disbelief,” in the present book). See also 5.44– 45, especially 5.45.2. On 1.70.2– 3, see Lateiner 1989, 80; Cooper 1975.  221. See Searle 1982, 3– 4. Cf. the judicial meaning of γρα␸ η.

Interpretation and Evaluation

223

traitor would have been far from definitive.222 But the narrator singles out in the logos an ordinary individual whom an impartial judicial body representing different cities of Greece had already officially convicted (7.213.2); he had subsequently fled under the weight of the aitie and died. His death, though unrelated on the human level, had suitably struck him after his crime, with the sort of timeliness Herodotus elsewhere regards as divine.223 By excluding all other possibilities, Herodotus contains the damage, resolves the controversy, and closes the case. His indictment of Ephialtes stems from the same stance as a reconciler that he outwardly assumes when his historie calls on him to arbitrate quarrels among the Greek city-states. Yet the text of the Histories encodes two different and competing messages: it implicitly exhorts all Greeks to put aside their mutual aitiai (accusations) at the same time as it forces them to face up to the various aitiai (guilts) attributed to them. Herodotus’ task of reconciling quarrels goes hand in hand with his role as their inexorable reporter in a discourse that entails pitting the Greeks against each other. The Salamis narrative includes a remarkable Athenian rumor according to which the Corinthians fled at the beginning of the battle and were only persuaded to return by a divine apparition after having remained absent for the entire time of the action (8.94). The narrative is in indirect speech with intrusive oblique infinitive224 and is followed by the tradition on the Corinthian performance from two other sources. The Corinthians themselves, first of all, do not agree with the Athenians and believe they have been “among the first” in the naval battle; the rest of Greece, moreover, bears witness  to the Corinthians’ claim (µαρτυρεει, 8.94.4). The interrogation of witnesses, in other words, discredits the Athenian story and acquits the Corinthians of cowardice. But what is the function of this type of inclusion in the Histories? We have seen contexts when Herodotus reports logoi he vigorously refutes to expose the ignorance or bias of his sources.225 Here, however, in light of Adeimantus’ bitter opposition against fighting at Salamis (8.59– 61), and in the absence of more detailed evidence of Corinthian prowess in 222. See How and Wells 1928, 2:215; Burn [1962] 1984, 413. 223. See “Divine Retribution” earlier in this chapter. 224. This is a syntactical device by which the narrator distances himself from the received logos. See Cooper 1974. 225. See, e.g., Herodotus’ refutations of Greek traditions concerning the Egyptians (discussed under “Strategies of Evaluation” earlier in this chapter.

224

Telling Wonders

Herodotus’ narrative of the battle, the Athenian story does damage the memory of their performance.226 At the same time, the story implicitly convicts the Athenians of slander and alludes to their current enmity with Corinth. The entire Salamis narrative bristles with episodes of disunity that establish a continuum between past and present discord.227 Though no one is explicitly accused or blamed in the voice of the narrator, the logos implicitly underlines the shortcomings of all the Greeks. With the story about the alleged Corinthian cowardice in mind, we should revisit the episode of Corcyrean Medism.228 The Corcyreans promise help to the Greeks with a “fair-faced” speech, but they have “other plans” (7.168.2). Though they send sixty ships, they hold them off the coasts of the Peloponnese, watching to see how the war will turn out. Visa-vis ` the Greeks, they plan to claim that the Etesian winds prevented their fleet from rounding off Malea (7.168.4). This mind-reading narrative is enriched by a speech introduced by a purpose clause and is never in fact uttered, but reported in direct discourse (an unparalleled device), with which the Corcyreans intended to court Xerxes’ favor if he had been victorious (7.168.3). What makes the Corcyrean episode especially instructive is the deliberate way in which it excludes the form of discourse that puts historie on display. The sequence on the Corinthians at Salamis explicitly mentions two versions and three sources: a protagonist source, a minority source (the Athenians), and “the Greeks.” Similarly, the Corcyrean narrative embeds three sources and two different versions of what happened: (1) the Corcyreans say that they tried to reach the rest of the Greeks but were frustrated by the Etesian winds (called an “excuse” at 7.168.4); (2) the Greeks accepted the Corcyrean version of events (the Corcyreans “misled the Greeks”; see 7.168.4); (3) a nameless source (or Herodotus himself)  still considers the Corcyreans aitioi (cf. α ι τιωµενων, 7.168.4) and tells a different story (corresponding to the direction of the whole at 7.168.1– 4). Because in this form of discourse the unequivocal condemnation of the Corcyreans would have depended on the histor’s endorsement of 226. Plutarch (De Malign. Herod. 39 ⫽ Mor. 870B– 871B) quotes several inscriptions that celebrate Corinthian valor at Salamis and says that Herodotus’ practice of telling damaging stories only to discredit them is in fact a device for slander. See, e.g., 3.56; cf. De Malign. Herod. 27 ⫽ Mor. 863A. 227. See 8.59– 62 (quarrel over strategy); 8.84.1– 2 (metanarrative controversy on who started the battle); 8.85 (Ionians vs. mainland Greeks); 8.92 (Themistocles vs. Polycritus). 228. See 7.168, discussed earlier in this section.

Interpretation and Evaluation

225

version 3, he withdraws from this capacity and is replaced by a semi– invisible narrator who reports motives and intentions as facts. Throughout the last four books of the Histories, the fluctuation of the discourse between a minimally narrated narrative and the form of historie/dispute allows the text to show Greek failings while preserving the visible narrator as benevolent arbitrator, refuter of vicious gossip, praiser of good deeds, and reconciler of the Greeks. This histor is present and conspicuous in a group of narratives concerning the Greek embassy to Argos for the purpose of requesting support against the Persians (7.148– 52). Here the story includes several elements that are also present in the reports of similar unsuccessful Greek missions to Gelon (7.153– 67), Corcyra (7.168), and Crete (7.169– 71). The past divisiveness described by the narrative is moreover concomitant with a current controversy, which emerges metanarratively from the report of four variant versions. A small sampler of Herodotus’ methods for handling sources, this is also the fullest instance of judicial arbitration in the Histories.229 It leads to the formulation of a general principle on the subjectivity of controversial positions. Version A is the protagonist version, which “the Argives say about themselves” (7.148.2). Despite a discouraging oracle, the Argives promised to participate in the Greek resistance on the conditions that they obtain a thirty-year peace with Sparta and at least half of the command of the armed forces (7.148.2– 4). The peace was especially important to them, because their city had lost many men during Cleomenes’ attack and needed time for its children to grow to adulthood; they feared that if  would befall them at the hands of the Persians, another disaster (κακον)
ηκοοι)  they would end up subjected (υπ to Sparta (7.149.1). To the Argive demands, the Spartan envoys replied that the matter of peace would have to be decided by their assembly at home; with regard to the hegemonie, they could only offer the Argive king a share equal to that of the two Spartan kings. Hearing this, the Argives found the greed of the Spartans unbearable; considering that it was preferable to be ruled by the barbari
ιξαι) to the Spartans, they bid the ans than to subject themselves (υπε heralds to depart (7.149.2– 3). Version B is “another logos current through Greece” about an alleged previous event (7.150.1– 3): Xerxes had apparently sent an embassy to urge the Argives to remain neutral and to promise them special regard if 229. See Nagy 1990, 315.

226

Telling Wonders

he were victorious. For this reason, when the Greeks tried to enlist them in their cause, the Argives demanded the arche: they knew the Spartans would refuse, and the Argives wanted to have an excuse (προ␸α σιος, 7.150.3) for remaining at peace. A third story (B1), introduced as “another logos that some of the Hellenes tell about something that happened many years after,” is offered next as a corroborating proleptic appendix to version B. When the Athenian Callias and his colleagues happened to be in Susa “on another mission,” Argive heralds were also present. The Argives asked Artaxerxes whether the friendship they had formed with his father, Xerxes, was still valid, and they received warm reassurances from the king (7.151). What follows is a composite gloss containing the generalization and ending in yet a fourth version (C). 1. Whether Xerxes sent a herald to Argos to say these things [B], and whether Argive heralds went up to Susa and asked Artaxerxes about the friendship [B1], I cannot say exactly, nor do I  express any other opinion [γνωµην] about these things except that which the Argives themselves say. 2. But this much I know, that if all men were to bring together all  κακα] in the middle, with the intentheir own evils [τα ο ι κηια tion of making an exchange with their neighbors, after bending down to inspect the evils [κακα] of their neighbors, each of them would gladly bring back home those that they had brought over to begin with. 3. So, not even the Argives behaved in the most shameful way. And I have the obligation to report the things that are said, but I do not in the least have the obligation to believe them, and let this rule be valid for my entire logos. C. Because also the following is said, that it was the Argives who called the Persians into Greece, because the conflict with the Lacedaemonians had gone badly for them and they wished to have  anything rather than their present grief [λυπης]. (7.152.1– 3) The first statement (7.152.1) begins by declining to corroborate the indictments in versions B and B1. In light of the relative transparency and complexity of the Argives’ own version A, the narrator makes their gnome his own. Among various factors that may have contributed to their (at least potential) Medism, the most compelling appear to be their weakness resulting from the defeat they had suffered at the hand of

Interpretation and Evaluation

227


ηκοοι)  Cleomenes and their fear of becoming subjects (υπ of the Spartans (7.149.1). As we have already seen, to avoid stasis, the Athenians yielded on the issue of the hegemonie (εικον, 8.3.1). For the Argives, however, to
ιξαι, 7.149.2), would yield to the Spartans, even in a limited way (τι υπε have been uncomfortably close to submission. The point of view of the Argives forms the specific context of the maxim at 7.152.2 and of the added-on version C that closes the gloss. Herodotus’ lack of knowledge with respect to the veracity of the various versions is compensated by his experience (ε πιστα µενος) of a more general nature, which he now brings to bear on his interpretation of this case. The passage is difficult, because it combines the caution of historie with the obscurity of the ainos, as Nagy saw.230 At the most immediate level of meaning, the kaka that men compare in the gnomic saying are “misfortunes,” like the kaka Herodotus elsewhere attributes to war.231 The specific primary referent of the phrase oikeia kaka (domestic misfortunes) is represented by the Argives’ defeat by Sparta (called a kakon at 7.149.1) and their current vulnerability.232 The narrator implies that whatever the Argives did, they were acting under the pressure of their misfortune, a pressure made more severe by the subjective (and exaggerated) perception that people have of their own misfortunes in the absence of a suitable “marketplace” opportunity for revising that perception. Seen in this light, the behavior of the Argives is not as culpable (or “shameful” [αισχιστα]) as it could have been (7.152.3). This evaluation leads the researcher/ arbitrator to illustrate in version C a theoretically possible behavior from the side of the Argives that would be even more shameful than anything prospected so far but that would be similarly motivated by their previous  misfortune and their own perception of it (λυπη). But again, because something is circumstantially plausible and is said does not mean that it really happened. To the subjectivity of people’s motives for action corresponds the partiality of the accusations. Between the reports of versions B and B1 and the even more damaging version C, Herodotus recalls two rules that govern both his scientific/historical inquiry and its judicial counterpart. First, sources need to be heard. Second, 230. Nagy 1990, 315, citing Aesop’s fable “The Two Packs” (266 Perry) as a parallel to Herodotus’ formulation. 231. See especially 6.98.2, 8.3.1. This interpretation is supported by Legrand (1946, 7.151) and others. 232. Cf. especially 6.21.2. Threat of war and conquest is also implied in πα␽εα . . .  παθεα at 1.153.1. ο ι κηια

228

Telling Wonders

the logoi have only inconclusive evidentiary value, for they are likely to be politically biased in a historical context, just as they are culturally determined in an ethnographic one.233 In this particular case, then, the researcher/arbitrator’s self-referential glosses acquit the Argives by appealing to the unreliability of the logoi, just as his referential glosses of interpretation and evaluation at least excuse their behavior on the basis of the subjective validity of their own perception. The reading I have just given is, however, one-sided. Understandably, Plutarch appears more exasperated than usual in the face of this passage
ες

εν  υγι  αλλ

α παν  and applies to it the unforgettable verdict
Ελικτα κουδ 234  περιξ [all twisted, nothing sound, all back to front]. One reason why our interpretation does not fully satisfy is that it requires that the main point of the generalization at 7.152.2 must be the implicit idea “[People’s behavior should be excused because] men have an exaggerated perception of their own misfortunes,” rather than the explicitly described stage when one would hypothetically compare one’s kaka with those of one’s neighbors and decide to keep them after all. For the comparative operation to confirm or produce, rather than correct, the subjective perception of “all men,” the kaka each person brings to the middle must be “wrongdoings.”235 Then, the meaning of the generalization for the specific case is, first, that the Argives do not consider what they did all that bad in comparison to the bad actions of others. Second, those ready to convict the Argives of the most serious wrongdoings are under the same subjective perception: they are excessively harsh toward others and underestimate their own wrongdoings. The subjectivity of the source-protagonist is again counterbalanced by that of the accusing logoi. From Herodotus’ objective standpoint, “not even the Argives did the most shameful things,” with αισχιστα in the evaluation picking up on κακα (in the sense of α ι σχρα , “shameful”) in the maxim.236 The suggestion that others besides the Argives have kaka (wrongdoings) to account for applies both to the world of the narrated and to the extratextual world of the performance. In Herodotus’ narrative, The233. Cf. in an ethnographic context, 2.123.1 and 4.195.2, with similar glosses of noncorroboration. 234. Plut. De Malign. Herod. 28 ⫽ Mor. 863E, quoting from Eur. Andr. 448. The translation is by Bowen (1992, 57). 235. See especially Macan 1908, 1.1.209, followed by How and Wells 1928, 2:191. 236. Cf. the sophistic Dissoi Logoi, 90 DK 2.18 and 2.26, where τα α ι σχρα (shameful things) is what is being compared and found καλα (beautiful/honorable) by different people. See “Funeral Customs and Other Nomoi” earlier in the present chapter.

Interpretation and Evaluation

229

ban actions certainly appear more “shameful” than those of the Argives. In the real world of the narration, Argos had settled down to peace and inactivity, in the least culpable position imaginable at least since 451 b.c.237 By contrast, each of “the leading states warring with one another about the arche” (6.98.2) had been exerting itself in the hope of obtaining the help of Persia against the other.238 The Spartan king Archidamus in Thucydides perhaps formulates a typical argument when he says, “For all those who, like us, are the objects of designs from the part of the Athenians, it is not a matter of reproach to provide for their own survival [διασωθη ναι] by acquiring for our side not only Greeks but also barbarians” (Thuc. 1.82.1). For Herodotus, to seek the cooperation of Persia for the purpose of making war against other Greeks is a bad action.239 But in a world where different groups of Greeks rationalize this type of behavior for the sake of their mutual wars, even the most unlikely worst actions of the Argives, who feared Spartan domination, were not αισχιστα, that is, without parallel in their baseness. The maxim at 7.153.2 contains one of three instances of the phrase “all  men” (πα ντες ανθρωποι) in Herodotus’ metanarrative. In one of the other cases, as we have already seen, it appears in the very similar passage that envisions people comparing nomoi (customs) and finally choosing their 237. Argos was at peace with Sparta (see Thuc. 5.14.4) as well as with Athens (from 462: see Thuc. 1.102.4). Cf. Thuc. 5.28.2. Herodotus’ leniency toward Argos contrasts with his negative (though, as usual, unmarked) representation of Thebes (see, e.g., 7.205.3, 222, 233). Correspondingly, on the Athenian tragic stage, Argos, in contrast to Thebes, emerges as a city that “can be saved”: see Zeitlin 1990. 238. See Thuc. 2.7.1; cf. 4.50, perhaps later than Herodotus’ Histories. See How and Wells 1928, 2:190. See also Aristoph. Acharn. 61– 134. For the fateful Spartan embassy to Persia mentioned by Herodotus at 7.137 (cf. Thuc. 2.67), see “Divine Retribution” earlier in the present chapter. 239. Callias’ embassy to Susa mentioned by Herodotus at 7.151 was almost certainly part of the negotiations that eventually led to peace with Persia in 449 b.c. According to Samons (1998, 135), on that occasion Athenians and Argives went to Susa together after stipulating their alliance with anti-Spartan intent in 462/461. If Herodotus were implying as much, his mention of that mission would be another point in favor of the equivalence between the actions of the Argives and those of other Greek powers. However, Herodotus 7.151 perhaps mentions Callias and the Athenians with him to provide a clue to the source of this story, but otherwise clearly implies that their meeting with the Argives at Susa was coincidental. We can also speculate that Herodotus would have approved a cessation of the offensive war against Persia in and of itself (pace Badian 1993, 134), and that his statement that the Athenians were there “on another matter” is designed to dissociate Callias’ mission from the self-serving Argive embassy. On this difficult passage, see also Macan 1908, 1.1.209. The role Herodotus attributes to the Callias family is problematic also at 6.121, a passage we will discuss in chapter 4.

230

Telling Wonders

 own (3.38.1; repeated with παντων [ανθρ ωπων] at 3.38.4). The third case is the generalization that “all men know equally about the gods.”240 We should consider the interconnections among these three generalizations on the topic of beliefs and opinions held by different people as the sign of the contribution of ethnographic historie to Herodotus’ political ideology. The use of the phrase “all men” also links historie in the sense of “inquiry” and historie in the sense of “arbitration.” In each of the three cases, the phrase “all men” refers to a human community whose overall identity both depends on and transcends its internal differentiation. While at 3.38.1 and 2.3.2 the internal differentiation of the entity “all men” from the point of view of practices and beliefs coincides with the subdivision of humankind into different ethnea, the generalization at 7.152.2 implicitly replaces membership in an ethnos with membership in a Greek polis to account for differences in perception. Whether they are misfortunes or wrongdoings, the kaka are chosen, discarded, and evaluated in different ways on the basis of men’s political outlook as citizens of Argos, Athens, Sparta, and so on. The narrator’s statement of cultural relativism at 3.38 is designed to promote a cosmopolitan attitude in the listeners; it encourages them to overcome their cultural subjectivity in the only way it can possibly be overcome, namely, by recognizing its compelling universality as objective proof of the equivalent value of the different nomoi. The maxim at 7.152.2 follows the same model for the purpose of mediating disagreements among the Greeks and to provide the audience with a higher vantage point for judging themselves and others. The histor pursues these goals by means of a verdict that, in a characteristic way, explicitly acquits and implicitly convicts. The narrative of the Histories spares none of the Greeks. In this context, to apply the norm of subjectivity to a political quarrel is to invite the different parties to rise above that norm and reexamine with a more impartial outlook their own “bad actions.” Conclusion The generalizations in the voice of the narrator represent sporadic but precious indicators of the substance of the message that the Histories as a whole wants to communicate to the Greeks. At the highest level of inquiry, we find a tenacious attempt to verify and demonstrate certain truths that 240. 2.3.2. See “Equal Knowledge” and “Funeral Customs and Other Nomoi” earlier in the present chapter.

Interpretation and Evaluation

231

belong to the supracultural theological sphere: the fluctuations of human fortune with respect to individuals and states, the moral participation of the divine in these processes, and the willingness of the gods to share some of their knowledge with humans. The central problem of human misfortune in the overarching context of the Histories is represented by war and the painful confusion between the need of any state to cultivate the values and the resources that will enable it to maintain its freedom, on the one hand, and the agonistic impulse to fight for the acquisition of rule, on the other. Concomitant with and symbolic of the unnecessary armed conflicts of Greeks against Greeks are the verbal and ideological quarrels in which claims, counterclaims, self-justifications, and accusations are all a part of a rhetoric of mutual aggression. Here Herodotus establishes an implicit parallel between Greek ethnocentrism and Greek polis particularism. Just as the Greeks cannot seem to find a balance between the awareness of their own cultural worth and the systematic misunderstanding and disparagement of foreign cultures, so they define their political identity on the basis of a hegemonic polis ideology. The connection Herodotus draws between antagonism toward the barbarian other and antagonism toward other Greeks may seem paradoxical if we consider that Panhellenism “in itself accentuated rather than softened the distinction between Greek and barbarian.”241 The connection is, however, fully justified, for example, in Athenian uses of Panhellenic ideology as an imperialistic tool vis-a-vis ` other Greeks and in the confusion of the contemporary Greek political discourse in general concerning who the other might be.242 Herodotus’ authority rests on his knowledge of the world and on his lack of allegiance to any particular state. From this vantage point, he teaches his audience the ambivalent character of reality and provides them with a model of arbitration for negotiating their differences.

241. Guthrie 1971, 162. 242. Cf. Antiphon DK 87 B44, frag. B: “we are barbarians to each other.” For the relation between Panhellenism and imperialism, see Perlman 1976.

Chapter 4

Thoma

Comparison and interpretation are associated with Herodotus’ overarching goal of explaining. The first sentence of the Histories, however, situates  α ι τι ην [for what reason]) within the explanation (represented by δι ην much broader context of a celebration of achievements great and wondrous (θωµαστα ) across the inhabited world. Celebration and narratability overlap with explanation but refuse to be restricted by it. Explanation provides direction without entirely controlling either the pace of the logos or the paths it will take. Among the advertisements that throughout the work keep reestablishing the autonomy of narratability vis-a-vis ` explanation, words of wonder occupy a special place: they alone name an emotional response the narrator wants the recipient to share.1 This is not to say that in Herodotus’ speech act, words like thoma are always strictly    rationed and precisely targeted.2 But θωµα, θωµα . . . µοι, θωµα µεγιστον, and the verb θωµα ζω are the most emphatic and mysterious directions the narrator provides. They tend to announce special semiotic challenges and occasionally mark the highest philosophical level of the inquiry. The wonder the Histories are designed to inspire is not (at least not openly) self-referential, that is, directed at the logos of Herodotus. Aside from one or two cases when wonder is directed at someone else’s logos, the object of wonder belongs to the world of the narrated, which in turn imitates the real world. Nevertheless, mention of the wonder response may create, from the point of view of the recipient, a wondrous discourse event, something that surprises primarily within the discourse. Thus, 1. On the issue of what wonder is and whether it constitutes an impression registered “in the heart” or “in the brain,” see Greenblatt 1991, 15– 22. For the notion of wonder in Herodotus, see especially Barth 1968; Dewald 1987, 154– 55, 165; Hartog 1988, 230– 37; Payen 1997, 117– 28.   2. E.g., the expressions αξιος θωµατος or θωµα σαι αξιος used as metanarrative qualifiers within narrative statements (see, e.g., 1.185.3, 3.47.2, 3.113.1, 4.199.1, 4.53.3) are    often as casual as αξιοθ εητος or αξιαπ ηγετος, αξιος λογου, and so on. See Barth 1968, 98– 99.

232

Thoma

233

when I examined Herodotus’ description of the leather boats of the Assyrians, I was proceeding backward.3 In cases of this sort, the phenomenon itself perhaps causes less wonder than the text’s injunction that we,  along with the narrator, should wonder (θωµα µεγιστον µοι). This in turn leads to interpretation and, retrospectively, to a redirection of our wonder response to the object. Herodotus presents to his audience both ethnographic and historical items as wonders. In the history, appeals to wonder draw attention especially to exceptional actions or agents—onetime behaviors of individuals and animals or occurrences that appear to reveal the intervention of the  divine. In the ethnographies, a θωµα tends to be a tangible foreign artifact, a phenomenon of the landscape or a feature of the flora and fauna of distant lands. Metanarrative thoma is rarely used to describe either the activities of foreign peoples in the ethnographies or those of foreigners in the history.4 In the context of their respective cultures and through Herodotus’ rhetoric of similarity or equivalence, exotic actions and customs want to be interpreted as normal. Unlike the narrator, characters are frequently in wonder at the behaviors, utterances, or appearances of foreigners, because they are different.5 On one occasion, Herodotus ex pects that some of his listeners will experience great wonder (θωµα µεγιστον) at a Persian action that is too similar, that is, reminiscent of the contemporary policy of a Greek, not a barbarian, superpower.6 He implies no doubt that some may not even believe it; but for those who do, this unexpected event will invite reflection. Whatever is or appears to someone exceptional or strange represents, 3. See my examination of 1.194 in my introduction. 4. At 2.35, the abundance of thomata in Egypt is paralleled, rather than illustrated, by the list of customs that are “opposite to those of the rest of the world.” The song of Linus (2.79.2– 3), the collection of spices in Arabia (3.111.2, 112), and the boats of the Assyrians (1.194) are the only other items termed thomata in metanarrative that have something to do with customs. As Payen observes (1997, 118), none of the Scythian customs, e.g., as strange as they may seem to the reader, belong to the semantic register of thoma: Herodotus denies that there are many thomata in Scythia, and customs are not included among the few (4.82). 5. E.g., the Fish Eaters wonder at the long life of the Ethiopians (3.23.2), Heracles at the Mixoparthenos (4.9.2), the Scythians at seeing the Amazons (4.111.1), Darius at the actions of the Paeonian woman (5.13.1), and Xerxes’ scout at seeing the Spartans combing their hair and doing exercises at Thermopylae (7.208.3). 6. The action is Mardonius’ establishment of democracies in Ionia after the revolt (6.43.3, with back reference to 3.80.1). See chap. 2, “The Sameness of the Lydians”; chap. 3, “Explicit Evaluation,” n. 170.

234

Telling Wonders

at any rate, a stimulus for further thinking. Whether in the ethnography or in the history, once the metanarrative labels any particular fact a thoma, that fact tends to jump out from its narrative context but nevertheless needs a context in which to make sense.7 It is not explained but demands somehow to be explained and to participate, in its turn, in the text’s network of explanations. It provides an impulse to mental inquiry, much as, in Aristotle’s formulation, wonder provides the impulse to philosophy.8 It is an inquiry, however, that the text declines to actualize but implicitly identifies as the task of the recipient. Herodotus and the Conventional Code of Ethnographic Wonders Scholars have not sufficiently explored Herodotus’ notion of wonder. The tendency has been rather to regard wonder as a self-justifying response in the face of what is unusual according to commonsense Greek assumptions of normalcy. The use of wonder words in the Histories even represents, according to Hartog, not so much the narrator’s signal for the audience to share his response to an object as Herodotus’ own response to audience expectations of the most conventional sort. This topos of traveler tales of all ages, he argues, was commonplace also in ancient ethnography. Because people like to be amazed, the narrator adopts a rhetoric to bolster his claim that he is qualified to amaze them.9 Hartog’s own analysis of narrative sections of the Histories, including reports of wonders, is clever and profound; but it proceeds against the text and with little regard for Herodotus’ communication of meaning to his audience. Wonder words in the Histories are often signals that something has a special meaning. This is not to deny the existence of a previous ethnographic wonder 7. See Greenblatt 1991, 2– 25, on the “anecdotal” character of all literature of marvels. 8. Arist. Metaph. 1.2.8– 9 [982b]; see also Plato Thaet. 155d. Cf. Redfield (1985, 103) on Herodotus: “Wonder is the beginning of wisdom when it leads to further thought.” 9. “Thoma may be reckoned among the procedures used by the rhetoric of otherness. Generally speaking, the impression it conveys is one of trustworthiness, for the narrator cannot fail to produce this rubric, which is expected by his public. To omit it would be, at a stroke, to ruin his credibility. It is as if it were postulated that far away, in these other countries, there were bound to be marvels/curiosities” (Hartog 1988, 231). Cf. Strabo’s accusation (11.6.3) that accounts of marvels by Herodotus and others pander to the desires of the audience. For wonder as a conventional topos of ancient ethnography, see Jacoby 1913, 331– 32; Jacoby maintains that ethnographic logoi were organized to include a description of the nature of the country, a survey of customs, the mention of thomasia, and finally a section on political history.

Thoma

235

tradition. Though we know too little about it to make a proper comparison, Herodotus’ awareness of this tradition especially emerges from three metanarrative statements that refer to audience expectations by saying that this or that country has or does not have many thomata or thomasia.10 The most expressive of these, for our understanding of Herodotus’ adaptation of the conventional code of wonders to his own celebratory code, is the following introduction.  the land of Lydia does not have that As for wonders [θωµατα δ ε], much suitable for description [ε ς συγγρα␸ ην], compared to other countries, with the exception of the dust washed down from Mount Tmolus. But it offers one building that is by far the biggest, not counting those in Egypt and Babylon. (1.93.1) Placed at the beginning of the sentence and followed by the name of the term θωµατα the country and the programmatic word συγγρα␸ ην, here functions as one of the summarizing elements in a title more or less equivalent to “Description of Lydia: Wonders.” But as it turns out, this announcement belongs to a programmatic correction. Not much else fits the topic in the normal sense, says the narrator, and not all available items are interesting. The absence of thomata in Lydia corresponds to this country’s cultural similarity to Greece. Difference from Greece is, then, one expected re quirement for θωµατα of foreign lands.11 “The dust from Tmolus” would qualify in this sense, but Herodotus mentions the phenomenon as notorious and immediately moves on. By the standards of ethnographic narratability, this golden sand washed down by the river Pactolus presumably belongs to the same order of phenomena as other natural treasures of the East. Herodotus gives a lengthy description of the harvesting of gold in India.12 This follows a pattern similar to that of the harvesting of Arabian spices by various methods, each more “wonderful” than the 10. See 1.93.1, 2.35.1, and 4.82; Jacoby 1913, 331; Hartog 1988, 231. θωµα σιος, even and θωµα,  more than θωµαστος seems to be a technical term in the ethnographic tradition. See Barth 1968, 108. But in Jacoby’s classification (see n. 9 in the present chapter), it is not clear, e.g., to what extent the rubric thomasia would have overlapped with the others in ethnographic writing before Herodotus. 11. See Hartog 1988, 231– 32. For Lydian customs similar to those in Greece, see 1.94.1. 12. 3.102.2– 105. Cf. Herodotus’ interest in the gold of Ethiopia (3.114) and northern Europe (3.116.1).

236

Telling Wonders

other.13 At the edges of the earth, where people live apparently untrammeled by societal structure and in direct contact with their environment, the ethnographer is closest to an anthropological mode of inquiry. The Indian-Arabian sections focus on the ingenuity and labor of man in his natural state, on the coexistence of opposite extremes, and on the relations and correspondences that appear to exist between animals and human beings.14 According to Herodotus’ interpretation, these ethnographic and zoological data indicate that the world of nature in general and particularly the customary behaviors of animals are regulated according to a balance that is intelligent and providential;15 this leads to the implicit suggestion that, at some level, the customary behavior of men translates into the area of nomos by virtue of this providential design. The methods for collecting rare products are perfectly good wonders in conventional terms. But for Herodotus in this “ends of the earth” passage, they also provide additional evidence for defining custom in its essence as paradoxically both natural and divine. This confirms the unitarian view, previously theorized, of “Custom king of all.”16 Lydia, by contrast, is a land of the center. The lesson it ethnographically teaches is not anthropological but political and historical. In this context, Herodotus rejects as not significant the natural phenomenon of the gold from the Tmolus. He even fails to discuss it as the source or emblem of the famous Lydian wealth that impressed the Greeks and that 13. See 3.111.1, 3.112; cf. 3.113.1. On the structural similarity of the methods of collecting spices, see Detienne 1977, 14– 20. 14. The last is a concern that, for different reasons, Herodotus was also able to pursue in Egypt. See chap. 2, “The Texture of Nomos.” 15. See 3.108; cf. 3.106.1. See Immerwahr 1966, 312. 16. See 3.38, again based on a testing case provided by a population of the southeastern edges, the Callatian Indians. See chap. 3, “Funeral Customs and Other Nomoi.” The ingenuity of the primitive harvesters of spices is at one point denoted with the cultured term σο␸ ι ζεσθαι (3.111.3). This term also describes the behavior of cats in the section on Egyptian animals (2.66.2). These two uses of the verb bridge the gap between cultural activity (“devices”) and elementary impulses. Similar theoretical interests govern Herodo tus’ “great wonder” [θωµα µεγα] (3.12.1) at the softness of Persian skulls and hardness of Egyptian skulls on the battlefield of Papremis. He attributes this contrast to the opposite customs of wearing a head covering or shaving the head: whereas his observation of primitives allows him to verify the ultimate derivation of culture from nature, here he notices changes in nature as a result of culture. Herodotus’ most theoretical glosses on anthropological subjects tend to occur in book 3 for two reasons, both connected with the progress of the historical narrative: here the actions of Cambyses, especially in relation to Egypt, raise the question of the absolute value of nomos (and therefore of its connection to phusis); and here the conquests of Darius bring the narrative to the extremities of the earth, a suitable field for such discussion.

Thoma

237

represents an important theme in his own historical narrative about  item the narrator agrees to valorize in this Lydia.17 The one and only (εν) section is the monumental tomb that the people built for one of their kings. This oversized building both qualifies as a thoma in the conventional sense and serves, as we have seen, as a more analytical symbol of Lydian society and its structure, of the single disturbing custom in which Lydia differs from Greece, and of the oppressive monarchy that will ensure the country’s eventual loss of freedom.18 Thus, the introduction to the first properly ethnographic section of our text indicates that Herodotus accepts a canonical aspect of ethnographic narrative, the report of thomata, but will subsume and adapt it to his representational needs. When Herodotus berates those Ionians who maintain that the Nile’s floods are due to its origin from Ocean (2.21), his polemic may be related to this redirection of the notion of ethnographic “wonders.” He calls the theory “more fit for an account of wonders”  [θωµασιωτερη] but also “more inept” [ανεπιστηµονεστ ερη] than a theory refuted earlier.19 Other than factually inaccurate, it belongs to those geographical speculations about the shape of the earth that Herodotus considers ideologically misleading.20 He therefore declares the “won αξι  ω µνησθηναι, drous” logos itself as “not worth mentioning” (ουδ 2.20.1) except to reject it. The oxymoron registers the narrator’s objection to an indiscriminate literature of wonders. The theory that the Nile derives from Ocean has the flaw of bringing  2.23) with no the account into the realm of the invisible (ε ς αφαν ης, possibility for verification. But just as a verifiable thoma may not be a useful tool of representation to Herodotus (the golden dust of the Tmolus is real but uninteresting), so the fact that a wonder is unverified does not automatically disqualify it from being meaningful. The oracle shrine of Apollo and Artemis at Buto, which is “worthy of logos” as a whole, includes two orders of thomata, visible and unseen. The temple in the precinct of Leto, made of a single block of stone, is programmatically ˆ 17. See, e.g., 1.14.3; 1.50.1– 3; 1.51.1– 3, 5; 1.52; 1.92.1. A heap of this golden ψηγµα is presumably what Alcmaeon falls on when he visits Croesus’ treasure house (6.125.4). For gold as an index of royalty, see Kurke 1995, 45– 51. 18. 1.94.7. See chap. 2, “The Sameness of the Lydians” in the present book. 19. Cf. 2.23. For the attribution of the theory to Hecataeus, see FGrHist 1 F 302; Lloyd 1976, 100. 20. Cf. 4.36.2– 45 and especially 4.36.2; at 4.42.1, Herodotus’ rejection of the Ionians’ geographic constructs is phrased in terms of his “being in wonder” [θωµα ζω]. See chap 2, “The Texture of the Earth,” in the present book.

238

Telling Wonders

advertised (␸ρα σω) as “as far as I am concerned, among the things that  can be seen, . . . the one that causes the greatest wonder” [το δ ε µοι των   ␸ανερων θωµα µεγιστον παρεχοµενον] (2.155.3). The combination of wonder “to me” and visibility “to me” is repeated in the conclusion of  των  νυν ο! νηος  ␸ανερωνµοι   the very brief description (ουτω µεν των ε στι θωµαστοτατον,  το ι ρον περ ι τουτο 2.156.1). After the temple, the nearby island of Chemmis, on a lake of the Delta, is a second wonder (2.156.1– 2). Here one finds a great temple of Apollo, three altars, and also “palm trees and many other trees, some bearing fruit, some not” (2.156.3). Although this is still all part of the visible, we also are told with that is, it great verbal redundancy that the island is supposedly πλωτη; floats. The narrator has not seen this phenomenon, but the Egyptians say it does float (2.156.2). They tell a story to explain this, namely, “that Leto came to this island at the time when it did not float, and having received Apollo in trust from Isis, she saved him from Typhon by hiding him in this island that now reportedly floats” (2.156.4). The sanctuary of Buto testifies to the prominence in Egypt of three gods who are also major divinities in the Greek pantheon and are similarly connected to one another in myth and cult. The flight of Leto to the (perhaps floating) island of Chemmis reproduces the wanderings that lead Leto to the floating island of Delos in the Greek tradition.21 The palm trees of Chemmis are singled out for mention because they expose what is for Herodotus the ultimate proof of the mutual interface between different religions, the correspondence of specialized signs.22 No glosses of comparison appear in this passage, but the subtext of the description of the Buto sanctuary has to do with the horizontal and synchronic similarity between Egypt and Greece that Herodotus pursues throughout the Egyptian ethnography. In a context of preexisting relations, palm trees and other features at Buto and Chemmis fulfill the same function as the existence in Egypt of the song of Linus. The latter is hard to fathom as an ethnographic thoma in conventional terms but is for this narrator ζειν µε).23 Buto, however, also encodes  occasion for wonder (αποθωµα the otherwiseness of Egypt, even its oppositeness from Greece. Over 21. See Hom. H. Ap. 3.25– 138. In a late version of the story (Hyg. Fab. 140), Leto is pursued by Python, which would increase the similarity with the Egyptian story told by Herodotus. The correspondence between Greek and Egyptian tradition here accounts for Herodotus breaking his own rule of silence concerning “divine things” (2.3.2). 22. See Hom. H. Ap. 3.18, 113. For the Greek association of the palm tree with Artemis, see Sourvinou-Inwood 1991, 99– 123. 23. See 2.79.1– 2 and chap. 2, “The Texture of Nomos,” in the present book.

Thoma

239

there, palm trees are “among many other kinds of trees, some bearing fruit some not”; here, a single palm tree sprouts in a rocky land.24 In Egypt, Leto is the nurse of Apollo and Artemis, not their mother, and these two are the children of Dionysus and Isis.25 The floating nature of Chemmis contributes to this tension between what is same and what is different: whereas Delos used to float but no longer does (it is in fact by definition akinetos: see 6.98.3), Chemmis did not float but is said to do so now.26 Unlike the Nile’s origin from Ocean, therefore, the supposedly floating Chemmis is a wonder to Herodotus, as it perhaps was to Hecataeus.27 Herodotus’ polemic against Hecataeus, if polemic is implied here, limits itself to an emphatic gloss of noncorroboration that acknowledges the scientific distinction between phenomena that are verified by eyewitness, phanera, from those that are learned through oral reports: “I personally did not see it either move or float, but, hearing this, I am AMAZED [τεθηπα], if it truly is a floating island.”28 In the description of Scythia, a land with no θωµα σια except for the number of its rivers and the vastness of its plain, something that is indeed visible but cannot be what people say it is counts nevertheless as worthy σαι . . . αξιον)   of wonder (αποθωµα and preserves its meaning. They show [␸αι νουσι] the footprint of Heracles, stamped on the rock; it resembles a man’s footprint but has a size of two cubits . . . (4.82) 24. The infertility of Delos is an essential element of the myth of the birth of Apollo (see Hom. H. Ap. 3.48, 55, 60, 72). 25. An opaque gloss of testimony recruits Aeschylus to the system of interconnections  (see 2.156.6). Here the verb ηρπασε both recalls the famous abduction of the Demeter and  Persephone myth (cf. ηρπαξεν in Hom. H. Dem. 2.3, etc.: by snatching the Egyptian logos that Artemis was the daughter of Demeter, Aeschylus has abducted Artemis to replace the Kore of Greek tradition) and participates in Herodotus’ polemic against Greeks who exploit Egyptian cultural knowledge without giving it due credit (2.123.3). For the view that Aeschylus’ version in this gloss has something to do with the substance of the Eleusinian mysteries, see Mazzarino 1966, 97 and n. 124. 26. See especially Pindar frag. 33d SM on the transformation of Delos as a nonfloating  ι νητον τερας). island (ακ On the symbolic significance of the immobility of Delos, see Wood 1972, 141 n. 55. 27. Hecataeus describes Chemmis in remarkably poetic terms: εστι δ ε η! νησος µετα  ρσι η κα ι περιπλει κα ι κινεεται ε π ι του υδατος (FGrHist 1 F 305). 28. 2.156.2. There is no evidence that Hecataeus expressed disbelief of the fact that the island floated. The verb τεθηπα appears only here in the Histories. Like θωµα ζω, it follows  See Smyth 1956, the most frequent construction of verbs of emotion with ε ι rather than οτι. 2247. The narrator is amazed at the fact that the island floats, if it does, not wondering whether it does.

240

Telling Wonders

Scythia’s scarcity of wonders means poverty of material culture in contrast to the powerful kingdoms of the East. The excepted items are natural wonders, the plain and the rivers, elsewhere described at length and representing the elusiveness and insularity of the Scythians, in whose land Darius’ army will roam unproductively and risk being trapped.29 But in relation to the Greeks, as we have seen, what makes Herodotus’ Scythia most insular and foreign does not preclude and sometimes even favors the notion of a special similarity between the two peoples. Elsewhere Herodotus even reports an uncorroborated story “the Black Sea Greeks say,” according to which the Scythian nation was born from the marriage of their wandering hero Heracles to the Mixoparthenos, a local maiden half woman and half snake.30 Here the footprint “they show” (appropriately huge to match the landscape and the hero’s stature) is not a sort of memorial but rather the visual symbol equivalent to that fictional event. In Herodotus’ description of Scythia, it is a pure narrative sign, and as a sign, it represents, from the point of view of the text, a thoma. The traditional topos of ethnographic wonders could provide Herodotus (and no doubt Hecataeus and others, though in ways we do not know) not with material for amazing his audience “ineptly” but with a rich reservoir of symbolic forms. One of the conventional attributes of thoma must have been µεγαθος, size, which is hardly devoid of significance, including in a political sense.31 Already in the archaic period, the Greeks perceived foreign lands, especially the East, as “big” and themselves as “small,” with the related invidious problem that bigger is evidently and devastatingly better—and has the better of you—though in the end “small” turns out to be better and stronger.32 Bigness in nature is dwarfing and is sometimes related to wilderness and fear of the unknown, as in the case of the Scythian rivers and plain. In cultural artifacts, it often signifies wealth, power, or both at once.33  ντων . . . θωµασιωτα 29. See 4.47– 59, 4.99– 101. So the Pontus is πελαγεων . . . απα τος (4.85.2). 30. 4.8– 10. See the indirect noncorroboration at 4.10.3– 11.1. See Vandiver 1991, 172; Georges 1994, 1– 9. 31. Hartog (1988, 234) observes that the quantitative aspect of thoma is confirmed by the frequency with which it is described in terms of measurements. $ κατα κοσµον/ο  σµικρη 32. Phocylides writes (frag. 4 Diehl), πολις ε ν σκοπελω ι κευσα  κρεσσων Νι νου α␸ραινο υσης [A city that is small but on a lofty promontory and well ordered is stronger than foolish Niniveh]. 33. Consequently, it signifies the “greatness” of the builder. See Immerwahr 1960, 265 on physical erga.

Thoma

241

Numerous large buildings in the Histories are recruited in the service of the monarchical code, and a few of these are designated with words of the thoma family. But Herodotus’ signs are more subtle and varied than this simple correspondence. Babylon and its dam built by Nitocris—this last an achievement that is wondrous specifically because it is big and tall     ι , 1.185.3)—are coun(αξιον θωµατος, µεγαθος κα ι υψος οσον τι εστ tered by the θωµα µεγιστον of the Assyrian boats, which are also, peacefully, oversized (1.194.3). In Egypt, the gigantic proportions of Eastern art and architecture are adapted to the grandiosity of nature and testify to the high degree of technical skill of this ancient culture.34 Here Herodotus insists on measurements as part of his polemic against what Froidefond calls the “malevolent snobbism” of the Greeks, who minimize Egyptian things and even shrink Egypt’s territory.35 The pyramids, a product of monarchic oppression, are outranked on the scale of thomata by the labyrinth, a collective monument of twelve just kings, and by the yet more wondrous Lake Moeris; one is the cultural match of the land of Egypt, the other of the Nile.36 Lloyd remarks how many aspects of Herodotus’ attitude toward Egypt are revealed in his description of the labyrinth (2.148)—in his display of sources, admiration of size, and disparagement of the Greeks.37 From a narratological perspective, we should notice how conspicuously the persona of the narrator has invaded his subject. He celebrates, groups, excepts, compares, compares again, and ranks. The labyrinth is “too great for logos” [λογου µεζω], and indeed there is relatively little logos here in the sense of description or “story.” Predominant are glosses of historie, that is, the narrative of the journey of the researcher/narrator, who soon multiplies into “we” for maximum interpretive and narra ι τε torial authority.38 Using all the available verbs of seeing (ε'ιδον, αυτο ! ωµεν, ! ωµεν),   ι θεησα µενοι, αυτο  ι ωρ  ωρ αυτο Herodotus reports that he/ 34. The propylaia of the temple of Athena at Sais built by Amasis are θωµα σια (2.175.1), but Herodotus especially wonders at Amasis’ achievement in transporting the monolithic chamber from Elephantine (µα λιστα θωµα ζω, 2.175.3). 35. On Greek diminutives for Greek things, see Froidefond 1971, 122– 23; Lloyd 1976, 310. For the size of Egypt, see especially Hdt. 2.6– 11, 15– 18. 36. See 2.148– 49; Benardete 1969, 63. 37. Consider 2.148.2: “for if one were to gather all the walls and display of works of the Greeks, they would appear to be inferior in labor and expense to this labyrinth.” See Lloyd 1993, 2:148. 38. See Chamberlain, forthcoming, on the use of “we” not in reference to an ethnographic group but as a projection, operating only on the surface of the text, of the narrator’s definitive voice and authority.

242

Telling Wonders

“they” personally saw at least the upper chambers, fifteen hundred in number and “superior to all human works”; he was not allowed to see the lower chambers but learned by hearsay (λογοισι ε πυνθανοµεθα,  η παραλαβοντες) that they contained the royal burials and sacred ακο$  crocodiles. Going through (διεξιοντες, διεξιουσι) “from the courtyard into the rooms and from the rooms into the columned porches, and from the columned porches into other vestibules, and from the rooms into other courtyards” (just as the narrator walks through cities great and small and along the paths of the logos), he/“they” experienced infinite 39  wonder (θωµα µυριον). The labyrinth represents the limiting case of glosses clustering around the term thoma to connote Herodotus’ ownership of the wonder response and consequently also his independence from the wonder tradition of ethnographic writing before him. The object of wonder is first and foremost something he sees, hears, crosses, learns, and tells.40 It is a wonder to him (µοι), and he is the one who experiences wonder (θωµα ζω). Even in the face of what I have called “conventional” wonders, Herodotus’ impulse to wonder is by definition idiosyncratic and proceeds from the one who is in charge of the logos—my wonder/my logos. Wonder theoretically frees Herodotus from tradition, chronology, his task to explain, and various other constraints; in fact, it competes with these factors as one of the requirements or appetites of the logos. Hartog (1988, 234) noticed the “connection between thoma and digression” in the following passage, where advertisement and program introduce an ethnographic thoma that lies clearly outside the expected range of marvels in distant lands. So, there [i.e., in Scythia] these phenomena occur because of the cold. But I wonder [θωµα ζω]—for indeed my story sought digres  δη µοι ο! λογος  ς sions from the beginning [προσθηκας γαρ ε ξ αρχη ε δι ζητο]—for what reason in the whole territory of Elis mules are not able to be born, when the place is not cold and there is no other visible cause. The Eleans themselves assert that it is as a result of a curse that mules are not born among them. (4.30.1) 39. 2.148.5– 6. The participles could imply an indefinite subject (“anyone going  through”), but the main verb in the past tense (“they caused wonder” [θωµα παρει χοντο]) and the previous first-person plural finite verbs suggest “to us as we were going through.” As Payen remarks (1997, 120), “la vue est associ´ee a` la marche de l’enquˆete . . . elle se confond mˆeme avec le proc`es de l’´ecriture.” 40. See Dewald 1987, 155, n. 21.

Thoma

243

Wondering Why The semantic field of thom- words, especially the verb thomazo, includes an interrogative mode. The question being asked is about the meaning of the thing in a broader, real-world context. In most cases, it remains implicit and ill defined, but at 4.30.1, it is explicit and formulated in terms of why: why are there no mules in Elis? There are no mules or asses in Scythia because they cannot withstand the cold. This belongs to a natural state of affairs, whereas the absence of mules (not asses) specifically in Elis (not in neighboring regions: see 4.30.2) remains a mystery. The divine, which ultimately rules nature (3.108), operates in predictable ways so that phenomena can be seen to derive from material causes. An  αλλου  effect that breaks the pattern and has no visible cause (εοντος ουτε ␸ανερου α ι τι ου) leads one, by default, to the immediate and extraordinary agency of a transcendent force.41 The local inhabitants of Elis resort precisely to this explanation by maintaining that there are no mules in their country because of some curse. The vagueness of their answer makes it all the more valuable. Because the historical and ethical circumstances of the Elean curse remain obscure, the hypothesis of supernatural causes rests primarily on the observation of the irregularity itself.42 This constitutes, therefore, an entirely disinterested and independent item of evidence, a tool of persuasion Herodotus can use at a distance to deal with an issue of great importance to the overarching message of the work. If direct divine intervention can be deduced ethnographically from chronic irregularities in the natural world, that assumption can be perhaps extended to sporadic historical occurrences, even when the laws of nature are not violated so drastically or at all.  It is a wonder to me [θωµα δε µοι] that when they were fighting next to the grove of Demeter, not even one of the Persians apparently went into the sacred precinct or died there, and it is around the sanctuary that most of them fell, on unhallowed ground. And I am 41. So, people in Scythia “are in wonder” when it thunders in winter and regard it as a prodigy (τερας, 4.28.3). A unique case of rain in Upper Egypt was a great portent (␸α σµα µεγιστον) concomitant with the Persian conquest (3.10.3). 42. This is unlike, e.g., the ethical and historical reason of the permanent ethnological abnormality of the Scythian enareis (see 1.105.4). Animals as well as human beings are momentarily affected by transcendent causes as a result of human wrongdoing at 1.167.1, 6.139.1, and 7.171.2 (cited by Smith 1992, 9, 53, and n. 5). For the absence of mules in Elis as a taboo rather than an exception in nature, see Nagy 1990, 336.

244

Telling Wonders

of the opinion, if one must have at all an opinion about divine things, that the goddess herself did not let them in, because they had burned her temple at Eleusis. (9.65.2) There are no mules in Elis, no Persians in Demeter’s precinct. The answer to the question of cause must both times remain provisional, but cases of the first type are scientific corroboration for the more ideologically charged cases of the second. From a scientific perspective, animals embody elemental normalcy in nature as projected by the divine and provide a standard to measure the more complicated world of humans. Animals display a range of customary behaviors in the same way as humans do, except that these animal behaviors are not subject to the cultural processing that results in human nomoi. As a field on which god operates more directly, the animal world represents an intermediary between the divine and the human realms.43 Translated into historical terms, this view encompasses the idea that, on the one hand, animal events reflect human events and, on the other hand, the ways in which they do so mysteriously register divine reaction. From the point of view of Herodotus’ narrative, animals are also mediators between ethnography and history; they participate, moreover, in both the theological and the symbolic codes of explanation. In an episode of the narrative of Xerxes’ march against Greece, lions in Thrace attack the camels of the Persian army (7.125). Whereas in the case of the absence of mules in Elis the discourse needed to justify an abrupt narrative transition to a different place, here it works just as hard, in a way that appears entirely optional, to break the logical continuity of the narrative for the purpose of bringing the event into prominence. It starts, While Xerxes was passing through, lions attacked the camels carrying the provisions. This sentence is pure nonnarrated narrative; as a moderately interesting entry in the journal of the march, it could have been left at that. But a gloss follows, attached by γα ρ, with added narrative details:44 43. See Smith 1992, 7, 32– 33, and passim. Smith has counted 804 references to fauna of some kind in the Histories, 600 if we exclude horses. 44. The structure of the discourse in the first two sentences of 7.125 is analogous to the sequence of summary introduction followed by narrative; but see chap. 1, “What Is Metanarrative,” in the present book.

Thoma

245

 . . . for, coming down at night and leaving their homes [ηθεα], the lions touched no other animal or man and worked havoc only on the camels. Here, the phrases “no other animal” and “only the camels” incline toward the interpretive. But the best is yet to come, in the form of a concluding gloss where explanation is replaced by interrogation in an advertisement of narratability with the narrator’s first person. I am in wonder as to the cause [θωµα ζω δ ε το αιτιον], whatever it ζον] the lions to stay away from the  was that compelled [το αναγκα rest of the army and attack the camels, animals that they had never seen before and of which they had no experience [ε πεπειρεατο]. By repeating the substance of the story three times, the discourse implies that these lions did not behave like normal animals. The gloss “I am in wonder as to the cause” [θωµα ζω δ ε το αιτιον] formulates an enigma parallel to that of the absence of mules in Elis (“I wonder why . . . in the  αλλου  absence of other visible cause” [θωµα ζω ο τι . . . . εοντος ουτε   φανερου α ιτ ιου], 4.30.1– 2). If this event also is supernatural, given the context, it perhaps constitutes an omen; if so, we are invited to decode it appropriately. What is its significance? The phrase “camels carrying the provisions” is bound to recall the prominence of the issue of supplies for Xerxes’ army in the preceding narrative, including Artabanus’ warning on this matter.45 His prediction will have no fulfillment, but it underlines Persian vulnerability abroad. More importantly, camels are the most exotic part of the Persian army: Cyrus uses them, for example, to frighten the Lydian horses and defeat Croesus at Sardis (1.80). They are a synecdoche for the strange invader himself. Lions symbolize lethal behavior and a display of strength, with polyvalent connotations.46 In a political and social setting, a lion signifies kingship and dominance or, in a negative sense, tyranny and any destructive power that acts from within.47 In response to a prophecy, Meles, king 45. See 7.49.5, 7.50.4. See also, in the voice of the narrator, 7.83.2, 7.118– 20, 7.187. 46. The lioness at 3.108.4 is “strongest and most daring” [ ι σχυροτατον κα ι θρασυτατον]; her cub, when still unborn, destroys her womb with its claws. 47. In the fight between the lion and the dog at 3.32, the fighting dog represents Cambyses’ brother Smerdis, the dog’s brother is the opposite of Cambyses, and the lion is

246

Telling Wonders

of Lydia, tries to make Sardis impregnable by bringing a lion around the walls. Hipparchus and Cypselus are represented as lions—the first in a dream, the latter in an oracle.48 Lions in the face of an external enemy, however, connote a valorous offensive stance and fierceness in war.49 The lions who slaughter the camels in Xerxes’ army leave their haunts (called  ηθεα) and take on an exotic opponent of which they have no experience  (ε πεπειρεατο). They are the short-range precursors of the Spartans at Thermopylae, who leave their Peloponnesian homes and customary ways  (i.e., their ηθεα) to fight against an army of which they are “inexperi enced” [απειρο ι ].50 Just as the lions are “compelled” to attack, so the Spartans have undertaken a “compulsory” war: this means a defensive war, but their valor at the time of the battle turns defense into legitimate offense.51 For they have the “courage of bulls and lions,” according to the oracle at 7.220.4, and are led by their king Leonidas, “son of lion,” the Heraclid heir of the Pelopids of Lydia—traditionally and heraldically connected with lions—who will be immortalized with the statue of a lion.52 The gloss of identification of Leonidas at Thermopylae, which introduces his Heraclid genealogy, qualifies him as ο! θωµαζοµενος µαλιστα [the object of greatest wonder], from the point of view not exclusively of the narrator but of all, then and now (7.204). In a nearby and thematically related section, metanarrative thoma praises the courageous actions and words of the two Spartan ambassadors, Sperthias and Boulis (7.135.1). The Persian scout at Thermopylae is in wonder at the Spartans combing their hair and doing exercises before the battle his analogue. See also Knox’s 1952 analysis of the parable of the lion cub in Aesch. Agam. 717– 36. 48. See 5.56.1 (Hipparchus), 5.92β3 (Cypselus), 1.84.3 (Meles: unfortunately, the lion was born to him from a concubine, and Meles neglected to bring him all the way around, so the device ultimately did not work. Did the king fail adequately to provide for his succession?). Pericles is represented by a lion in a dream at 6.131.2; see “Wonder and Disbelief” later in the present chapter. 49. See, e.g., Il. 11.113– 21, 170– 78. 50. 9.46.2. Their inexperience is due to the Spartans’ absence at Marathon, as Pausanias says. I am grateful to Deborah Boedecker for suggesting the double meaning of  the term ηθεα in this passage, with allusion to the Spartans’ exceptionally giving up their insularity. This is not the first instance of this wordplay: see chap. 2, “Identification with the Other,” in the present book. 51. Cf. Il. 11.473– 86, where the lion’s offensive stance is used to characterize the defensive battle situation of Aias moving against the Trojans who crowd around wounded Odysseus. For Spartan aggressiveness at Thermopylae, see 7.211.3 and 7.223.2– 3. On  γκαι η in reference to the Spartans and “compulsory” war, see Demaratus at 7.104.3, ανα discussed in chap. 2, “The Evils of War,” in the present book. 52. See 7.225.2; Immerwahr 1966, 260– 61, n. 69; Georges 1994, 141– 42.

Thoma

247

(ε θωµαζε , 7.208.3). Herodotus’ wondering in the lion episode is related to and foreshadows the broader sense of wonder that he wants the listener to experience at the almost numinous epiphany of heroes in the first battle of this Persian war.53 In the passage about the absence of mules in Elis and other cases, perhaps including the lions episode at 7.125, wondering about cause is equivalent to being struck by the possibility that the divine is manifesting itself through an irregularity of nature.54 What is unambiguously a miracle—that is, a τερας, φα σµα, or σηµηιον—can be termed, from the point of view of the emotional reaction it elicits, a thoma.55 The attack of the Thracian lions, however, represents less of a violation of natural laws than do other animal omens in the Histories, especially the two reported in this same narrative of Xerxes’ march (7.57.1– 2). We should perhaps think of the lions phenomenon as what Homer would call a πελωρ: this occurs whenever animate beings, animals or heroes, reveal themselves pervaded by the presence of the divine, a notion that would well agree with Herodotus’ scientific and ethnographic observation of the continuity among the animal, the human, and the divine realms.56 But the question of why such and such happened and the answer to that question are in this particular case secondary. The story about the lions and the camels is first and foremost a narrative event, a metaphor. The narrator’s intensely subjective and disproportionate wonder at this minor incident in the world of the narrated first and foremost creates a mise en abˆıme for the larger picture of the Thermopylae narrative and what that narrative represents in the larger picture still. It invites the listener not so much to speculate on cause as, less specifically and concretely, to interpret meaning. Why Wonder Another minute instance illustrates the extent to which Herodotus’ communication relies on the symbolic code. This passage also foregrounds 53. See chap. 3, “Specific Glosses of Interpretation.”  54. Cf. 7.153; Herodotus mentions that the feat of Teline is a “wonder to me” [θωµα µοι] in light of the man’s effeminate nature (πε␸υκεναι); the implements of the goddesses in the narrative provide the clue. 55. See 6.117.2– 3 (blinding of Epizelus at Marathon), 8.135 (Mys receives oracle in Carian). The miracles at Delphi (8.37– 38) are objectively called τερατα, ␸α σµατα, and   θεια, but Herodotus also glosses the narrative with wonder words (θωµα, αξια θωµα σαι, 8.37.2). See Nenci 1957, 281– 89, for the various terms for miracles in Homer. 56. Cf., e.g., the “divine behavior” of cats (θεια πρηγµατα) at 2.66.3. See Nenci 1957, 189– 293, on the Homeric concept of πελωρ.

248

Telling Wonders

the behavior of animals: during Darius’ campaign against the Scythians, the mules and donkeys in the Persian army caused disarray among the Scythian horses (4.129). Once again, the mode of narration forces the recipient of the narrative to perform interpretive operations beyond the literal meaning of the text. The summary introduction to the episode contains an embedded program with an advertisement of narratability. I.

One thing that was helpful to the Persians and of hindrance to the Scythians when they were assaulting Darius’ camp—I am  going to tell a great wonder [θωµα µεγιστον ε ρεω]—was the voice of the asses and the sight of the mules. (4.129.1)

An explanatory gloss follows, with a back reference to the ethnographic section about Scythian winters I have already mentioned (4.28). The gloss repeats the same thing twice. II. For the land of Scythia produces neither ass nor mule, as I have explained before. In the whole country of Scythia there is neither ass nor mule because of the cold. (4.129.2) Next we find another summary introduction, resumptive of the first and in ring composition after the gloss. ! ι σαντες], they threw into confuIII. So, when the asses brayed [υβρ sion the Scythian horses. Now comes what we may call the narrative core of the episode, with factual details and in the mode of iteration.57 The negative causal clause  at the end (actually participles preceded by ατε) constitutes yet another gloss that essentially rephrases the idea, already expressed in sentence II, that there are no mules or asses in Scythia. IV. Often, in the middle of an attack against the Persians, when the horses would hear the voice of the asses, wheeling around they were thrown into confusion and were in wonder [ε ν θωµατι εσκον], pricking up their ears, for they had neither heard such a voice nor seen that sight. 57. Iteration occurs when the narrative represents once what happened many times. See Genette 1980, 116.

Thoma

249

The whole passage is capped by a conclusion that reduces the import of the interpretive statement “helpful to the Persians and of hindrance to the Scythians” in the initial introduction (I). V. Because of this, then, they gained a small advantage in the war. (4.129.3) This brief narrative is even more redundant than the one about lions attacking camels. The notion of Persian advantage is repeated twice (I, V), as is the central function that the sound of asses and the appearance of mules troubled the horses (III, IV—both times with ταρα σσω). The background information about the absence of asses and mules in Scythia appears three times (II twice, IV). As he is emphasizing the event, the narrator makes clear that it is small and inconsequential to the outcome (ε π ι τι ε ␸ εροντο). σµικρον This time it is most definitely not an omen; it violates no natural law. There is no wondering why, since the rational cause of the event is clear and reiterated in the glosses. Both the story about Cyrus exploiting the strategic advantage of the camels in his battle against the Lydians (1.80) and the case of the Thracian lions, presented as an exception (7.125), lead us to expect precisely what happened in this case, namely, that animals will be fearful of other animals unfamiliar to them. If the unexpected is an essential feature of thoma, the question is, rather, why should the horses’ being “in wonder” at unfamiliar animals constitute a “great wonder” to the narrator or to us.58 The anthropomor! ι σαντες, ε ταρα σσοντο) is an addi phism of the animals (ε ν θωµατι, υβρ tional sign that the narrative is metaphorical and designed somehow to illuminate the historical context at a higher level of meaning.59 The Per! sians, the only nation in the Histories who are called υβριστα ι by nature (by Croesus at 1.89.2), have symbolic associations with the mule, especially at the beginning of their history. Two mule prodigies refer to Persian actions, but more importantly, Cyrus, the founder of the Persian empire, is called a king-mule by the Delphic oracle.60 Mules (as well as 58. For the notion that the expected is “no wonder,” see 7.187.1. ! ι ζω (behave insolently) has the sense of “braying” at 4.129.2 only; the 59. The verb υβρ root is elsewhere applied to animals at 1.189.1 only (see n. 63 in the present chapter). The verb ταρα σσω is frequently used of armies throwing each other into disarray in this campaign (see especially 4.125.1– 5). 60. For the mule prodigies, see 3.151.2 with 3.153, 7.57.2. For Cyrus as mule, see 1.55.2; cf. 1.91.5– 6.

250

Telling Wonders

donkeys) connote poverty and an inferior social condition; they appear in Herodotean biographies of social upstarts who rise to power.61 Horses connote very different things. On the one hand, they are the sign of a “hard” culture, primitive or nomadic.62 On the other hand, horses can also signify power, wealth, and luxury. By the time of Xerxes’ expedition against Greece, the Persian is a “horse” in this second sense. This is clear from the omen of a mare giving birth to a hare. Hereby it was shown clearly that Xerxes would lead forth his host  with mighty pomp and splendor [αγαυρ οτατα κα ι µεγαλοπρε πεστατα] but that in order to return home, he would have to run for his life.63 In the episode of mules and donkeys, it is mildly ironical that the wealthy Persians gain their only advantage against the rude Scythians thanks to the most modest of their resources. At the symbolic level, however, the narrative suggests that mules and donkeys had what it would have taken for the Persians to be successful vis-a-vis ` the Scythians.64 In the narrative sequence that immediately follows, the bird, mouse, and frog the Scythians send to Darius are negative symbols signifying what the Persians are not (4.132.2). In the same way, here the Persians are not “mules,” and will shortly become “hares” (see 4.134.1; cf. 7.57.1, mentioned earlier). They have lost their original simplicity, such as the Lydian Sandanis described to Croesus (1.71.2– 4). No longer a “hard” people, as they were in the early time of Cyrus, they are a “soft” people attacking a “hard” people, and they will lose the war.65 The proliferation of animal incidents in this group of narratives and the metaphorical value of the animals in the Scythian message constitute one 61. See 1.59.4 (Pisistratus), 6.68– 69 (illegitimacy of Demaratus). Donkeys have equivalent connotations (e.g., at 1.194, 2.121δ, 5.68.1), and at 4.129, the fact that they can bray makes them more essential to the plot. 62. See 1.215.1– 2, 216.4. For the connection of horses with nomadism, see Hartog 1988, 18, on Hdt. 2.108. 63. 7.57.1. At 1.189.2, Cyrus’ white horse steps into the river “out of hybris” when Cyrus is attempting to cross (διαβαι νειν, the verb of violation of boundaries). On Persians and horses, see also 1.192.3, 7.40.2– 4, Smith 1992, 120– 22. Persian riding at 1.136.2 is, however, a trait of a “hard” culture. 64. In fact, the braying donkeys subsequently provide at least the means of Persian escape: see 4.134.3– 135.3. 65. This is an implicit rule of history: see chap. 2, “Implicit and Explicit Difference in the Ethnographies.”

Thoma

251

guarantee of a metaphorical understanding of 4.129, particularly by an audience more accustomed than we are to the functioning of the symbolic code. An additional guarantee is the disproportionate advertisement of  θωµα µεγιστον. Herodotus’ scientific attention to the animal world, which is part of his work as an ethnographer, joins the multifarious traditions that associate animals to humans. Poetic metaphor, such as we find in Homeric similes and animal metaphors in tragedy, verbally signifies specific human behaviors through animals; it is exploited in the religious tradition, which reads ominous meanings in real animal occurrences. Ionian storytelling includes “historical” anecdotes in which the cooperation of animals with humans shows either human natural ability to control lower forms of life or the granting of divine assent to human enterprises.66 Not least important, the animal fable features animals for the purpose of conveying a moralistic message about humans.67 Familiarity with the cultural conventions on which these different types of animal stories are based allows Herodotus’ listeners to integrate into an ethical structure individual events that do not in themselves need to signify anything beyond their literal meaning. Herodotus’ narration does its share to make these events fulfill the function of narrative omens or signs. The word the discourse uses to celebrate them, thoma, is a sign of signs. Vertical Analogy: Wondering Because The animal narratives I have considered show the overlap between the ethnographic notion of wonder and its historical counterpart and illustrate the rules of Herodotus’ pursuit of wonder in the account of past events. The thoma may raise a question concerning its cause or encourage reflection on its meaning. In either case, it often is a small item that the discourse retrieves from the side or magnifies on the way because of its potential to illuminate larger issues. This potential is fully realized by a spectacular image that springs up fully formed in the midst of the narrative of campaigns undertaken by one 66. See Charon of Lampsacus (FGrHist 262 F 1), analogous to Herodotus 1.80. Cf. 3.85– 86. 67. In Herodotus, an animal fable appears at 1.141. The fight between the lion cub and the puppy at 3.32 is an enacted ainos, decoded in the text and demonstrating the “natural” character of solidarity based on blood kinship. In the case of Cambyses, violations of custom entail violations of nature.

252

Telling Wonders

 of the Lydian kings. This is the θωµα µεγιστον once half-witnessed by Periander of Corinth (who was a friend of the enemy of the king), namely Arion of Methyma on a dolphin being carried off to Taenarum. Arion was a harp player, second to none of his contemporaries, and the first we know about who composed the dithyramb, gave it its name and taught the genre in Corinth. (1.23.1) The transition to the story epitomizes the freedom of Herodotus’ logos to pursue wonder on the lateral paths opened up by factual connec tions.68 The glosses of sources and cautionary λεγουσι (governing a narrative all in indirect discourse) contradict Periander’s verification in the narrative of the fact as true.69 Both off track and unbelievable, the story ranks high in the logos for the force with which the narrator applies the celebratory code to its protagonist and to its central event. Modern readers—and at least one ancient—have responded to these signals. The variety of available interpretations testifies to the depth and indeterminacy of what Herodotus regards “a great wonder.”70 The connection of narratability and symbolism, achieved in the previous examples through the cultural code of animals, emerges even more clearly in this case, owing to the narrative’s mythical cast. Attacked by sailors who want to steal his gold, Arion dons his professional robes and sings the nomos orthios (a technical term in the musical code, but meaning “correct law/custom” in ordinary language); he then jumps into the sea and is rescued by a dolphin, which carries him to Taenarum. In the poetic tradition, Apollo jumps on a Cretan ship in the guise of a dolphin and leads it to Taenarum and then on to Crisa, where he initiates the dumbfounded sailors to his priesthood by bidding them, among other things, to 68. See Pagel 1927, 4. Unlike most other anachronic narratives in Herodotus, the Arion episode cannot claim the function of an explanatory gloss in relation to its surroundings. 69. See 1.24.1, 6, 8. Packman (1991, especially 400) argues that Periander, whose initial incredulity is corrected by inquiry and verification ( ι! στορεεσθαι, 1.24.7), is analogous to Herodotus vis-a-vis ` his sources and that Arion, who is on the receiving end of Periander’s disbelief, is analogous to Herodotus vis-a-vis ` his audience, so that “on one level, the Arion story can be read as a plea for a specifically historical suspension of disbelief.” 70. See, e.g., Benardete 1969, 14– 16; Cobet 1971, 145– 51; Flory 1978; Munson 1986; Packman 1991, 399– 401. Plato exploits the metaphorical character of the story in the Republic (454d), by making Socrates say that when his positions come under the wave of an attack, he will “hope for a dolphin to take us on its back or some other impossible    means of rescue.” Here the phrase τινα αλλην απορον σωτερι αν echoes απορ ι ην in Hdt. 1.24.4.

Thoma

253

sing the paean (Hom. H. Ap. 3.388– 544). Dionysus turns into dolphins the Tyrrhenian pirates who have kidnapped him, but he saves and rewards with prosperity the righteous helmsman (Hom. H. Dion. 7). The legend reported by Herodotus uses similar themes to heroize Arion into an almost sacral figure, whose immunity from danger represents the triumph of the fundamental values he embodies: intellectual worth, adherence to nomos, loyalty to the gods, deliberate pursuit of one’s assigned task in the face of danger, and refusal to be subjugated.71 The gloss of testimony appended to the conclusion records the existence at Cape Taenarum of a tangible representation of the rescue, the bronze statue of a man riding a dolphin (1.24.5). The offering is “not  large” [ου µεγα] and is antithetical, therefore, to the impressive monuments of the East. It does less to corroborate the veracity of the legend than to confirm its meaning:72 Arion is also “small” in the face of a stronger opponent, who is ethically (though not ethnically) barbaric. The wonder of his survival joins the mystery of a prodigy to the significance of a natural event. In the context of Herodotus’ scientific and historical work, the agency of the dolphin points again to the participation of nature in a divine plan that is ethically rational according to the standards of men. The ainos of Arion is inserted in the first detailed report in the Histories of the aggression against a small Greek state by a large Eastern power. It independently confirms the evidence provided by that narrative for the causality of success and failure.73 Its message, however, also reflects preoccupations that will come into better focus later on with the rescue, almost miraculous but natural and rational, of the Greeks from the Persian danger.74 The symbol of Arion himself will in fact return at that point, this time on the main path of the logos, just before the Greeks’ first confrontation with the Persian fleet, and in a more realistic incarnation. Gone are the dolphin, the “best musician of his time,” the sacred robes and accoutrements of his profession, the nomos orthios sung among the rower’s benches, and the prodigious reappearance on dry land 71. See Flory 1978. As Benardete remarks (1969, 15), “nowhere else in Herodotus does  νοµος mean ‘tune.’” 72. See Bowra 1963 on the actual religious background of this statue. 73. See Cobet 1971, 149. Alyattes’ failure to conquer the Milesians, due to a divinely induced sickness (1.19.1– 21.1), is just as unpredictable in human terms as is the failure of the Corinthian sailors. 74. See Munson 1986, 99.

254

Telling Wonders

of the rescued musician, utterly intact and “just as he was when he jumped” (1.24.7). The new Arion is Scyllias of Scione, “the best diver of his time” (8.8.1). He deserts from the Persian camp and does what he is accustomed to do and does best: he dives into the sea. Having reached the Greeks at Artemisium, he gives them intelligence of the enemy side. A predominantly metanarrative passage in the mode of historie replaces the mythical narrative of Arion (8.8.2– 3). In what way [Scyllias] at that point arrived among the Greeks, I cannot tell precisely, but I am in wonder if the things that are said   ε στι αληθ   are true [θωµαζω δ ε ε ι τα λεγοµενα εα]. For it is said that from Aphetae, having dived into the sea, he did not reemerge until he arrived to Artemisium, having crossed a distance of about eighty stades by sea. Indeed, other things are said about this man   ικελα . . . that are similar to lies, but some are true [αλλα ψευδεσι    τα δ ε µετεξετερα αληθ εα]. On this matter, however, let me express my opinion that he arrived at Artemisium by boat. Scyllias is almost a contemporary and an Arion translated into history. The narrator intertwines the code of refutation with that of celebration and still expresses the narratability of the unverified event in terms of wonder. He does not so much “wonder whether” as he is “in wonder if”—the formulation he applies to the allegedly floating island of Chemmis.75 The phrase “other things similar to lies, . . . but some true” assigns Scyllias’ “swim” to Artemisium to an uncertain zone between these unbelievable but true facts and the antithetical “lies similar to truth” of epic poetry, meaningful fictions with the power to communicate what is essentially true.76 True or untrue, prodigious or merely exceptional, the rides to safety and freedom of Arion and Scyllias easily become the subject of ainoi that replay the rescue in the larger plot of the Histories as a concluded whole. These passages are not symbolically univocal, for they can accommodate references to more recent extratextual in75. See 2.156.2, discussed under “Herodotus and the Conventional Code of Ethnographic Wonders” earlier in this chapter. 76. Cf. the Muses in Hes. Theog. 27: “We know how to say many lies similar to truth,   but we also know how to sing true things when we want to” [ιδµεν ψευδεα πολλα λεγειν  ε θελωµεν   ηθεα   ε τυµοι σιν οµοι α, / ιδµεν δευτε αλ γηρυσασθαι] (cf. Od. 19.203, Theognis 713). See Bowie 1993, 17– 23. Herodotus’ quotation is a nod to Scyllias’ heroic status among his contemporaries, on which see Masaracchia 1977, 161, How and Wells 1928, 2:238.

Thoma

255

stances of abuse and self-interest among the Greeks. Thus, Arion’s aggressors are not barbarian pirates, as are those of Dionysus in the Homeric hymn; they are Corinthians whom Arion“trusted more than anyone else” to convey him safely home (1.24.2). The gloss of identification of Scyllias informs us that he made use of his diving ability after the shipwreck off Mount Pelion to rescue valuable property for the Persians and gain a great deal for himself as well (8.8.1). This is a nonidealized narrative element whose connotations clash with the symbolism of Scyllias’ feat at Artemisium.77 In the case of Arion, the symbolic indeterminacy is enhanced by the implicit self-referential aspects of the protagonist: he is, like Herodotus, a skilled performer who must eventually confront hostile audiences.78 One of these listens to him for the sake of pleasure but plans to kill him after the song (1.24.5). The other meets his narrative with the same disbelief the narrator of the Histories experiences from his listeners.79 Herodotus’ patterns from small to large or from far to near and across narrative levels create an open-ended system of competing associations that we modern readers have only begun to reproduce. We can only do so by taking interpretive risks. This brings us to another object of wonder, the mysterious Artemisia of Halicarnassus. More daringly than in his narrative on Arion, here Herodotus combines a self-referential element with a special exploitation of the vertical analogy between individuals and states.80 Let us begin with the second aspect first. As an ally of Xerxes in his expedition against Greece, Artemisia was for fifth-century Greeks the historical reincarnation of the mythical Amazon, the enemy of the civilized male world of the polis, the invader of Attica, where she found defeat and obliteration at the hand of the Greeks, specifically the Athenians.81 As we have already 77. Cf. the story of Ameinocles of Magnesia (7.190), criticized by Plutarch (De Malign. Herod. 30, 39 ⫽ Mor. 864C, 871C). 78. See Bernadete 1969, 14– 16; Packman 1991; Thomson 1996, 151 n. 20, 167. Benardete quotes Plato Rep. 454d (see n. 70 in the present chapter), where Socrates also applies the Arion persona to himself in the face of opposition from his listeners. I call a character “self-referential” when it emerges as a double of the narrator Herodotus. But I have been considering a term or a sentence self-referential when it speaks about the narrative either of Herodotus or of his source (see especially my definitions in chap. 1). 79. See Packman 1991 (see n. 70 in the present chapter). Cf. Payen 1997, 58. For reactions attributed to the audience of the Histories, see chap. 1, “Self-Referential Glosses,” in the present book. 80. See especially chap. 2, “Analogy as an Interpretive Tool.” 81. On the connection between Artemisia and the Amazons, see Aristoph. Lys. 671ff.

256

Telling Wonders

seen, in book 4 Herodotus reverses this politically significant tradition by inserting into his Scythian narrative the paradox of marriageable, reasonable, and peaceful Amazons.82 For the purposes of his narrative of the battle of Salamis, he creates a different Amazon paradox, one that turns same into other rather than, as in the Sauromatian logos, other into same. Unlike Herodotus’ Amazons in book 4, Artemisia is aggressive, eager for war, and equipped with excellent ships.83 Unlike the Amazons of tradition, she is Greek, not barbarian; cultured, not wild; and renowned for strategic ability. She is also a winner, and she is a winner over her own side. The Amazon of tradition, defeated by Athens, is here reborn as a symbol of the Athens that was born with the victory of Salamis.84 The triumph of Artemisia is the opposite of that of Arion and is enough to make the narrator throw up his arms in a very different sort of amazement. The narrative has its comic side: reduced in a tight spot during the battle, Artemisia undertakes to save herself in a flash by ramming an allied ship; she gets lucky and wins, and in the process, she gets doubly lucky and increases her power and prestige (8.87– 88). As an open-and-shut case, at least, Artemisia provides evidence that contradicts so much other evidence Herodotus has been accumulating for an ethically rational order of things. Can ill-gained success possibly be permanent? Artemisia is a symbol and an enigma, one of several that punctuate the beginning of the end, and the end of the end, of Herodotus’ investigation.85 We are now ready to complicate the picture further and examine the peculiar way in which the narrator signals this thoma and his own involvement with it. The following is the programmatic introduction to a long gloss of identification for Artemisia. Of the other taxiarchs, I make no mention/memorial, on the  grounds that I am not compelled by necessity [ου παραµεµνη    µαι . . . ως ουκ αναγκαζοµενος], but [I do] of Artemisia, whom   especially I regard with wonder [της µαλιστα θωµα ποιευµαι], a woman marching against Greece, who, since the death of her hus82. See chap. 2. “The Other Is Same.” 83. See 7.99.3. The inexperience in seafaring that the Sauromatian narrative attributes to the Amazons (4.110.1) becomes more intelligible at the symbolic level in light of their opposite and analogue Artemisia: see chap. 2, n. 245. 84. See Munson 1988 for a detailed argument. On personification of poleis in Greek art, see Holscher ¨ 1995, 174. 85. See, e.g., 7.133.2 and chap. 3, “Divine Retribution” in the present book.

Thoma

257

band herself holding the tyranny [τυραννι δην] and having a young son, participated in the expedition out of daring and manly cour ς age, there being no necessity/compulsion for her to do so [ουδεµιη    ο ι ε ουσης αναγκα ι ης]. (7.99.1) The last clause brings into focus Artemisia’s aggressiveness. As a fe  male τυραννος (and, we might add, as a symbol of the city-τυραννος), she does what despotic powers normally do, waging an “unnecessary”— that is, offensive—war.86 At a more immediate level, however, the expres  ς ο ι ε ουσης   sion ουδεµιη αναγκα ι ης bears positive connotations: unlike other allies compelled by Xerxes, Artemisia is, for unspecified reasons (and again like Athens), free.87 This special position is part of her paradox as a wonder and ostensibly the first reason for her inclusion in the logos. In a previous negative program, Herodotus has stated that since the local infantry commanders in the Persian force were not free but   slaves, he was “not constrained by necessity” [ου γαρ αναγκα ι η  ε ξεργοµαι] to mention them (7.96.1– 2). The combination of this passage with 7.99.1, which closely follows, yields the sequence “I am not compelled to mention other commanders who were slaves [i.e., who were under compulsion], but [I am compelled to/will mention] Artemisia, whom I consider a thoma and who was not under compulsion.” The rhetorical figure, underlined by the repetition of metanarrative selfreferential anank- words, creates an antithesis between those other foreign subjects of Xerxes and the free narrator (the prominent ε γω of 7.96.1), who is not compelled to say in his logos what he does not consider worth telling. By the same token, Artemisia of Halicarnassus, at least to the extent that she is free and “not under compulsion,” becomes here (independently from whatever else she is already beginning to symbolize) implicated symbolically in the world of the narration also as an analogue of “Herodotus of Halicarnassus,” the histor.88 86. See, e.g., Artabanus at 7.10δ1 and chap. 3, “The Evils of War,” in the present book.  87. In the historical narrative, words of the αναγκfamily most frequently refer to compulsion applied by a monarchical ruler, especially in the narrative of Xerxes’ expedition (7.103.4; 7.108.1; 7.110; 7.132.2; 7.136.1; 7.139.3; 7.172.1; 8.22.2; 8.140α2; 9.17.1). See Munson forthcoming. The unique position of Artemisia as a free agent is evidenced in the subsequent narrative of her advice to Xerxes (8.67– 69). See Munson 1988, 95– 98. 88. Georges (1994, 301 n. 93) reminds us that she may have been the great-aunt of Herodotus the real author.

258

Telling Wonders

From what sort of compulsion does Herodotus declare himself free? Implicitly it must include something similar or vertically analogous to the monarchical compulsion that the local commanders in the Persian army experience and Artemisia does not. That means all external pressure exercised by persons, for political or other reasons. Herodotus’ immunity from that type of pressure consists here, as many other times, in his being free to magnify the minute and meaningful narrative element that is a thoma to him. This will entail, as it turns out, robbing the real protagonist at the battle of Salamis of a starring role, in favor of a minor participant in the enemy camp, with five ships to her name. Episodes of Greek— and especially Athenian—skill, valor, patriotism, and love of freedom are conspicuously absent in the battle report, in striking contrast with Aeschylus’ version of the same battle and with Herodotus’ own narratives of Thermopylae or even Marathon. First, Herodotus’ narrative of Salamis attributes the Greek victory in great part to the hopeless strategic disarray of the Persian force. Second, but more importantly, Herodotus narrates this battle to make it mean what it meant as seen from the perspective of later stasis developments among the Greeks.89 The central Artemisia section (8.87– 88) contributes to the first purpose in its literal import; by constructing a metaphor for later Athenian unscrupulous behavior, it contributes to the second. Herodotus’ narrative of Salamis, in other words, adds two implicit qualifications to his earlier explicit praise of the Athenians as the saviors of Greece, largely owing to that victory (7.139). This is a radical choice, almost as surprising as Artemisia’s turnabout maneuver on the battlefield. Although daring and free from external compulsion, however, the narrator of the Histories is, unlike Artemisia, subject to an ananke of a higher order. In voicing his praise of Athens at 7.139.1, he counters   political pressure by declaring himself compelled (αναγκα ι η ε ξεργοµαι) to express an unpopular opinion. Here the compulsion is the moral duty  to tell the truth.90 In two additional cases of self-referential αναγκin the passive voice, the compulsion to tell is exercised by the logos and has to do with the narrator’s self-imposed task to provide a didactically effective 89. Cf. chap. 3, “Disputes, Arbitration, and the Subjectivity of Opinions.” The suggestion that the Persian failure was largely due to Persian strategic errors would be part of the argument of those who wished to minimize Athenian merit. See the Corinthians in Thuc. 1.69.5.   ταληθ   90. See 7.139.5: αληθ ες, εος. See also chap. 3, “Specific Glosses of Interpretation,” in the present book.

Thoma

259

account.91 Herodotus “is” Artemisia, but he also “is” Arion. The requirement of his logos, what we may call the nomos of the logos, is his obligation as well as his choice.92 This includes, with explanation, the pursuit of wonders, which are signs for things that need to be explained. Warned by the marker thoma, a sign of signs, his listeners will take note, stop and think, and explain on their own. Wonder and Disbelief The interrogative mode of the root θωµ-, as we have seen, usually does not concern the existence of the object of wonder (is it really true?) but rather has to do with its meaning. Even when the expression of wonder combines with glosses of noncorroboration, Herodotus implicitly encourages the audience to make sense of the wondrous phenomenon as if it were true. A few times, however, θωµ- is more closely connected with disbelief. If some Greeks consider the Persian support of democracies in Ionia a thoma megiston, that means they will find it hard to believe (6.43.3). In the two passages where Ionian theories are a cause of wonder because they are implausible and absurd, θωµ- (exceptionally self-referential in the sense that it identifies the reported logos) functions as a term of the code of refutation, not of the code of celebration.93 The last case I will consider perversely combines the reading directions “thoma, absurd logos, and untrue fact: disregard” and “thoma, untrue fact, interesting logos: pay attention,” perhaps even adding (according to some interpretations) “thoma, unbelievable logos, but perhaps true fact.” The last possibility is hard to accept because it entails understanding the text as saying the exact opposite of what it clearly says. That several readers have advanced it testifies to the greater than usual discordance of what the text communicates at different levels of metanarrative. If there is  ο του λογου   91. See 2.3.2 (υπ ε ξαναγκαζοµενος) and 2.65.2. See also chap. 2, “The Texture of Nomos,” in the present book. For other expressions of obligation, see chap. 1, n. 38 and corresponding text. 92. In Herodotus, moral obligation and the compulsion of nomos often conflict with monarchic obligation and are strictly related to voluntary choice. See, e.g., the case of  and declares himself compelled to tell the Prexaspes, who both decides voluntarily (ε κων) truth (3.75.1– 2). 93. See 2.21, 4.42.1 In the first passage, θωµασιος almost functions as an advertisement of nonnarratability: the theory is wondrous and therefore not worth reporting. See discussion under “Herodotus and the Conventional Code of Ethnographic Wonders” earlier in this chapter.

260

Telling Wonders

a performance in Herodotus that we would most like to observe live, complete with body language and tone of voice, this is certainly it. The passage in question ranks among the quarrels of the Histories, where the histor/arbitrator sets forth an accusation and evaluates the merits of a case. Here the accusation is presented not as a current logos but as a charge belonging to the world of the narrated, made in Athens at the time of Marathon or shortly thereafter (α ι τ ι η . . . εσχε ε ν Αθηναι οισι, 6.115). This aitie claimed that after the fighting was over, the Alcmaeonids arranged for a shield to be flashed from Athens to signal to the Persians at sea that they should round Sounion and sail to the city, which in fact they did. Nothing came of the alleged incident, however, because the victorious Athenian army raced back to Athens by land and arrived before the Persians. So the enemy gave up their designs and, from Phalerum, sailed back to Asia (6.115– 16). After the narrative of other events in the aftermath of the battle (6.117– 20), Herodotus goes back to the Alcmaeonids issue in a long gloss of refutation introduced by a rejection of the charge. Here, wonder at the fact reported leads to disbelief in the report and to its rejection based on unlikelihood:  It is a wonder to me and I do not accept the rumor [θωµα δε µοι κα ι  λογον]   ε νδεκοµαι  ουκ τον that the Alcmaeonids would have ever lifted up a shield to the Persians as a signal according to an agreement, wishing the Athenians to be ruled by the barbarians and Hippias. . . . (6.121.1) The participial clause “wishing the Athenians . . .,” reported as a logical implication of the charge, has the effect of underlining its absurdity. The refutation is attached directly here in a crescendo of rhetorical animation: “they who [ο$ιτινες] were tyrant-haters, just as much or more than Callias, the son of Phaenippus and the father of Hipponicus.” An explanatory gloss follows, demonstrating the extent of Callias’ hostility toward the Pisistratids (6.121.2); this concludes with the restatement of the comparison that had caused the mention of Callias in the first place: “So also  the Alcmaeonids were tyrant-haters [µισοτυραννοι] no less than this man” (6.123.1). So far the refutation seems to express straightforward outrage, even though the comparison that brings the family of Callias into the discourse employs a cultural code of Athenian politics that is not entirely transpar-

Thoma

261

ent to us.94 The second movement of the defense, introduced with a rejection closely analogous to the first (“It is therefore a wonder to me and I do not believe the slander,” 6.123.1), takes off again with the animated “they who” [ο$ιτινες] of the first movement. This time it adduces as proof of innocence the Alcmaeonids’ exile during the whole period of tyranny and their success in bringing the regime to an end: “it & %  Αθηνας   was they who freed Athens [τας ουτοι ησαν ο ι ε λευθηρωσα ντες] more than Harmodius and Aristogeiton, in my judgment [ως ε γω κρι νω].” Herodotus then anticipates the objection that the family might have turned to treason against their fatherland because they had some grudge against the Athenian people, but he counters this point by arguing that no one enjoyed more renown and honor among the Athenians than the Alcmaeonids (6.124.1). So not even the report that they would have   ε λογος made signals with the shield for that reason makes any sense (ουδ   α ιρεει). “A shield was raised and that cannot be denied; it happened” Herodotus concludes, “but as to who did it, I cannot say anything further” (6.124.2). The rhetoric about the Alcmaeonids as liberators of Athens recalls the encomium of the Athenians, “saviors of Greece” (7.139.5); the conclusion of the refutation has the same definitive tone of the indictment of Ephialtes (7.214.3), except that here Herodotus generalizes the indictment and acquits the individual culprit (far more prominent than Ephialtes!) instead of the other way around.95 But the mention of the Alcmaeonids’ renown generates a long explanatory gloss on the history of their wealth and status, where eager argumentation gives way to amiable mimetic narrative and, at the same time, political history gives way to folklore, or what Thucydides would call “the mythical/romantic ele ment” [το µυθωδες] (Thuc. 1.22.4). This gloss consists of three connected narratives, the first of which tells the story of how the family’s eponymous ancestor Alcmaeon enjoyed 94. Cf. the ambiguity of 7.151 (see chap. 3, “Disputes, Arbitration, and the Subjectivity of Opinions,” text and n. 239, in the present book). Here the point may simply be that, like the Alcmaeonids, they were loyal democrats and had long enjoyed good relations with Persia: the identification of Callias as “father of Hipponicus” recalls the son of Hipponicus named Callias, the negotiator of the peace. Callias II may have earned Herodotus’ approval. Probably a religious conservative (diadouch at Eleusis), he was related to Aristides (see Plut. Arist. 24.4– 8; Davies 1971, 257), the only Athenian politician whose portrayal in Herodotus does not contain elements connoting “tyranny” (see 8.79.1). 95. He acquits the Alcmaeonids but lays charge on others, complains Plutarch (De Malign. Herod. 27 ⫽ Mor. 862E).

262

Telling Wonders

friendly relations with Croesus and visited him at his court in Sardis, where he was given a large amount of gold (6.125). The second is about Cleisthenes, the tyrant of Sicyon, choosing a husband for his daughter, Agariste, among the best and brightest young men from all over Greece. He ended up giving her to Megacles, the son of the Alcmaeon of the previous story, who thereby greatly increased the family’s prestige (6.126– 30). The third section briefly reports the birth from Megacles and Agariste of the Athenian democratic reformer Cleisthenes and another son, named Hippocrates. The offspring of this Hippocrates included a younger Agariste, who married Xanthippus. When she was pregnant, she dreamed she had given birth to a lion; after a few days, she became the mother of Pericles (6.131). Scholars have observed that these narratives undermine the defense to which they are appended by highlighting the Alcmaeonids’ long-standing connections with Eastern monarchs and Greek tyrants.96 It is important, however, to establish to what extent they do so and in what way. The possibility of the Alcmaeonids’ closeness to the Persians is not even taken into consideration.97 Herodotus rather refutes the charge on the basis of the Alcmaeonids’ hostility toward the Pisistratids (6.121– 23), a point that the subsequent “renown” narratives never undermine. We find no reference, for example, to the fact that Megacles, who here marries Agariste and fathers the sons Cleisthenes and Hippocrates, gave a daughter in marriage to Pisistratus, as we learn elsewhere in the Histories.98 We have in these sections not an implicit retraction of the defense but a cheerful reflection on the charge, marked by a striking change in the mode, tone, and level of the discourse. In the first two narratives—on Alcmaeon and on Agariste’s wedding, 96. See especially Strasburger 1955, 15– 18; also Thomas 1989, 264– 72. Thomas attributes the jarring effect to the combination of an apologetic family tradition (adapted to the patriotic polis tradition) in the first part of the defense and popular traditions in the second part. 97. We should perhaps infer a pro-Persian policy on the part of Cleisthenes and the Alcmaeonids on the basis of the embassy Herodotus reports at 5.73.3. But Herodotus does not mention Cleisthenes or the Alcmaeonids in that passage, nor does he ever suggest that the aitie at the time of Marathon is somehow connected to the aitie previously incurred by the ambassadors. See Fornara and Samons 1991, 19– 20. 98. See 1.61.1. This silence is noted by Plutarch (De Malign. Herod. 27 ⫽ Mor. 863A– B). Herodotus’ claim that the Alcmaeonids were in exile “for the entire time of the tyranny” (6.123.1) is not inconsistent with his statement at 1.64.3 and with the apparent unawareness of other ancient sources concerning Cleisthenes’ archonship under Hippias in 525 b.c. (ML 6). See Davies 1971, 372; Fornara and Samons 1991, 17.

Thoma

263

respectively—concrete novelistic details proliferate, some very funny. Let loose in Croesus’ treasure-house, Alcmaeon is the ethical opposite of Solon on a similar occasion.99 But he is also his opposite from the point of view of his dramatic role, verbally incapacitated by the gold he has packed into his mouth, his logos replaced by body language. “Stuffed up” with gold in his robes, hair, and boots, he staggers out of the room “similar to anything but a human being” (6.125.4). We are similarly made to see the suitors (µνηστηρες) of Agariste “stuffed up” with lineage and pride.100 They stick to their best behavior in the gracious and tense atmosphere of Cleisthenes’ court while the tyrant assesses their performance in the games and at dinner. During the last and most lavish of an excruciating yearlong series of banquets, in the midst of civilized competitions in public speaking and song, and in front of all the Sicyonians who have been invited to the party, the favored contender, Hippoclides, throws it all away when he starts dancing bottom-up on the table, moving his legs about in the air as if they were arms (6.129.4). “Son of Tisander,” Cleisthenes bursts out, “you have just danced away your marriage.” Hippoclides replies, “Hippoclides does not care,” with a phrase that has become proverbial, as the narrator’s gloss explains (6.130.1). This Hippoclides, who gets tired of good manners, is not just a crude  finally showing his true colors. He is one who, by Homeric µνηστηρ breaking the rules of the polite competition, rebels against the political constraints of his tyrannical audience. His performance is just as whimsical and unconventional as the turnabout of Herodotus from the earnest polis patriotism of his Alcmaeonid defense to the irreverent and factually dubious novelistic narratives that follow. As Plutarch observes, in his criticism of a different passage, It seems to me that, like Hippoclides standing on his head on the table and waving his legs like arms, Herodotus would dance away the truth and say, “Herodotus doesn’t care.”101 An ancient reader here exploits the proverbial character of Hippoclides’ utterance reported by Herodotus to respond to Herodotus’ own 99. See 1.30– 33; Strasburger 1955, 18. 100. Chamberlain (1997, 66– 67) notices the verbal correspondence between παντα   ε ξωγκωτο at 6.125.5 and ε ξωγκωµενοι at 6.126.3. 101. Plut. De Malign. 33 ⫽ Mor. 867B, on Hdt. 7.233.3. In reference to the Alcmaeonid defense, Plutarch uses another ainos, comparing Herodotus to a predator who catches a crab and then promises to let it go (De Malign. Herod. 27 ⫽ Mor. 862F– 863A).

264

Telling Wonders

subversive narrative methods. The self-referential aspect of Hippoclides seems especially apt when applied to the very context where Hippoclides does his dance, that of Herodotus’ extravaganza on the topic of the Alcmaeonids.102 But the transition from argument to lighthearted narratives that work through connotation and symbolism, and from explicit praise to implicit ambivalence, is not unique to this passage. We have already noticed the shift in tone and substance from the encomium of Athens at 7.139 to the report of the antics of Artemisia and others in the narrative of Salamis. Like Artemisia, the narrator of the Histories is a free agent. Like Hippoclides, who frees himself from the constraints of the tyrant Cleisthenes, he displays his freedom as a performer vis-a-vis ` the political establishment of the leading city (the tyrant city) of his time. Herodotus dances away not “the truth,” as Plutarch says, but the question of what the truth is, in the old controversy of the shield at Marathon. The specific accusation against the Alcmaeonids stands rejected, and he no  , as Plutarch paraphrases). That allegalonger cares (ου ␸ροντ ι ς Ηροδοτω tion of the past is now a thoma good for thinking in metaphorical terms about the present. Just as Artemisia at Salamis deconstructs the antithesis between male and female, Greek and Barbarian, friend and enemy, in relation to the polis turannos, so the old rumor about Alcmaeonid medism at Marathon leads Herodotus to a survey of family history that confounds the distinctions between East and West, democracy and tyranny, citizen and foreigner, in reference to its leaders. Proceeding from the eponymous ancestor Alcmaeon, to his son, Megacles, (to whom a foreign tyrant gives  “according to the nomoi of the his daughter in marriage [verb ε γγυω] Athenians” [6.130.2]), to the democratic reformer who bears the name of his tyrant grandfather, the Alcmaeonids’ ambivalent line leads directly to the ambivalent lion Pericles, the one and only Alcmaeonid who was still relevant in Herodotus’ time.103 As a case that engages Herodotus in his role of arbitrator of differences, the discussion of the aitie against the Alcmaeonids conforms in its own way to the rule of explicit acquittal 102. Chamberlain (1997, 34– 81, especially 52– 65) is the first, as far as I know, who saw Hippoclides as a double of Herodotus. Cf. Dewald 1987, 151. 103. Dewald (1998, 691) remarks that Cleisthenes’ remarkable phrase at 6.130.2 “reminds H’s contemporary readers of Pericles’ citizenship law of c. 450 bce, demanding that an Athenian citizen have two Athenian parents. Here Pericles’ own great-grandmother is the foreigner involved.” Pericles claimed an exception from his own law for the sake of his son from Aspasia of Miletus (Plut. Per. 37.5). For the lion imagery, see “Wondering Why” earlier in the present chapter. On Pericles as a lion, see chap. 2, “The Monarchical Model in Athens.”

Thoma

265

and implicit warning or blame.104 This combination is here achieved through a mixture of historie with the mode of the ainos.105 From the point of view of Herodotus’ notion of wonder, however, the Alcmaeonid passage features the narrator more clearly than elsewhere demonstrating through his own discourse what he wants his listeners to do with a thoma—freely associate and reflect, to find a broader context or a different plane of experience where the absurd becomes intelligible and the abnormal meaningful.

104. See chap. 2, “Disputes, Arbitration, and the Subjectivity of Opinions.” 105. See Nagy 1990, 310– 13.

Conclusion

The discussion in this book has sought access to the Histories first and foremost through the discourse: it has attempted to understand what the text says by examining how it says it. I have analyzed three functions of referential metanarrative (comparison, interpretation, and, in a more limited way, celebration) to show how they contribute to transforming historical narrative and ethnographic description—the surface declarative modality of the text—into a lesson for the here and now of narration. Largely based on Herodotus’ experience of foreign societies through time and space, this lesson includes a warning to the Greeks to take stock of essential similarities beyond contingent differences and to recognize the likely human responses, common values, and constraints that emerge from the cultural norms and historical vicissitudes of “all men.” Herodotus’ warning is moralistic, because it is implicitly based on the idea that human beings have control over their behavior and consequently, to some extent, over their fortunes. The text also leaves room, however, for the adviser’s ultimate aporia with regard to the ways in which the recipients of the narrative can usefully apply the teachings he provides. It conveys no confidence that they will in fact be able, or still in time, to benefit from them. Thus, the imperative and interrogative features of the Histories radically modify the movement toward a happy resolution in which the narrative, when considered only from the point of view of its declarative aspects, may appear to be emplotted. Herodotus’ performance wants to be considered from the point of view of the audiences for which it was intended and in light of the circumstances that obtained at the time. This is a risky undertaking for us modern readers: precisely what times, what places, what circumstances, and what audiences are we talking about? Herodotus no doubt composed and performed the Histories piecemeal. They must represent a considerable portion of his life’s work. Some parts perhaps originated as independent expositions and have been only imperfectly adapted to the final form 266

Conclusion

267

of the whole. The whole itself appears at once both finished and provisional. Nevertheless, we can safely bracket off the third quarter of the fifth century b.c. as an extratextual context of Herodotus’ performances. We are relatively well informed about the most memorable events of that time and about some of the contemporary reactions and ideological predispositions of the public. Against this background, I have tried to show that the ideological underpinnings of the ethnographic descriptions represent a well-suited complement to the political thought that emerges from the history, including from those historical narratives that reflect domestic concerns and belong to the most recent identifiable layer. A more precise reconstruction of the circumstances of performance would require relating the histor Herodotus to the real author Herodotus at a particular time. If we do so on the basis of the evidence that is available to us, however, we may end up exercising our imagination to a greater degree than is appropriate to a scholarly task. The external tradition about the life of Herodotus is fragmentary, composite, sometimes derived from the interpretation of hints in the Histories, and for the most part late. It portrays an itinerant lecturer, an Ionized Dorian, possibly half-Carian, born from a distinguished family but perhaps not an aristocrat, politically active but not a military man.1 A figure on the margins in the most literal sense, he started out on the eastern border of the Greek world, a Persian subject; he moved away (somewhat like those Ionians of whom the Histories approve) and eventually ended up in a mixed Greek democratic community in the far west. The Thurii Herodotus joined was not an Athenian imperialistic venture but a unique Athenian initiative with Panhellenic aims; it was inspired, at least for some of those involved, by the atmosphere of conciliation that briefly followed the stipulation of the Thirty Years’ Peace.2 What we   1. On Herodotus’ life, see especially Suda, s.vv. Ηροδοτος and Πανυασσις, and the sources collected and discussed by Jacoby (1913, 205– 47) and Legrand (1932, 5– 15). See also Brown 1988, especially 4: “Unlike the three major historians with whom he is usually compared—Thucydides, Xenophon and Polybius—Herodotus was not a military man, he remained a civilian all his life. And that should not be forgotten if we are to understand the way he thought” (emphasis mine). I also agree with Brown (1988, 8) that the statement in the  ε πι␸ανων)  does not necessarily Suda that Herodotus came from a distinguished family (των imply a noble birth. The personal narrative at 1.143 seems ironically to imply that, unlike his predecessor Hecataeus, Herodotus was not in a position to show off his lineage. 2. The main primary sources for the foundation of Thurii are Diod. 12.9– 11 and Strabo 6.1.13– 15. The original aims and meaning of the enterprise are controversial. See Cloch´e 1945, 95– 103; Ehrenberg 1948; Wade-Gery 1958; Kagan 1969, 154– 78, 382– 84. Kagan (1969, especially 168) attributes the initiative to the desire of Pericles to deflate the

268

Telling Wonders

know about the foundation of Thurii reveals practical and political motives;3 it also suggests, however, that a utopian impulse may have played a role in this enterprise. Plato, after all, would later find inspiration for his ideal state precisely in this part of the world. The architect of Thurii was Hippodamus of Miletus, who reorganized Piraeus and codified the orthogonal subdivision of urban space on the model of the Greek cities of Asia. This eccentric polymath was also versed in political theory, concerned with the ideal size and proportions of the democratic city-state and the best form of government. His proposed constitution envisioned equal political rights for a population of ten thousand citizens divided into three classes: artisans, farmers, and an armed force for the purposes of defense. We can perhaps relate these features to Herodotus’ interest in such subdivisions (e.g., at 2.164– 68), his essentially democratic preferences, his concern vis-a-vis ` the excessive growth of cities, and his approval of a state’s readiness against external attacks. Hippodamus also apparently maintained that in a well-run state, judges should be allowed to render qualified verdicts. This measure, objects Aristotle, would turn a judge into an arbitrator and obliterate the clarity of the judicial process. But Herodotus’ own subtle attempts to settle discords may indicate that he shared Hippodamus’ concerns about the possible unfairness of absolute condemnations or acquittals. Similarly, Hippodamus’ idea that there should be a law for awarding honors to any citizen who should make an invention of benefit to the state may throw light on Herodotus’ exploration of specious or wise inventions by reformers and nations around the world.4 Whatever Hippodamus’ influence might have been on the constitution of Thurii, Protagoras of Abdera undertook the task of writing the new city’s laws. This is the thinker, as we have seen, with whom Herodotus’ text carries on an implicit dialogue on the subjects of relativism and the knowability of the gods in relation to the world’s different religious rhetoric of his domestic opponents by demonstrating “his moderation, his lack of imperial ambition, and his Panhellenic sentiments.” 3. The practical impulse for the first Athenian settlement at Sybaris would have been represented by the opportunity to provide for surplus citizens. The second dispatch of Athenian colonists and individuals from other Greek city-states to the new location of Thurii partly filled the need to reinforce the original Athenians of Sybaris. An allotment of land would have provided a strong economic incentive for the colonists. See Kagan 1969, 157, 166– 67. 4. See Arist. Pol. 2.5, 1267b– 1269a. On inventions (verb ε ξευρι σκειν), see especially chap. 2, n. 35 and corresponding text, in the present book.

Conclusion

269

traditions. Equally remarkable is the fact that one of the oikists was Lampon, a diviner (µαντις) and friend of Pericles.5 Plutarch anecdotally relates how Lampon once explained a unicorn as an ominous phenomenon, while Anaxagoras showed that the malformation of the animal’s cranium was due to natural causes. Modern historians maintain that Pericles doubtless accepted the scientific explanation as correct, while recognizing that the conservative religious perspective was comforting and necessary for the uneducated masses.6 Herodotus, for one, would not have been so ready to interpret the disagreement between the natural philosopher and the seer in this manner. The two explanations reflect the double level of causality that, as Herodotus teaches us, is applicable to the same phenomenon. In the new enterprise of Thurii, the planned cooperation of Protagoras and Lampon may have represented a deliberate match of the best sort of innovation with the healthiest respect for religious tradition.7 The invitation to Greeks of different cities to participate in the settlement, the decision to start over in a new site after eliminating the supremacist group of the original Sybarites,8 the constitution of democratic stamp—all these features are consistent with an experiment in diversity, equality, and harmony. The settlement seems to have employed a number of advisers renowned for their experience of the world at large. Herodotus, the sophos and the traveler, had much to tell about Greeks and non-Greeks, the dangers of leadership in all forms of government, the importance of negotiation and arbitration, the validity of different ways and points of view, the causality and the evils of war, the communality of experience and the role of the divine in the affairs of all men: for him, an unofficial role in the Thurii project was the perfect job. Here was a brand-new city that could learn from the wisdom of foreigners; it had the potential to fulfill the best values of the Greeks, avoid the mistakes of the past, be eudaimon without pleonexie, and grow in moderation without engineering its own decay. Here, the audiences from almost all over Greece, such as Herodotus had addressed in the past at the ephemeral gatherings of 5. See Graham 1964, 36– 37. 6. See Plut. Per. 6, on which see Ehrenberg 1948, 164– 65, quoted with approval by Kagan (1969, 168– 69). 7. Cf. the religious aspect of Hippodamus’ planned state, see Arist. Pol. 2.5.1267b. 8. For the troubles between the Sybarites and the Athenians who had taken part in the first settlement of Sybaris (probably in 446/445) and the move to the new site of Thurii, see Diod. 12.10– 11; Strabo 6.1.13.

270

Telling Wonders

Olympia or separately in their own city, were represented in a single place by a permanent Panhellenic community of fellow citizens.9 The new state must have soon turned out to be a disappointment to the original settlers.10 At about the same time as the conflict of Corcyra with Corinth that Thucydides regarded as one of the precipitating factors for the outbreak of the Peloponnesian War, Thurii experienced internal struggles of its own: perhaps the developments in mainland Greece increased the tension between the Athenian and the Peloponnesian elements of the colony’s diversified population.11 We may fantasize that Herodotus applauded both the decision of the Thurians to seek the arbitration of the oracle of Delphi and the Delphic response. Asked who was to be called the oikist of Thurii, the Pythia named none other than Apollo himself. This made the colony responsible for resolving its own internal troubles and gave it divine sanction for managing its external policy autonomously from the political pressures back home. The colony distanced itself from Athens and was content with a cautious and mostly inactive alliance.12 But Herodotus’ adoptive city now reflected in a microcosm contemporary intercity discords in the rest of Greece, just as his native Halicarnassus had been a sample of the earlier fights of the Greeks on behalf of constitutional freedom. The stasis in Thurii, as well as the stasis outside, 9. The population of Thurii was divided into ten tribes. One was for Athens; three were for Euboea, Ionia, and the islands, respectively; three were Peloponnesian tribes (Arcadians, Eleans, and Achaeans); and three tribes included other Dorians (from Boeotia, Amphictyonis, and Doris). See Diod. 12.10– 11. As Kagan observes (1969, 162– 63), Spartans and Corinthians must have been few and grouped with other Dorians—not surprisingly, since Sparta was underpopulated and Corinth had ample means for supporting her citizens. 10. Most scholars assume that Herodotus was one of them. See, e.g., Legrand 1932, 15. However, according to Parke (1946, 88– 89), Herodotus only went to Thurii just after the publication of the Histories in their current form (i.e., after 430), when the grant of citizenship and allotment of land made him able to support himself. Parke’s argument and the speculation that Herodotus gave up public recitation after his departure for Thurii are far from convincing, but the fact remains that we do not know when Herodotus joined the colony or, for that matter, whether or not he moved elsewhere at the end of his life. See Legrand 1932, 18– 19. 11. This is so at least according to Diod. 12.35, though Graham (1964, 198) remarks, “this may be an over-simplification, since there were settlers from many other parts of Greece.” The civil struggles in Thurii ca. 434 b.c. were preceded, early after the city’s foundation, by an unsuccessful war with the neighboring Spartan colony of Taras over the territory of Siris (see Strabo 6.1.15; cf. SIG 69). 12. On the subsequent history of Thurii and its alignment during the Peloponnesian War, see Graham 1964, 198– 99.

Conclusion

271

must have affected both Herodotus’ relationship with his audiences and the import of his message to them. Either as a citizen or as a performer, the Herodotus of the biographical tradition always seems to have attracted both praise and blame.13 At Thurii as at Halicarnassus and other cities of Greece in his earlier years, his position may not have been entirely stable and comfortable if the honors accorded to him were mixed 14 The charm of Herodotus’ narrative,  with hostile reactions and ␸θονος. the caution with which he expresses himself, his silences about current affairs, his diplomatic elusiveness in arbitrating differences, and his adoption of the mode of the ainos are joined to a sometimes breathtaking tactlessness and irony. The narrator of the Histories seems to both follow and disdain the Pindaric “norm of the polyp.”15 The allusive features of his discourse may have obscured for his audiences the true worth of his political message. At the same time, as even Plutarch’s later reaction demonstrates, they were not destined to hide his ambivalence (or what Plutarch unfairly calls viciousness) toward all sides. At the end of our study of the Histories, the best we can do is observe that “Herodotus of Halicarnassus” and “Herodotus of Thurii,” both attested in the programmatic first sentence, emerge as a strong unified 13. The biographical tradition is insistent in registering episodes of politically motivated public approval or disapproval to Herodotus. At Athens, he allegedly received an exorbitant honorarium granted to him on the proposal of one Anytus (according to Diyillus, FGrHist 73 F 3, cited at Plut. De Malign. Herod. 26 ⫽ Mor. 862A– B), and he was honored by the Council as a result of a reading from his work (Eus. Chron. Olymp. 83.4 ⫽ 83.3 in the Armenian version). According to Aristophanes the Boeotian (FGrHist 379 F 5, cited at Plut. De Malign. Herod. 31 ⫽ Mor. 864D), the Thebans refused payment to Herodotus, and their magistrates prevented him from conversing with the city’s young. See also n. 14 in the present chapter. 14. The Vita of Herodotus from the Suda relates that after helping to drive the tyrant out of Halicarnassus, Herodotus migrated to Thurii “after he saw himself being the object  of resentment/envy [␸θονουµενον] on the part of the citizens.” For the narrator of the  Histories as an object of ␸θονος, see 7.139.1, discussed especially in chap. 3, “Specific Glosses of Interpretation,” in the present book; his alter ego Artemisia is similarly the object  of ␸θονος (8.69.1). Xerxes’ odd generalization about strangers being less invidious to one another than are fellow citizens (7.237.2– 3) perhaps functions as an indirect reassurance of Herodotus’ goodwill to the Greeks. 15. See Gentili 1988, 132: “the difficult external circumstances in which Pindar’s professional duties have placed him are a visible source of embarrassment. It is as if the poet were unable to conceal completely his efforts to follow the precept of Amphiaraus and imitate the polyp that assumes the color of the rock to which it clings.” Gentili goes on to quote Pindar frag. 43 SM: “Let your mind, my son, behave like the skin / of the rockclinging beast of the sea / and consort with men of all nations; go along willingly with those around you, / change your thoughts to suit the seasons” (trans. Gentili and Cole).

272

Telling Wonders

subject to whom the logos as a whole belongs.16 The image of the road he travels as researcher and narrator shows him to be transient, unestablished, and uncommitted to one place. He is comfortable away from home and comfortable with the fragmented diversity of the world. Though powerless to coerce and unlikely to persuade, he is also more qualified than most of his fellow Greeks to undermine current assumptions. He is free like Artemisia, an original like Hippoclides, and subject, like Arion, only to the technical and ethical requirements of his task. He sets himself up as a foil to a variety of types that populate his logos and potentially his public: partisan citizens, conventional thinkers, the politically unfree, performers working for a patron, scientific or juridical histores with vested interests or perverted methods, poets who invent myths, tourists and mindless colonists who demean native traditions, and the overly rationalistic deniers of the work of the divine. At the center of Herodotus’ concern is the need to place the experience of the Greeks in a broader context. Only cultural comparison will allow them to define and assess correctly their language, their customs, their values, their public and private actions, and the cultural tendencies these reveal. The problematic embodied in the questions “What is Greek?” and “What is not Greek?” competes with the question about the legitimacy of formulating them in this way. The importance of these issues is related to the political significance, as it emerges from Herodotus’ text, of both the historical and the ethnological activity. The professional histor, his sources, his audience, and men in general are all retellers of the past as well as observers of customs. The thoughts they hold on these matters are bound to affect their actions toward others. Herodotus achieves a demythologized reconstruction of Greek resistance to the Persian invader and the later attempts of different groups of Greeks to interpret that event in their own ways. He portrays what the Greeks conceive Greekness to be as it actually manifested itself in precious moments at Marathon, Thermopylae, Salamis, and Plataea. His is,   16. The unanimous reading in the manuscript tradition is Ηροδοτου ‘Αλικαρνησσεος. But earlier sources and especially the quotation by Aristotle (Rhet. 3.9.2.1409a) give  Ηροδοτου Θουρι ου. Some later authors testify to his migration to Thurii by saying specifically that he eventually “was called Thurian” (see especially Strabo 14.2.16; Plut. De Malign. Herod. 35 ⫽ Mor. 868A). Plutarch attests that in his day, some manuscripts gave Θουρι ου (De Ex.13). Jacoby (1913, 205) and Legrand (1932, 12– 15) argue that Θουρι ου is the original reading, and Legrand has adopted it in his edition. It would be fitting, however, and no less a possibility, that both versions were authored by Herodotus and in circulation during his lifetime.

Conclusion

273

however, a qualified portrayal that reveals at least the fragility, if not the utter falsity, of the cultural superiority to which the Greeks lay a claim. From their success against the Persians, the Greeks have derived a selfconfidence and a sense of entitlement able to transform a defensive stance into aggression, first against the enemy, then against each other. This always occurs in the name of the same old antithesis between “us” and “them” that the conflict with the Persians seemed historically to embody. Who plays the civilizing role now and who are the Amazons, however, will depend on the point of view. To recognize one’s own subjectivity is equivalent to fulfilling the Delphic imperative “Know thyself” and is in turn the basis for acting according to “nothing in excess.” Not to recognize it falls under the heading of what we would call ethnocentrism and leads to oppressive, tyrannical behavior. For Herodotus of Halicarnassus and Thurii, who is centered in no place and belongs to nowhere, what is wrong with the Greeks of his time in their dealings with one another was already visible during their common resistance against the Persians and even earlier still. Their difficulties are partially due to a cognitive error he attempts to correct: their inadequate understanding of themselves and others. The subversiveness of his logos was perhaps destined to exceed Herodotus’ intention. It set out to explain causes, celebrate achievements, and present realistic paradigms drawn from long stretches of time and space; in the end, it produced a statement that the Greeks and Greekness itself were bound for an uncertain future. What remains in the Histories, so irresistibly seductive to this day, is the cumulative representation of the many different ways in which a civil society can be what it ought to be, coupled with an astonishing awareness of that goal’s always imperfect fulfillment.

Bibliography

Aly, W. 1921. Volksmarchen, ¨ Sage und Novelle bei Herodot und seinen Zeitgenossen. Gottingen. ¨ Reprint, 1969. ———. 1929. Formprobleme der fruher ¨ griechischen Prosa. Philol. Suppl. Leipzig. Amandry, P. 1950. La mantique apollinienne a` Delphi. Paris. Andrewes, A. 1938. “Eunomia.” CQ 32:89– 102. Armayor, O.K. 1987. “Hecataeus’ Humor and Irony in Herodotus’ Narrative of Egypt.” AW 16, nos 1– 2:11– 18. Asad, T. 1986. “The Concept of Cultural Translation in British Social Anthropology.” In Clifford and Marcus 1986, 141– 64. Asheri, D., ed. 1988. Erodoto: La Lidia e la Persia, Libro I delle storie, testo e commento. Milan. ———, ed. 1990a. Erodoto: La Persia, Libro III delle Storie, testo e commento. Milan. ———. 1990b. “Herodotus on Thracian Society and History.” In Nenci 1988, 131– 63. Austin, J.L. 1962. How To Do Things with Words. Cambridge, Mass. Austin, M.M., and P. Vidal-Naquet. 1977. Economic and Social History of Ancient Greece: An Introduction. Berkeley and Los Angeles. Austin, N. 1994. Helen of Troy and Her Shameless Phantom. Ithaca. Bach, K., and R. Harnish. 1979. Linguistic Communication and Speech Acts. Cambridge. Badian, E. 1993. From Plataea to Potidaea: Studies in the History and Historiography of the Pentecontaetia. Baltimore. ———. 1994. “Herodotus on Alexander I of Macedon: A Study in Some Subtle Silences.” In Hornblower 1994b, 107– 30. Bal, M. 1985. Narratology: Introduction to the Theory of Narrative. Toronto, Buffalo, and London. Baldry, H.C. 1965. The Unity of Mankind in Greek Thought. Cambridge. Barley, N. 1992. The Innocent Anthropologist. New York. Barth, H. 1968. “Zur Bewertung und Auswahl des Stoffes durch Herodotus (Die   Begriffe θωµα, θωµαζω, θωµασιος und θωµαστος).” Klio 50:93– 110. Barthes, R. 1970. S/Z. Paris. ———. 1986. “The Discourse of History.” In The Rustle of Language, trans. R.

275

276

Telling Wonders

Howard, 127– 40. New York. Originally published as “Le discours de l’histoire,” Social Science Information 6 (1967): 65– 75. Basso, K. 1976. “‘Wise Words’ of the Western Apache: Metaphor and Semantic Theory.” In Meaning in Anthropology, ed. K. Basso and H. Selby, 93– 121. Albuquerque. Beck, I. 1971. Die Ringomposition bei Herodot. Hildesheim and New York. Belsey, C. 1980. Critical Practice. London. Reprint, 1991. Benardete, S. 1969. Herodotean Inquiries. The Hague. Benedict, R. 1989. The Chrysanthemum and the Sword: Patterns of Japanese Culture. 1946. Reprint, with a foreword by E.F. Vogel, Boston. Benveniste, E. 1971. Problems of General Linguistics. Trans. M.E. Meek. Miami. Originally published as Probl`emes de linguistique gen´erale (Paris, 1967). Bernabo, ` L. 1977. “Oracoli come messaggio: Erodoto testimone di una dimensione orale dei responsi oracolari?” BIFG 4:157– 74. Bett, R. 1989. “The Sophists and Relativism.” Phronesis 34/2. 139– 69. Bickerman, E.J. 1952. “Origenes Gentium.” CP 47:65– 81. Bischoff, H. 1932. Der Warner bei Herodot. Diss., Marburg. Bloomer, W.M. 1993. “The Superlative Nomoi of Herodotus’s Histories.” CA 12:30– 50. Boedeker, D. 1987a. “The Two Faces of Demaratus.” In Boedeker 1987b, 185– 201. ———. 1993. “Hero Cult and Politics in Herodotus.” In Dougherty and Kurke 1993, 164– 77. ———, ed. 1987b. Herodotus and the Invention of History. Arethusa 20, nos. 1–2. ———. Forthcoming. “Mythical Patterns and Epic Heritage.” In de Jong, Bakker, and van Wees forthcoming. Boedeker, D., and K. Raaflaub, eds. 1998. Democracy, Empire, and the Arts in Fifth-Century Athens. Cambridge, Mass. Booth, W.C. 1983. The Rhetoric of Fiction. 2d ed. Chicago. Bornitz, F. 1968. Herodot Studien: Beitra¨ ge zum Versta¨ ndnis der Einheit des Geschichtswerkes. Berlin. Bowen, A.J., ed. 1992. Plutarch: The Malice of Herodotus. Trans. A.J. Bowen. Warminster. Bowie, E.L. 1993. “Lies, Fiction, and Slander in Early Greek Poetry.” In Gill and Wiseman 1993, 1– 37. Bowra, C.M. 1963. “Arion and the Dolphin.” MH 20:121– 34. Briant, P. 1988. “H´erodote et la societ´e perse.” In Nenci 1988, 69– 104. Broadhead, H.D., ed. 1960. The Persae of Aeschylus. Cambridge.  Brown, F.S., and W.B. Tyrrell. 1985. “Εκτιλ ωσαντο: A Reading of Herodotus’ Amazons.” CJ 80:297– 302. Brown, T. 1988. “Early Life of Herodotus.” AW 17:3– 15. Burkert, W. 1985. “Herodot uber ¨ die Namen der Gotter: ¨ Polytheismus als historisches Problem.” MH 43:121– 32. ———. 1988. “Herodot als Historiker fremdem Religionen.” In Nenci 1988, 1– 32.

Bibliography

277

Burn, A.R. [1962] 1984. Persia and the Greeks: The Defense of the West, c. 546– 478 b.c. 2d ed., with a postscript by D.M. Lewis. London. Castriota, D. 1992. Myth, Ethos, and Actuality: Official Art in Fifth-Century b.c. Athens. Madison. Chamberlain, D. 1997. “Herodotean Voices: Reading Characters in the Histories.” Ph.D. diss., University of California, Berkeley. ———. 1999. “On Atomics Onomastic and Metarhythmic Translations in Herodotus.” Arethusa 32, no. 3:263– 312. ———. Forthcoming. “‘We the Others’: Interpretive Community and Plural Voice in Herodotus.” Chatman, S. 1978. Story and Discourse: Narrative Structure in Fiction and Film. Ithaca. Chiasson, C.C. 1986. “The Herodotean Solon.” GRBS 27:249– 62. Christ, M.R. 1994. “Herodotean Kings and Historical Inquiry.” CA 13, no. 2:167– 202. Cingano. 1985. “Clistene di Sicione: Erodoto e i poemi del ciclo tebano.” QUCC 49:31– 40. Clifford, J. 1986. “On Ethnographic Allegory.” In Clifford and Marcus, 1986, 98– 121. Clifford, J., and G.E. Marcus, eds. 1986. Writing Culture: The Poetics and Politics of Ethnography. Berkeley and Los Angeles. Cloch´e, P. 1945. “P´ericl`es et la politique ext´erieure d’Ath`enes entre la paix de 446– 445 et les pr´eludes de la guerre du P´eloponn`ese.” AC 14:93– 128. Cobet, J. 1971. Herodots Exkurse und die Frage der Einheit seines Werke. Historia Einzelschriften 17. Wiesbaden. ———. 1977. “Wann Wurde Herodots Darstellung der Persekriege Publiziert?” Hermes 105:2– 27. Cole, T. 1967. Democritus and the Sources of Greek Anthropology. Chapel Hill. Reprint Atlanta, 1990. Connor, W.R. 1971. The New Politicians of Fifth-Century Athens. Princeton. ———. 1984. Thucydides. Princeton. ———. 1985. “Narrative Discourse in Herodotus and Thucydides.” In The Greek Historians: Literature and History, Papers Presented to A.E. Raubitschek, ed. M.H. Jameson, 1– 17. Saratoga, Calif. ———. 1987. “Tribes, Festivals, and Processions: Civic Ceremonial and Political Manipulation in Archaic Greece.” JHS 107:40– 50. ———. 1993. “The Histor in History.” In Rosen and Farrell 1993, 127– 40. Cook, A. 1976. “Herodotus: The Act of Inquiry as a Liberation from Myth.” Helios 3:23– 66. Cook, A.B. 1907. “Nomen Omen.” CR 21:69. Cooper, G.L. 1974. “Intrusive Oblique Infinitives in Herodotus.” TAPA 104:23– 76.  Constructions  (ως) ———. 1975. “The Ironic Force of the Pure Optative in οτι of the Primary Sequence.” TAPA 105:29– 34. Corcella, A. 1984. Erodoto e l’analogia. Palermo.

278

Telling Wonders

———, ed. 1993. Erodoto le Storie: La Scizia e la Libia, Libro IV testo e commento. Milan. Crahay, R. 1956. La litt´erature oraculaire chez H´erodote. Paris. Darbo-Peschanski, C. 1987. Le discours du particulier: Essai sur l’enquˆete H´erodot´eenne. Paris. Davies, J.K. 1971. Athenian Propertied Families. Oxford. Davison, J.A. 1962. “Literature and Literacy in Ancient Greece.” Phoenix 16:141– 56, 219– 33. Defradas, J. 1954. Les th`emes de la propagande delphique. Paris. de Jong, I.J.F. 1987. Narrators and Focalizers: The Presentation of the Story in the Iliad. Amsterdam. ———. 1998. “Aspects Narratologiques des Histoires d’H´erodote.” Lalies: Actes des sessions de linguistique et de litt´erature 19. de Jong, I.J.F., E. Bakker, and H. van Wees, eds. Forthcoming. A Companion to Herodotus. Leiden. Demand, N. 1987. “Herodotus’ Encomium of Athens: Science or Rhetoric?” AJP 108:746– 58. Denniston, J.D. 1934. The Greek Particles. Reprint, Oxford, 1987. de Romilly, J. 1971. “La vengeance comme explication historique dans l’oeuvre d’H´erodote.” REG 84:314– 37. de Sanctis, G. 1933. “Intorno al razionalismo di Ecateo.” RFC, n.s., 11: 5– 15. Detienne, M. 1973. Les maˆıtres de la verit´e dans la Gr`ece archa¨ıque. Paris. ———. 1977. The Gardens of Adonis: Spices in Greek Mythology. Trans. Janet Lloyd. New York. Reprint, Princeton, 1994. Originally published as Les jardins d’Adonis (Paris, 1972). ———. 1989. “Culinary Practices and the Spirit of Sacrifice.” In Detienne and Vernant 1989, 1– 20. Detienne, M., and J.P. Vernant. 1989. The Cuisine of Sacrifice among the Greeks. Trans. Paula Wissung. Chicago. Originally published as La cuisine du sacrifice en pays grec (Paris, 1979). Dewald, C. 1981. “Women and Culture in Herodotus’ Histories.” Women’s Studies 8, nos. 1– 2:93– 126. ———. 1985. “Practical Knowledge and the Historian’s Role in Herodotus and Thucydides.” In The Greek Historians: Literature and History, Papers Presented to A.E. Raubitschek, ed. M.H. Jameson, 47– 63. Saratoga, Calif. ———. 1987. “Narrative Surface and Authorial Voice in Herodotus’ Histories.” In Boedeker 1987b, 147– 70. ———. 1990. Review of Hartog 1988. CP 85, 217– 24. ———. 1993. “Reading the World: The Interpretation of Objects in Herodotus’ Histories.” In Rosen and Farrell 1993, 55– 70. ———. 1997. “Wanton Kings, Pickled Heroes, and Gnomic Founding Fathers: Strategies of Meaning at the End of Herodotus’ Histories.” In Classical Closure: Reading the End in Greek and Latin Literature, ed. D.H. Roberts, F.M. Dunn, and D. Fowler, 62– 82. Princeton. ———. 1998. Herodotus: The Histories. Trans. Robin Waterfield. Oxford.

Bibliography

279

———. 1999. “The Figured Stage: Focalizing the Initial Narratives of Herodotus and Thucydides.” In Contextualizing Classics: Ideology, Performance, Dialogue; Festschrift Peradotto, ed. N. Felson, D. Konstan, and T. Faulkner, 229– 61. Lanham, Md. ———. Forthcoming a. “Emoi ou genelogesanti emeouton: Herodotus and the Authorial Persona.” In de Jong, Bakker, and van Wees forthcoming. ———. Forthcoming b. “Form and Content: The Question of Tyranny in Herodotus.” In Popular Tyranny, ed. K. Morgan. Austin. Diels, H. 1910. “Die Anfa¨ nge der Philologie bei den Griechen.” Neue Jahrbucher ¨ fur ¨ das Klassische Altertum, Geschichte und Deutsche Literatur 25, no. 13:1– 25. Reprinted in H. Diels, Kleine Schriften zur Geschichte der Antiken Philosophie, 68– 92 (Hildesheim, 1969). Diller, H. 1961. “Die Hellenen-Barbaren-Antithese im Zeitalter der Persekriege.” In Grecs et Barbares, 39– 68. Fondation Hardt, Entretiens 8. Geneva. Dillery, J. 1992. “Darius and the Tomb of Nitocris (Hdt. 1.187).” CP 87:30– 38. ———. 1996. “Reconfiguring the Past: Thyrea, Thermopylae, and Narrative Patterns in Herodotus.” AJP 117:317– 54. Donadoni, S. 1947. “Erodoto, Plutarco e l’Egitto.” Belfagor 2:203– 8. Dorati, M. 2000. Le Storie di Erodoto: etnografia e recconto. Pisa and Rome. Dougherty, C., and L. Kurke, eds. 1993. Cultural Poetics in Archaic Greece: Cult, Performance, Politics. Cambridge. Drews, R. 1970. “Herodotus’ other logoi.” AJP 91:181– 91. ———. 1973. The Greek Accounts of Eastern History. Cambridge, Mass. Drexler, H. 1972. Herodot Studien. Hildesheim. DuBois, P. 1982. Centaurs and Amazons: Women and the Pre-History of the Great Chain of Being. Ann Arbor. Durand, J.-L. 1989. “Greek Animals: Towards a Topology of Edible Bodies.” In Detienne and Vernant 1989, 87– 118. Easterling, P.E., and J.V. Muir, eds. 1985. Greek Religion and Society. Cambridge. Eco, U. 1962. Opera aperta: Forma e indeterminazione nelle poetiche contemporanee. Milan. Ehrenberg, V. 1948. “The Foundation of Thurii.” AJP 59:149– 70. Elayi, J. 1979. “Deux oracles de Delphes: Les r´eponses de la Pythie a` Clisth`ene de Sicyon et aux Ath´eniens avant Solamine.” REG 92:224– 30. Else, G.F. 1958. “‘Imitation’ in the Fifth Century.” CP 53:73– 90. Erbse, H. 1956. “Der erste Satz im Werke Herodots.” In Festschrift B. Snell, 209– 22. Munich. ———. 1961. “Tradition und Form in Werke Herodots.” Gymnasium 68: 339– 57. Euben, J.P., ed. 1986. Greek Tragedy and Political Theory. Berkeley. Evans, J.A.S. 1961. “The Dream of Xerxes and the ‘Nomoi’ of the Persians.” CJ 57:109– 11. ———. 1965. “Despotes Nomos.” Athenaeum 43:145– 53. ———. 1968. “Father of History or Father of Lies: The Reputation of Herodotus.” CJ 64:11– 17.

280

Telling Wonders

———. 1976. “Herodotus and the Ionian Revolt.” Historia 25, no. 1:31– 37. ———. 1979. “Herodotus and Athens: The Evidence of the Encomium.” AC 48:112– 18. ———. 1980. “Oral Tradition in Herodotus.” Canadian Journal of Oral History 4, no. 2:8– 16. ———. 1981. “Notes on the Debate of the Persian Grandees in Herodotus, 3.80– 82.” QUCC 45:79– 54. ———. 1982. Herodotus. Boston. ———. 1988. “The Medism of Pausanias: Two Versions.” Antichthon 22: 1– 11. ———. 1991. Herodotus, Explorer of the Past. Princeton. Fehling, D. 1989. Herodotus and His “Sources”: Citation, Invention, and Narrative Art. Trans. J.G. Howie. Leeds. Originally published as Die Quellenangaben bei Herodot (Berlin and New York, 1971). Ferrill, A. 1978. “Herodotus on Tyranny.” Historia 27:385– 98. Fetterman, D.M. 1989. Ethnography Step by Step. Newbury Park, Calif. Figuera, T.J. 1985. “Herodotus and the Early Hostilities between Aegina and Athens.” AJP 106:49– 74. ———. 1986. “Population Patterns in Late Archaic and Classical Sparta.” TAPA 116:165– 212. Finley, M.I. 1968. “Sparta.” In Probl`emes de la guerre en Gr`ece ancienne, ed. J.P. Vernant, 143– 60. Paris. ———. 1980. Ancient Slavery and Modern Ideology. London. Fisher, N.R.E. 1988. “Drink, Hybris and the Promotion of Harmony in Sparta.” In Powell 1988b. Flory, S. 1978. “Arion’s Leap: Brave Gestures in Herodotus.” AJP 99:441– 21. ———. 1980. “Who Read Herodotus’ Histories?” AJP 101:12– 28. ———. 1987. The Archaic Smile of Herodotus. Detroit. Flower, H. 1991. “Herodotus and Delphic Traditions about Croesus.” In Georgica: Greek Studies in Honor of George Cawkwell, ed. M.A. Flower and M. Toher, 57– 77. London. Ford, A. 1990. “Unity in Greek Criticism and Poetry.” Review of Unity in Greek Poetics, by M. Heath. Arion 3, no. 1:125– 54. Fornara, C.W. 1966. “Some Aspects of the Career of Pausanias of Sparta.” Historia 15, no. 3:257– 71. ———. 1971a. Herodotus: An Interpretive Essay. Oxford. ———. 1971b. “Evidence for the Date of Herodotus’ Publication.” JHS 91: 25– 34. ———. 1981. “Herodotus’ Knowledge of the Archidamian War.” Hermes 109: 149– 56. ———. 1983. The Nature of History in Ancient Greece and Rome. Berkeley. ———. 1990. “Human History and the Constraint of Fate in Herodotus.” In Conflict, Antithesis and the Ancient Historian, ed. J.W. Allison, 25– 45. Columbus, Ohio. Fornara, C.W., and L.J. Samons II. 1991. Athens from Cleisthenes to Pericles. Berkeley.

Bibliography

281

Forrest. W.G. 1966. The Emergence of Greek Democracy: 800– 400 b.c. New York. ———. 1984. “Herodotus and Athens.” Phoenix 38:1– 11. Fowler, R.L. 1996. “Herodotos and His Contemporaries.” JHS 116:62– 87. Fra¨ nkel, H. 1924. “Eine Stileigenheit der fruhgriechischen ¨ Literatur.” NAWG 63– 126. Reprinted in Wege und Formen der fruhgriechischen ¨ Denkens, 2d ed., 40– 96 (Munich, 1962). French, A. 1972. “Topical Influences in Herodotus’ Narrative.” Mnemosyne 25:9– 27. Frisch, P. 1968. Die Traume ¨ bei Herodot. Beitrage ¨ zur klassische Philologie 27. Meisenheim am Glam. Froidefond, C. 1971. Le mirage e´ gyptien dans la lit´erature grecque d’Hom`ere a` Aristote. Paris. Gammie, J.G. 1986. “Herodotus on Kings and Tyrants: Objective Historiography or Conventional Portraiture?” JNES 45:171– 95. Gantz, T. 1993. Early Greek Myth: A Guide to Literary and Artistic Sources. Baltimore and London. Garlan, Y. 1988. Slavery in Ancient Greece. Trans. J. Lloyd. Ithaca. Originally published as Les esclaves en Gr`ece ancienne (Paris, 1982). Garland, R. 1985. The Greek Way of Death. Ithaca. Geertz, C. 1973. The Interpretation of Cultures. New York. ———. 1983. Local Knowledge. New York. ———. 1984. “Anti-anti Relativism.” American Anthropologist 86:263– 78. ———. 1986. “The Uses of Diversity.” In The Tanner Lectures in Human Values 7:253– 75. Cambridge. ———. 1988. Works and Lives: The Anthropologist as Author. Stanford. Genette, G. 1980. Narrative Discourse: An Essay in Method. Trans. J.E. Lewin. Foreword by J. Culler. Ithaca, NY. Originally published as Discours du r´ecit (Paris, 1972). Gentili, B. 1988. Poetry and Its Public in Ancient Greece. Trans. A.T. Cole. Baltimore. Originally published as Poesia e pubblico nella Grecia antica (Rome, 1985). Georges, P. 1994. Barbarian Asia and the Greek Experience. Baltimore. Gigante, M. 1956. Nomos Basileus. Naples. Gill, C., and T.P. Wiseman, eds. 1993. Lies and Fiction in the Ancient World. Austin. Gillis, D. 1979. Collaboration with the Persians. Historia Einzelnschriften 34. Stuttgart. ´ Giraudeau, M. 1984. Les notions juridiques et sociales chez H´erodote: Etudes sur le vocabulaire. Paris. Goff, B., ed. 1995. History, Tragedy, Theory: Dialogues on Athenian Drama. Austin. Gomme, A.W. 1956. A Historical Commentary on Thucydides, Vol. II. Oxford. Gould, J. 1985. “On Making Sense of Greek Religion.” In Easterling and Muir 1985, 1– 33. ———. 1989. Herodotus. London.

282

Telling Wonders

———. 1994. “Herodotus and Religion.” In Hornblower 1994b, 91– 106. Graham, A.J. 1964. Colony and Mother City in Ancient Greece. Manchester. Gray, V. 1996. “Herodotus and Images of Tyranny: The Tryants of Corinth.” AJP 117:361– 89. ———. 1997. “Reading the Rise of Pisistratus: Herodotus 1.56– 69.” Histos 1 (August). Available at ⬍http://www.dur.ac.uk/histos⬎ (May 2000). Greenblatt, Stephen. 1991. Marvelous Possessions: The Wonder of the New World. Chicago. Grene, D., trans. 1987. Herodotus, the History. Chicago. Griffith, G.T. 1966. “Isegorie in the Assembly in Athens.” In Ancient Society and Institutions: Studies Presented to Victor Ehrenberg on his Seventy-Fourth Birthday, ed. E. Badian, 115– 38. Oxford. Griffiths, A. 1988. “Was Cleomenes Mad?” In Powell 1988b, 51– 78. Guthrie, W.H.C. 1950. The Greeks and Their Gods. Boston. ———. 1965. A History of Greek Philosophy. Vol. 2. Cambridge. ———. 1969. A History of Greek Philosophy. Vol. 3. Cambridge. ———. 1971. The Sophists. Cambridge. Hall, E. 1989. Inventing the Barbarian: Greek Self-Definition through Tragedy. Oxford. Hall, J.M. 1997. Ethnic Identity in Greek Antiquity. Cambridge. Hanson, F.A. 1979. “Does God Have a Body? Truth, Reality, and Cultural Relativism.” Man, n.s., 14:515– 29. Hanson, V.D. 1989. The Western Way of War: Infantry Battle in Classical Greece. Oxford. Harvey, D. 1994. “Laconica: Aristophanes and the Spartans.” In Powell and Hodkinson 1994, 35– 58. Halperin, D.M. 1990. One Hundred Years of Homosexuality and Other Essays on Greek Love. New York. Harmatta, J. 1988. “Herodotus, Historian of the Cimmerians and the Scythians.” In Nenci 1988, 115– 30. Harrison, T. 1998. “Herodotus’ Conception of Foreign Languages.” Histos 2 (March). Available at ⬍http://www.dur.ac.uk/histos⬎ (May 1998). ———. 2000. Divinity and History: The Religion of Herodotus. Oxford. Hart, J. 1982. Herodotus and Greek History. London. Hartog, F. 1988. The Mirror of Herodotus: The Representation of the Other in the Writing of History. Trans. J. Lloyd. Berkeley. Originally published as Le miroir d’Herodote (Paris, 1980). Hatch, E. 1983. Culture and Morality: The Relativity of Values in Anthropology. New York. Havelock, E.A. 1982. The Literate Revolution in Greece and Its Cultural Consequences. Princeton. Heidel, W.A. 1935. “Hecataeus and the Egyptian Priests in Herodotus, Book II.” American Academy of Arts and Sciences 18, no. 2:53– 134. Heinimann, F. 1945. Nomos und Physis. Basel. Reprint, 1965. Hellmann, F. 1934. Herodots Kroisos-Logos. Berlin. Helms, L.V. 1882. Pioneering in the Far East and Journeys to California in 1843 and to the White Sea in 1848. London.

Bibliography

283

Hohti, P. 1974. “Freedom of Speech in Speech Sections in the Histories of Herodotus.” Arctos 8:19– 27. ¨ ———. 1975. “Uber die Notwendigkeit bei Herodot.” Arctos 9:31– 37. ———. 1977. “συµβαλλεσ␽αι: A Note on Conjectures in Herodotus.” Arctos 11:5– 14. Holscher, ¨ T. 1998. “Images and Political Identity: The Case of Athens.” In Boedeker and Raaflaub 1998, 153– 83. Hornblower, S. 1987. Thucydides. Baltimore. ———. 1991. A Commentary on Thucydides. Vol. 1, Books I– III. Oxford. ———. 1992. “Thucydides’ Use of Herodotus.” In Philolakon: Lakonian Studies in Honor of Hector Catling, ed. J.M. Sanders, 141– 54. Athens. ———. 1994a. “Narratology and Thucydides.” In Hornblower 1994b, 130– 63. ———, ed. 1994b. Greek Historiography. Oxford. How, W.W., and J. Wells. 1928. A Commentary on Herodotus. 2 vols. Oxford. Reprint, 1964. Howald, E. 1923. “Ionische Geschichtsschreibung.” Hermes 58:113. ———. 1944. Vom Geist antiker Geschichtsschreibung. Munich. Humphreys, S. 1987. “Law, Custom, and Culture in Herodotus.” In Boedeker 1987b, 211– 20. Hunter, V. 1982. Past and Process in Herodotus and Thucydides. Princeton. Immerwahr, H.R. 1954. “Historical Action in Herodotus.” TAPA 85:16– 45. ———. 1956. “Aspects of Historical Causation in Herodotus.” TAPA 87: 247– 80. ———. 1960. “Ergon: History as a Monument in Herodotus and Thucydides.” AJP 81:261– 90. ———. 1966. Form and Thought in Herodotus. Cleveland. Jacobson, R. 1960. “Closing Statement: Linguistics and Poetics.” In Style in Language, ed. T.A. Sebeok, 350– 76. New York. Jacoby, F. 1913. “Herodotus.” RE Suppl. 2: 205– 520. ———. 1944. “GENESIA: A Forgotten Festival of the Dead.” CQ 38:65. Johnson, W.A. 1994. “Oral Performance and the Composition of Herodotus’ Histories.” GRBS 35:229– 54.  Jones, C.P. 1996. “Εθνος and γενος in Herodotus.” CQ 46:315– 20. Jordan, B. 1986. “Religion in Thucydides.” TAPA 116:119– 47. Jouanna, J. 1981. “Les causes de la defaite des barbares chez Eschyle, H´erodote et Hippocrate.” Ktema 7:3– 15. Kagan, D. 1969. The Outbreak of the Peloponnesian War. Ithaca. Kallet, L. 1998. “Accounting for Culture in Fifth-Century Athens.” In Boedeker and Raaflaub 1998, 43– 58. Kennedy, G.A., trans. 1991. Aristotle, On Rhetoric: A Theory of Civic Discourse. Oxford. Kerferd, G.B. 1981. The Sophistic Movement. Cambridge. Kindstrand, J.F. 1981. Anacharsis: The Legend and the Apophthegmata. Uppsala. Kirchberg, L. 1965. Die Funtion der Orakel im Werke Herodots. Gottingen. ¨ Klees, H. 1964. Die Eigenart des griechischen Glaubens an Orakel und Seher. Stuttgart.

284

Telling Wonders

   : Untersuchungen zur Geschichte einer ευρετ ης Kleingunther, ¨ A. 1933. Πρωτος Fragestellung. Leipzig. Reprint, 1975. Kleinknecht, H. 1940. “Herodot und Athen. 7.139, 8.140– 144.” Hermes 75:241– 64. Knox, B. 1952. “Lion in the House.” CP 47:17– 25. Reprinted in B. Knox, Word and Action: Essays on the Ancient Theater, 27– 38 (Baltimore, 1979). ———. 1961. “The Ajax of Sophocles.” HSCP 65:1– 36. Reprinted in B. Knox, Word and Action: Essays on the Ancient Theater, 125– 60 (Baltimore, 1979). ———. 1971. Oedipus at Thebes: Sophocles’ Tragic Hero and His Time. New York. Originally published in New Haven, 1957. Konstan, D. 1983. “The Stories in Herodotus’ Histories: Book I.” Helios 10, no. 1:1– 22. ———. 1987. “Persians, Greeks, and Empire.” In Boedeker 1987b, 59– 70. Kothe, H. 1969. “Der Skythenbegriff bei Herodot.” Klio 51:15– 88. Krischer, T. 1965. “Herodots Prooimion.” Hermes 93:159– 67. Kurke, L. 1995. “Herodotus and the Language of Metals.” Helios 22, no. 1: 36– 64. Labov, W. 1972. Language in the Inner City: Studies in the Black English Vernacular. Philadelphia. Labov, W., and J. Waletzky. 1966. “Narrative Analysis: Oral Versions of Personal Experience.” In Essays on Verbal and Visual Arts: Proceedings of the 1966 Annual Spring Meeting of the American Ethnological Society, ed. J. Helm, 12– 44. Seattle. Lachenaud, G. 1978. Mythologies, religion, et philosophie de l’histoire dans H´erodote. Lille and Paris. Lang, M.L. 1984. Herodotean Narrative and Discourse. Cambridge, Mass. Lardinois, A.P. 1995. “Wisdom in Context: The Use of Gnomic Statements in Archaic Greek Poetry.” Ph.D. diss., Princeton University. Lasserre, F. 1976. “H´erodote et Protagoras: Le d´ebat sur les constitutions.” MH 33, no. 2:65– 84. Lateiner, D. 1977. “No Laughing Matter: A Literary Tactic in Herodotus.” TAPA 107:173– 82. ———. 1980. “A Note on dikas didonai in Herodotus.” CQ 30:30– 32. ———. 1984. “Herodotean Historiographical Patterning: The Constitutional Debate.” QS 20:257– 84. ———. 1985. “Polarita` : Il principio della differenza complementare.” QS 22: 79– 103. ———. 1986. “The Empirical Element in the Methods of Early Greek Medical Writers and Herodotus: A Shared Epistemological Response.” Antichthon 20:1– 26. ———. 1989. The Historical Method of Herodotus. Toronto. ———. 1990. “Deceptions and Delusions in Herodotus. CA 9, no. 2: 229– 46. Lattimore, R. 1939. “The Wise Advisor in Herodotus.” CP 34:24– 39. ———. 1958. “The Composition of the History of Herodotus.” CP 53:9– 21. Legrand, E. 1932. H´erodote: Introduction. Paris. ———, ed. 1946– 60. H´erodote. Vols. 2– 11. Paris.

Bibliography

285

Lesky, A. 1966. A History of Greek Literature. Trans. J. Willis and C. de Heer. New York. Originally publsihed as Geschichte der griechischen literatur (Bern, 1957). L´evˆeque, P., and P. Vidal-Naquet. 1996. Cleisthenes the Athenian: An Essay on the Representation of Space and Time in Greek Political Thought. Trans. D.A. Curtis. Atlantic Highlands, N.J. Originally published as Clisth`ene l’ath´enien (Besan¸con, 1964). Levy, E. 1981. “Les origines du mirage Scythe.” Ktema 6:57– 68. Lewis, D.H. 1985. “Persians in Herodotus.” In The Greek Historians: Literature and History, Papers Presented to A.E. Raubitschek, ed. M.H. Jameson, 101– 17. Saratoga, Calif.  Linforth, I.M. 1918. “Οι αθανατ ι ζοντες (Hdt. 4.93– 96).” CP 13:23– 33. ———. 1924. “Herodotus’s Avowal of Silence in his Account of Egypt.” UCPCPh 9:269– 92. ———. 1926. “Greek Gods and Foreign Gods in Herodotus.” UCPCPh 9:1– 25. ———. 1928. “Named and Unnamed Gods in Herodotus.” UCPCPh 9:201– 43. Lissarague, F. 1990. The Aesthetics of the Greek Banquet. Trans. A. SzegedyMaszak. Princeton. Originally published as Un Flot d’Images: une esth´etique du banquet grec (Paris, 1987). Lloyd, A.B. 1975. Herodotus, Book II: Introduction. Leiden. ———. 1976. Herodotus, Book II: Commentary, 1– 98. Leiden. ———. 1988a. Herodotus, Book II: Commentary, 99– 182. Leiden. ———. 1988b. “Herodotus on Egyptians and Libyans.” In Nenci 1988, 215– 44. Lloyd, G.E.R. 1966. Polarity and Analogy: Two Types of Argumentation in Early Greek Thought. Cambridge. Lloyd-Jones, H. 1971. The Justice of Zeus. Berkeley. Lombardo, M. 1988. “Erodoto, Storico dei Lidi.” In Nenci 1988, 171– 203. Long, T. 1984. Barbarians in Greek Comedy. Carbondale and Edwardsville. ———. 1987. Repetition and Variation in the Short Stories of Herodotus. Frankfurt am Main. Loraux, N. 1977. “La ‘belle mort’ spartiate.” Ktema 2:105– 20. ———. 1986. The Invention of Athens: The Funeral Oration in the Classical City. Trans. A. Sheridan. Cambridge, Mass. Originally published as L’invention d’Ath`enes: Histoire de l’oraison fun`ebre dans la “cit´e classique.” (Paris, 1981). Lovejoy, A.O., and G. Boas. 1935. Primitivism and Related Ideas in Antiquity. Baltimore. Reprint, New York, 1965. Macan, R.W. 1895. Herodotus: The Fourth, Fifth, and Sixth Books. 2 vols. London. Reprint, New York, 1973. ———. 1908. Herodotus: The Seventh, Eighth, and Ninth Books. 2 vols. London. Marg, W., ed. 1965. Herodot: Eine Auswahl aus der neueren Forschung. Darmstadt. Marincola, J. 1987. “Herodotean Narrative and the Narrator’s Presence.” In Boedeker 1987b, 121– 38. ———. 1997a. Authority and Tradition in Ancient Historiography. Cambridge. ———. 1997b. “Odysseus and the Historians.” Histos 1 (October): 1– 26. Available at ⬍http://www.dur.ac.uk/histos⬎ (May 1999).

286

Telling Wonders

Martin, R.P. 1989. The Language of Heroes: Speech and Performance in the Iliad. Princeton. Masaracchia, A. 1975. Studi Erodotei. Rome. ———, ed. 1977. Erodoto: La Battaglia di Salamina, Libro VIII delle Storie, testo e commento. Milan. ———, ed. 1978. Erodoto: La Sconfitta dei Persiani, Libro IX delle Storie, testo e commento. Milan. Matthews, V. 1974. Panyassis of Halicarnassus. Leiden. Mazzarino, S. 1966. Il Pensiero Storico Classico. Vol. 1. Bari. McGlew, J.F. 1993. Tyranny and Political Culture in Ancient Greece. Ithaca. Meier, C. 1990. The Greek Discovery of Politics. Trans. D. McLutock. Cambridge, Mass. Originally published as Die Entstehung des Politischen bei den Griechen (Frankfurt, 1980). Mitchell, L.G. 1997. Greeks Bearing Gifts: The Public Use of Private Relationships. Cambridge. Moles, J. 1993. “Truth and Untruth in Herodotus and Thucydides.” In Gill and Wiseman 1993, 88– 121. ———. 1996. “Herodotus Warns the Athenians.” Papers of the Leeds International Seminar 9:259– 84. Momigliano, A.D. 1954. “On the Causes of War in Ancient Historiography.” In Acta Congressus Madvigiani: Proceedings of the Second International Congress of Classical Studies, 1:199– 211. Reprinted in A.D. Momigliano, Studies in Historiography (New York, 1966), 112– 26. ———. 1958. “The Place of Herodotus in the History of Historiography.” History 43:1– 13. Reprinted in A.D. Momigliano, Studies in Historiography (New York, 1966), 127– 41. ———. 1975a. Alien Wisdom: The Limits of Hellenization. Cambridge. ———. 1975b. “The Fault of the Greeks.” Daedalus 104, no. 2:1– 15. ———. 1978. “The Historians of the Classical World and Their Audiences: Some Suggestions.” ASNP, 3rd ser., 8:59– 75. Mora, F. 1985. Religione a Religioni nelle Storie di Erodoto. Milan. Morris, I. “Beyond Democracy and Empire: Athenian Art in Context.” In Boedeker and Raaflaub 1998, 59– 86. Muller, ¨ D. 1980. Satzbau, Satzgliederung und Satzverbindung in der Prosa Herodots. Meisenheim am Glam. Muller, ¨ K.E. 1972. Geschichte der antiken Ethnographie und ethnologischen Theoriebildung, Vol. 1. Wiesbaden. Munson, R.V. 1983. “Transitions in Herodotus.” Ph.D. diss., University of Pennsylvania. ———. 1986. “The Celebratory Purpose of Herodotus: The Story of Arion in Histories 1.23– 24.” Ramus 15, no. 2:93– 104. ———. 1988. “Artemisia in Herodotus.” CA 7, no. 1:91– 106. ———. 1991. “The Madness of Cambyses (Herodotus 3.16– 38).” Arethusa 24:43– 65. ———. 1993a. “Herodotus’ Use of Prospective Sentences and the Story of Rhampsinitus and the Thief in the Histories.” AJP 114:27– 44.

Bibliography

287

———. 1993b. “Three Aspects of Spartan Kingship in Herodotus.” In Rosen and Farrell 1993, 39– 54. ———. Forthcoming. “Αναγκη in Herodotus.” JHS. Murray D. 1987. “Herodotus and Oral History.” In Achaemenid History, vol. 1, Sources, Structures, and Syntheses, ed. H. Sancisi-Weerdenberg, 93– 115. Leiden. Myres. J.L. 1953. Herodotus, Father of History. Oxford. Reprint, Chicago, 1971. Nagy, G. 1979. The Best of the Achaeans: Concepts of the Hero in Archaic Greek Poetry. Baltimore. ———. 1990. Pindar’s Homer: The Lyric Possession of an Epic Past. Baltimore. Nenci, G. 1957. “La concezione del miracoloso nei poemi omerici.” Atti dell’Accademia di Scienze 2, Classe di scienze morali, storiche e filologiche 92:275– 311. ———. 1962. “La concezione del miracoloso in Esiodo.” Critica Storica 31: 251– 57. ———, ed. 1988. H´erodote et les peuples non grecs. Fondation Hardt, Entretiens 35. Geneva. ———. 1994. Erodoto, le Storie: La rivolta della Ionia, Libro V, testo e commento. Milan Nilsson, H.P. 1955. Geschichte der griechischen Religion, Vol. 1. Munich. North, H. 1966. Sophrosune. Ithaca. Nussbaum, M. 1986. The Fragility of Goodness. Cambridge. Ober, J. 1989. Mass and Elite in Democratic Athens. Princeton. ———. 1993. “The Athenian Revolution of 508/7 B.C.E.: Violence, Authority, and the Origins of Democracy.” In Dougherty and Kurke 1993, 215– 32. Ogden, D. 1993. “Cleisthenes of Sicyon, ΛΕΥΣΤΗΡ.” CQ 43:353– 63. Ostwald, M. 1965. “Pindar, ΝΟΜΟΣ, and Heracles (Pindar frg. 169 [Snell3] and P.Oxy. No. 1450, frg. 1).” HSCP 69:109– 38. ———. 1969. Nomos and the Beginning of the Athenian Democracy. Oxford. ———. 1986. From Popular Sovereignty to the Sovereignty of Law. Berkeley. ———. 1988. Αναγκη in Thucydides. Atlanta. ———. 1991. “Herodotus and Athens.” ICS 16:137– 48. ———. 1995. “Freedom and the Greeks.” In The Origins of Modern Freedom in the West, ed. R.W. Davis, 35– 63. Stanford. Packman, Z.M. 1991. “The Incredible and the Incredulous: The Vocabulary of Disbelief in Herodotus, Thucydides, and Xenophon.” Hermes 119:399– 414. Pagel, K.A. 1927. “Die Bedeutung des aitiologischen Moments fur ¨ Herodots Geschichtschreibung.” Ph.D. diss., Berlin. Parke, H.W. 1946. “Citation and Recitation: A Convention in Early Greek Historians.” Hermathena 67:80– 92. Parsons, P.J. 1992. “3965. Simonides, Elegies.” Oxyrhynchus Papyri 59:4– 50. Payen, Pascale. 1997. Les ˆıles nomades: Conqu´erir et r´esister dans l’Enquˆete d’H´erodote. Paris. Pearce, T.E.V. 1981. “‘Epic Regression’ in Herodotus.” Eranos 79:87– 90. Pearson, L. 1939. Early Ionian Historians. Oxford. ———. 1941. “Credulity and Scepticism in Herodotus.” TAPA 72:335– 55.

288

Telling Wonders

Pelling, C. 1997. “East Is East and West Is West—or Are They? National Stereotypes in Herodotus.” Histos 1 (March). Available at ⬍http://www.dur.ac.uk/ histos⬎ (September 1998). Pembroke, S.G. 1967. “Women in Charge: The Function of Alternatives in Early Greek Tradition and the Ancient Idea of Matriarchy.” JWCI 30:1– 35. Perlman, S. 1976. “Panhellenism, the Polis, and Imperialism.” Historia 25:1– 30. Podlecki, A.J. 1977. “Herodotus in Athens?” In Greece and the Eastern Mediterranean: Studies Presented to F. Schacherneyer on the Occasion of His Eightieth Birthday, ed. K.H. Kinzl, 246– 65. Berlin. Pohlenz, M. 1937. Herodot, der erste Geschichtsschreiber des Abendlandes. Leipzig. Reprint, Darmstadt, 1973. Powell, A. 1988a. “Mendacity and Sparta’s Use of the Visual.” In Powell 1988b, 173– 92. ———, ed. 1988b. Classical Sparta: Techniques Behind Her Success. Oklahoma Series in Classical Culture. London. Powell, A., and S. Hodkinson, eds. 1994. The Shadow of Sparta. London. Powell, J.E. 1938. A Lexicon to Herodotus. Cambridge. Reprint, Hildesheim, 1950. ———. 1939. The Histories of Herodotus. Cambridge. Pratt, M.L. 1977. Toward a Speech-Act Theory of Literary Discourse. Bloomington. ———. 1986. “Fieldwork in Common Places.” In Clifford and Marcus 1986, 27– 50. Prince, G. 1973. “Introduction a` l’´etude du narrataire.” Po´etique 14:178– 96. ———. 1977. “Remarques sur les signes m´etanarratifs.” Degr´es 11, no. 12: 1– 10. ———. 1980. “Notes on the Text as Reader.” In The Reader in the Text: Essays on Audience and Interpretation, ed. S.R. Suleiman and I. Closman, 225– 40. Princeton. ———. 1982. Narratology: The Form and Function of Narrative. Berlin, New York, and Amsterdam. ———. 1987. A Dictionary of Narratology. Lincoln. Pritchett, W.N. 1979. The Greek State at War. Vol. 2, Religion. Berkeley. ———. 1993. The Liar School of Herodotos. Amsterdam. Raaflaub, K.A. 1979. “Polis Tyrannos: Zur Entstehung einer politischen Metapher.” Arktouros: Hellenic Studies Presented to B.M.W. Knox, ed. G.W. Bowersock et al., 239– 41. Berlin and New York. ———. 1987. “Herodotus, Political Thought, and the Meaning of History.” In Boedeker 1987b, 221– 48. ———. Forthcoming. “Herodotus and Current Intellectual Trends.” In de Jong, Bakker and van Wees forthcoming. Raubitschek, A. 1939. “Εργα µεγαλα τε κα ι* θωµαστα.” RAE 41:217– 22. Rawlinson, G., trans. 1880. History of Herodotus. 4th ed. 4 vols. New York. Redfield, J. 1985. “Herodotus the Tourist.” CP 80:97– 118. Richardson, S. 1990. The Homeric Narrator. Nashville. Robinson, T.M. 1979. Contrasting Arguments: An Edition of the Dissoi Logoi. New York.

Bibliography

289

Romm, J.S. 1989. “Herodotus and Mythic Geography: The Case of the Hyperboreans.” TAPA 119:97– 117. ———. 1992. The Edges of the Earth in Ancient Thought, Geography, Exploration, and Fiction. Princeton. ———. 1998. Herodotus. New Haven. Rood, T. 1998. Thucydides: Narrative and Explanation. Oxford. Rosaldo, R. 1986. “From the Door of His Tent: The Fieldworker and the Inquisitor.” In Clifford and Marcus 1986, 77– 97. ———. 1989. Culture and Truth: The Remaking of Social Analysis. Boston. Rose, P.W. 1995. “Historicizing Sophocles’ Ajax.” In Goff 1995, 59– 90. Rosellini, M., and S. Sa¨ıd. 1978. “Usages de femmes et autres nomoi chez les ‘Sauvages’ d’H´erodote.” ASNP, 3d ser., 8:849– 1005. Rosen, R.M., and J. Farrell, eds. 1993. Nomodeiktes: Greek Studies in Honor of Martin Ostwald. Ann Arbor. Rosler, ¨ W. 1991. “Die ‘Selbhistorisierung’ des Autors: Zur Stellung Herodots zwischen Mundlichkeit ¨ und Schriftlichkeit.” Philologus 135:215– 20. Sa¨ıd, E.W. 1978. Orientalism. London. Sa¨ıd, S. 1980. “Guerre, Intelligence et Courage dans les Histoires d’H´erodote.” Ancient Society: 11– 12, 83– 117. ———. 1998. “Tragedy and Politics.” In Boedeker and Raaflaub 1998, 275– 95. Salmon, A. 1956. “L’Experience de Psamm´etique (II.2).” EC 24:321– 29. Samons, L. 1998. “Kimon, Kallias, and Peace with Persia.” Historia 47:129– 40. Schmid, W., and O. Sta¨ hlin. 1934. Geschichte der Griechischen Literatur. Munich. 1/2. 550– 673. Schmitt, R. 1976. “The Medo-Persian Names of Herodotus in the Light of the New Evidence from Persepolis.” Acta Antiqua Academiae Scentiarum Hungaricae 24:25– 35.   Schroeder, O. 1917. “Νοµος ο παντων βασιλευς.” Philologus 74:195– 204. Searle, J.R. 1969. Speech Acts: An Essay in the Philosophy of Language. Cambridge. ———. 1976. “A Classification of Illocutionary Acts.” Language in Society 5: 1– 23. ———. 1979. Expression and Meaning: Studies in the Theory of Speech Acts. Cambridge. Segal, C. 1971. The Theme of the Mutilation of the Corpse in the Iliad. Mnemosyne Suppl. 17. Leiden. ———. 1982. Dionysian Poetics and Euripides’ Bacchae. Princeton. Seitel, P. 1969. “Proverbs: A Social Use of Metaphors.” In The Wisdom of Many: Essays on the Proverb, ed. W. Weider and A. Dundes, 122– 39. New York. Shapiro, S. 1994. “Learning through Suffering: Human Wisdom in Herodotus.” CJ 89, no. 4:349– 55. ———. 2000. “Proverbial Wisdom in Herodotus.” TAPA 130:89– 118.   η µει ς ιδµεν.” Eranos 71:45. Shimron, B. 1973. “Πρωτος των ———. 1979. “Ein Wortspiel mit homoioi bei Herodot.” RhM, n.s., 122: 131– 33. ———. 1989. Politics and Beliefs in Herodotus. Historia Einzelschriften 58. Stuttgart.

290

Telling Wonders

Sinos, R.H. 1993. “Divine Selection: Epiphany and Politics in Archaic Greece.” In Dougherty and Kurke 1993, 15– 45. Smith, S.M. 1992. “Herodotus’ Use of Animals: A Literary, Ethnographic, and Zoological Study.” Ph.D. diss., Ohio State University. Smyth, H.W. 1956. Greek Grammar. Revised by G.M. Messing. Cambridge, Mass. Sourvinou-Inwood, C. 1991. Reading Greek Culture: Texts and Images, Rituals and Myths. Oxford. Stadter, P. 1992. “Herodotus and the Athenian arche.” ASNP, 3d ser., 22: 781– 809. ———. 1997. “Herodotus and the North Carolina Oral Tradition.” Histos 1 (January). Available at ⬍http://www.dur.ac.uk/histos⬎ (February 1999). Stahl, H.P. 1975. “Learning through Suffering? Croesus’ Conversations in the History of Herodotus.” YCS 24:1– 36. Stein, H. 1877– 1883. Herodot. 5 vol. Berlin.   Stier, H. 1928. “Νοµος βασιλευς.” Philologus 83:225– 58. Strasburger, H. 1955. “Herodot und das perikleische Athen.” Historia 4:1– 25. Strassler, R.B., ed. 1996. The Landmark Thucydides. New York. Svenbro, J. 1976. La parole et le marbre: Aux origines de le po´etique grecque. Lund. ———. 1993. Phrasikleia. An Anthropology of Reading in Ancient Greece. Trans. Janet Lloyd. Ithaca. Originally published as Phrasikleia: Anthropologie de le lecture en gr`ece ancienne. (Paris, 1988). Thomas, R. 1989. Oral Tradition and Written Record in Classical Athens, Cambridge. ———. 1997. “Ethnography, Proof, and Argument in Herodotus’ Histories.” PCPS 43:128– 48. ———. 2000. Herodotus in Context: Ethnography, Science and the Art of Persuasion. Cambridge. Thomson, G., ed. 1938. Aeschylus, Oresteia. 2 vols. Cambridge. Thomson, N. 1996. Herodotus and the Origins of the Political Community: Arion’s Leap. New Haven. Tourraix, A. 1976. “La femme et le pouvoir chez H´erodote.” DHA 2:366– 86. Tyrrell, W.B. 1984. Amazons: A Study in Athenian Mythmaking. Baltimore. Tyrrell, W.B., and F. Brown. 1991. Athenian Myths and Institutions: Words in Action. Oxford. Untersteiner, M. 1967. I Sofisti. 2 vols. Milan. van Dijk, T. 1976. “Pragmatics and Poetics.” In Pragmatics of Language and Literature, ed. T. Van Dijk, 23– 57. Vandiver, E. 1991. Heroes in Herodotus: The Interaction of Myth and History. Frankfurt. Van Groningen, B.A. 1958. La composition litt´eraire archa¨ıque grecque: proc´ed´es et realisations. Amsterdam. van Leyden, V. 1949– 50. “Spatium Historicum: The Historical Past as Viewed by Hecataeus, Herodotus, and Thucydides.” DUJ 11:89– 104. Van Otterlo, W.A.A. 1944. Untersuchungen uber ¨ Begriff, Anwendung und Entstehung der griechischen Ringkomposition. Amsterdam.

Bibliography

291

Van Paassen, C. 1957. The Classical Tradition of Geography. Groningen. Vernant, J. 1988. “City-State Warfare.” In Myth and Society in Ancient Greece, trans. J. Lloyd, 29– 53. New York. ———. 1989a. “At Man’s Table: Hesiod’s Foundation Myth of Sacrifice.” In Detienne and Vernant 1989, 21– 86. ———. 1989b. “Food in the Countries of the Sun.” In Detienne and Vernant 1989, 164– 69. Veyne, P. 1971. Comment on e´ crit l’histoire. Paris. Vidal-Naquet, P. 1981. “Slavery and the Role of Women in Tradition, Myth and Utopia.” Myth, Religion, and Society, ed. R.C. Gordon, 187– 200. Cambridge. Von Bothmer, D. 1957. Amazons in Greek Art. Oxford. von Fritz, K. 1936. “Herodotus and the Growth of Greek Historiography.” TAPA 67:315– 40. Wade-Gery, H.T. 1958. “Thucydides the Son of Melesias.” In Essays in Greek History, 239– 70. Oxford. Walbank, F.W. 1951. “The Problem of Greek Nationality.” Phoenix 5:41– 60. Waters, K.H. 1971. Herodotus on Tyrants and Despots: A Study in Objectivity. Weisbaden. ———. 1985. Herodotus the Historian: His Problems, Method, and Originality. Norman, Okla. West, M.L. 1992. Iambi et Elegi Graeci. Vol. 2. Oxford. ———. 1993. “Simonides Redivivus.” ZPE 98:1– 14. West, S. 1985. “Herodotus’s Epigraphical Interests.” CQ 35:278– 305. ———. 1988. “The Scythian Ultimatum (Herodotus IV 131, 132).” JHS 58: 207– 11. Williams, B. 1993. Shame and Necessity. Berkeley. Wolff, E. 1934. “Das geschichtliche Verstehen in Tacitus Germania.” Hermes 69:121– 66. Wood, H. 1972. The Histories of Herodotus: An Analysis of the Formal Structure. The Hague. Zeitlin, F. 1984. “The Dynamics of Misogyny: Myth and Mythmaking in the Oresteia.” In Women in the Ancient World: The Arethusa Papers, ed. J. Peradotto and J.P. Sullivan, 159– 94. Albany. ———. 1990. “Thebes: Theater of Self and Society in Athenian Drama.” In Nothing to Do with Dionysus? Athenian Drama in Its Social Context, ed. J.J. Winkler and F. Zeitlin, 130– 67. Princeton.

General Index

Names of modern scholars are included only when they appear in the text. Addressee, 28n. 21, 29n. 22, 37. See also Audiences  αδικος, etc., 31n. 25, 71, 85, 147, 182, 186, 187, 188, 189– 90, 195, 196, 211– 12 Advertisements of narratability. See Glosses of celebration; Narratability Adyrmachidae, 82 Aeacidae, 191n. 146 Aegina, 42, 189– 91, 193 Aeginetans as sources, 222n. 220 Aeschylus, 239n. 25, 258 Aesop, 5, 7, 227n. 230 Agathyrsi, 79– 80 Ainos, 5– 8, 13, 50, 180n. 122, 227, 251, 253, 254, 263n. 101, 265, 271 α ι τ ι η, 31, 182, 188, 192, 221, 222, 223, 260, 262n. 98 αιτιος, 31n. 25, 182, 188, 189n. 142, 192, 207n. 182, 211– 12, 222, 224 η.

See Glosses of historie; Hearsay ακο Alcmaeonids, 55n. 37, 222n. 220, 237n. 17, 260– 64 Alexander of Macedon, 29, 39 Alexander of Troy, 142– 44, 185

etc., 173– 74, 178, 197– 98, αληθ ης, 258– 59 “All men,” 100, 169– 70, 171, 226, 228, 229– 30  αλλος, 31, 50

Also (ka ι
), 50, 56, 118, 137 Alterity. See Other and otherness Alternative versions, 35n. 42, 41, 222– 23, 224, 225– 26 Alyattes, tomb of, 104– 5, 106, 235, 237 Amazons, 109, 123– 32, 144, 202n. 169, 212, 220, 233n. 5, 255– 56 Anacharsis, 115– 16, 117, 118– 19, 120 αναγκ-, 213, 246, 257 of the narrator, 34n. 38, 94, 166, 173, 257– 59 Analepsis. See Narratives, analeptic Analogy, 46, 47– 50, 95 explicit, 50– 66, 73, 82, 90 horizontal, 46, 50, 73, 132 implicit, 66– 73, 123, 100– 132, 191n. 146, 219, 238, 255 vertical, 46, 50, 56, 61– 62, 73, 86, 88, 90, 122, 132, 169, 251– 59 See also Comparison; Similarity Anaximander, 85n. 125 ανδραγαθ ι η, 114, 150, 154 ανδρη ι η, 67, 102– 3, 137, 154n. 54 Androphagoi, 80, 108, 136– 37, 170 Animals, 46, 80, 88n. 131, 90nn. 135, 136, 95– 96, 97, 122n. 232, 236, 243– 51 worship of, 93– 95 See also Asses; Camels; Hares; Horses; Lions; Mules 293

294

General Index



ανθρωπ ηιον, etc., 99n. 169, 101, 107 Apis bull, 51n. 25 αποδεκ-, 7, 32, 119n. 221, 182  αρα, 92n. 151, 159, 187 Arabia, 75n. 88, 85, 233n. 4, 235– 36 Arbitration, 7– 8, 166, 217, 222– 31, 260 η,

50, 55, 203, 204, 216– 17, 226 αρχ Athenian, 58, 203 Arete (moral excellence or courage), 43, 44n. 68, 51, 114n. 205, 208, 211, 213. See also

ανδραγαθ ι η; ανδρη ηη Argippaeans, 79, 108, 137 Argives, 225– 29 as sources, 225, 226– 27 Argos, 61, 64, 112, 120n. 225, 225, 229 Arimaspi, 79, 83n. 117 Arion, 252– 55, 256, 259 Aristagoras, 69, 209– 10 Aristides, 43, 261n. 94 Aristotle, 178– 79, 234 Artabanus, 48n. 9, 101n. 171, 122, 146, 213, 245, 257n. 86 Artaxerxes, 203, 205, 226 Artemisia, 41n. 62, 112, 126n. 245, 255– 59, 271 Asses, 55, 64n. 65, 243, 248– 50 Assyria, 86 Assyrians, 8, 10–13, 91n. 143, 138n. 10 Astrabacus, 64n. 65 Astyages, 23– 24 Ate, 184, 200 Athenian demos, 53, 56– 58, 208– 10 Athenians, 38, 42, 57– 59, 86, 92n. 149, 105n. 179, 116– 17, 125, 132, 136n. 3, 189– 94, 197, 212n. 194, 213– 17, 218, 219, 220, 221 as sources, 174, 222n. 220, 223– 24 Athens, 3, 4, 6n. 16, 48n. 9, 52– 59, 63, 68, 90, 109– 10, 124, 172– 75, 178, 181, 189– 94, 198n. 158, 199, 202– 5, 206– 11, 255– 56, 260– 65

Athletics, 49, 61 Audiences encoded in metanarrative, 37– 38, 139, 233 historical, 4, 38, 40, 219, 266, 269– 71 implied, 6, 10n. 25, 14, 16, 19 internal, 4n. 10, 252n. 69, 248, 251, 255, 259, 263 knowledge, expectations, and attitudes of, 38, 74, 98, 101, 102, 137, 147, 161, 174, 198, 251 Autopsy, 35, 86, 143, 164, 241 Babies, 63n. 62 Babylon, 9, 12, 14, 51– 52, 86, 104, 156n. 63, 241 Babylonians, 77, 78, 92n. 151, 98, 167n. 91 “Bad” as label applied to actions and/ or customs, 42, 137, 140, 156, 163, 209 Bald Men. See Argippaeans Balinese, 157– 58 Banquets, 69, 70, 155, 263 Barbarians (non-Greeks) attitudes of, 145– 54, 161– 63 Greek attitudes toward, 8, 12, 69, 79– 80, 81, 87, 93n. 152, 103, 107– 10, 114, 121, 124, 141– 42, 143, 147, 154– 55, 209– 10 as positive paradigms, 8, 12– 13, 106, 136– 37, 141, 142– 44, 212 in relation to Greeks, 8, 18, 31, 69n. 80, 76– 77, 91, 92, 97, 101– 32, 133, 135– 36, 155– 56, 211, 239– 40 Barbarity, 58, 122, 135, 141, 155, 253 Barbaros, 101, 135 Beliefs cultural, 34n. 40, 148, 149, 158– 59, 161

General Index Beliefs (continued) of the narrator (see Glosses, selfreferential, of opinion;

νοµηζειν) religious, 164– 66 Benedict, R., 13, 14 Blame, 16, 42, 218, 224. See also Glosses, referential, evaluations of worth Boas, G., 79 Boats, 90 Assyrian, 1, 8, 10– 14, 15, 26, 233, 241 Boundaries, 85– 86, 103, 204n. 176 violations of, 11n. 26, 48, 49, 50, 131, 135, 200, 201, 250n. 63 See also Rivers Bravery, 43 See also ανδραγαθ ι η; ανδρη ι η; Arete Buto, sanctuary of Apollo and Artemis at, 237– 39 Callatians, 167n. 91, 170, 236n. 16 Callias of Athens, 226, 229n. 237, 260– 61 peace of, 131n. 258, 229n. 237, 261n. 94 Callias of Elis, 72n. 83 Callicles, 171 Cambyses, 51n. 25, 57n. 43, 63n. 61, 65, 166, 168– 72, 229n. 237, 245n. 47 Camels, 244– 47, 249 Candaules, 28n. 20, 40, 42n. 64 Cannibalism, 77, 137, 161– 63, 170, 172 Causality, 30, 172, 183, 188, 190, 192, 243– 44, 245, 246, 249, 251, 253, 269 Celebration, 13, 29n. 22, 43– 44, 145n. 32, 232– 65 Celebratory code. See Codes, celebratory Celebratory glosses. See Glosses, referential, of celebration

295

Character identification of, 39 narrator as, 35 self-referential, 255, 264 Chauvinism. See Ethnocentrism Chemmis, 238– 39, 254 Chians, 195 Choaspes River, 9, 11 Classification, 46, 47– 48 Cleisthenes of Athens, 52– 57, 68, 262 Cleisthenes of Sicyon, 52, 54– 56, 68, 262, 263 Cleomenes, 65, 68, 169n. 95, 177, 209, 225 Climate, 75n. 88, 87– 88 Codes, 10– 11nn. 25, 26, 28, 41, 48, 102, 219 of animals, 252 celebratory, 31, 32, 44, 135, 235, 252, 254, 259 conventional, of wonders, 235 of customs, 40 ethical, 11nn. 27, 28, 30, 31, 70, 87, 208, 215 historical, 31 of historie, 33, 143n. 24, 159 juridical, 31, 70 of kingship (monarchical), 11, 28, 49n. 21, 64, 65, 104, 122, 154n. 54, 171n. 100, 241 mythical, 60 of narration, 33 political, 11n. 27, 44n. 20, 146, 174n. 110, 126 of refutation, 254, 259 of sacrifice, 161 scientific, 87, 244 Scythian, 114– 15 semic (indices), 10n. 25, 12n. 30, 32n. 28, 46, 49, 54, 61– 62, 63, 68– 69, 70n. 81, 122, 155, 254, 257, 264 Spartan, 112n. 198, 114– 15 symbolic, 10n. 25, 11nn. 27, 28, 49, 61– 62, 131, 218, 244, 246, 251 theological, 187, 244

296

General Index

Cognitive statements, 120, 140, 146– 48, 150– 51, 160, 162n. 72 definition of, 34n. 40 Coincidences, 50– 51, 193, 194 Common sense, 12 Comparison, 15– 16, 45– 133, 228, 232 diachronic, 16, 46– 47, 47– 73, 132 synchronic, 16, 46– 47, 73– 133 See also Analogy; Difference; Glosses, referential, of comparison Conclusions, 17, 23, 24, 41 mixed, 27 programmatic, 26, 30, 44, 164, 248 retrospective, 25, 28– 29, 42, 69, 74 summary, 25, 27– 28, 70, 248, 249 types and rhetorical value of, 24– 32 Conjecture, 36, 143 See also Glosses, self-referential, of opinion; συµβα λλεσθαι Connotation. See Codes, semic Constitutional Debate, 50n. 22, 101, 207, 210 Contemporary rhetoric, 174, 204n. 175, 215, 216, 217– 18, 229, 231 Contexts, 45n. 1, 48, 51, 90, 178, 205, 216– 17, 221, 234, 243, 267 Continents, 84– 85 Convictions and indictments, 222– 23, 224, 230 Corcyreans, 221, 224– 25 as sources, 224 Corinthians, 97, 174n. 107, 203n. 170, 223– 24, 258n. 89 as sources, 223 Croesus, 6n. 16, 11, 13, 25, 28, 30, 39, 50n. 24, 80n. 104, 102– 4, 106, 119n. 221, 135, 162n. 72, 182– 85, 188n. 138, 211– 12, 237n. 17, 245, 249, 262, 263 Culture, 40, 136, 168 material, 88– 89, 110, 137

in relation to climate, 87– 88 in relation to history, 106 See also Nomos Cultures hard, 77– 78, 79, 250 primitive, hard and soft. See Primitives soft, 77– 78, 80, 104– 6, 250 Customs. See Codes of customs; Nomoi Cypselus, 63n. 62, 246 Cyrene, 187, 188n. 136 Cyrus, 9– 11, 25, 33n. 32, 39, 41n. 62, 50n. 24, 51n. 27, 63, 102, 103, 130, 188n. 138, 245, 249, 250n. 63 Darius, 11, 63, 69, 88n. 131, 112, 171n. 100, 187, 188n. 138, 192, 202, 205, 233n. 5, 240, 248, 249

24 δε, Deception in Herodotus, 49, 57n. 44, 63, 71, 72, 135n. 2, 147, 148, 210 of Herodotus, 7, 19n. 49 δει , 34 Deioces, 48n. 8, 63– 64 De¨ıphonus, 70 Delos, 201, 204n. 175, 238– 39 earthquake of, 201– 5 Delphi, 6, 7, 53, 104, 124n. 239, 218, 247n. 55. See also Oracles Demaratus, 41n. 62, 64, 67, 111, 112n. 198, 208, 213, 246n. 50, 250n. 61 Demeter, 190, 194n. 151, 239n. 25, 243– 44 Democracy, 52– 53, 56– 59, 207– 11 Demonstratives, 25– 26, 28 Description, 21, 39, 74 nonevaluative, 156– 63 Detienne, M., 173 διαβαι νειν, 131, 214, 250n. 63 Difference, 46– 47, 51, 74– 76, 79, 89,

General Index 92n. 145, 95, 102, 126, 132, 133, 142n. 21, 163, 166, 167, 230, 235 See also Comparison; Other and otherness; Uniqueness Digressions, 242 See also Narratives, inserted δι κη, etc., 70– 73, 80, 130, 136– 37, 143– 44, 189– 90, 199– 200, 211 Dionysus, 55, 92, 119– 21, 146n. 33 Discourse, 14, 17n. 44, 19, 22, 24, 43, 51, 73, 93– 94, 95n. 156, 98, 156, 223– 25, 232, 244– 45, 248– 49, 251, 265, 266 definition of, 20n. 4 Disputes, 217– 31, 260 metanarrative, among sources, 35, 219, 221– 23, 225– 28 narrative, among historical agents, 35, 188, 219– 21, 222– 23, 224– 25 See also Stasis Dissoi Logoi, 169– 70, 228n. 236 Divine action, 71– 73, 121, 181, 195, 230– 31, 243, 253. See also Retribution Divine communication, 24, 71– 73, 181, 194– 206, 231 See also Dreams; Omens; Oracles Divine principle, 165, 169, 185, 243– 44, 247 Divine representations, 166, 238 Divine sphere, 95, 244, 247 δι ζοµαι, ε πιδι ζοµαι, 33, 37n. 55, 242

42n. 63, 54 δοκεω, Donkeys. See Asses Dorians, 55, 56, 203n. 172 Dorieus, 65n. 66, 72n. 83

etc., 28, 63, 207n. 183 δουλοω, Dreams, 23, 41, 49, 196, 197, 246 Dress, 69, 121, 122, 128 ethnographic description of, 81, 92n. 151, 121

δυναστευειν, 53, 209

297

Ecbatana, 64n. 63, 90 Egypt, 51, 75, 84, 86, 92, 103, 104, 142, 143, 168, 185– 86, 236n. 16, 237– 39, 241 Egyptians, 7, 13, 63, 92– 97, 98– 100 as critics, 145n. 31 as ethnographic subjects, 75– 76, 77, 78, 81n. 109, 87– 88, 118, 136n. 3, 141– 44, 167n. 91, 196, 223n. 225 as sources, 143, 239 ε ι κα ζω, 30, 60– 61, 62, 83n. 114, 184 Ephialtes, 17n. 47, 222– 23 Ephorus, 108, 141n. 17 Epitaphios. See Funeral oration ε πιθυµ- (desire), 91, 94, 98, 122, 164– 65, 171n. 100, 184 Eros, 65n. 67, 69 Ethea, 99, 119 in relation to nomoi, 137n. 7 Ethiopians, 75n. 88, 77, 79, 98n. 165, 108n. 186, 145n. 31, 167n. 91 Ethnocentrism, 118, 145– 46, 151, 164, 166, 171, 231 Ethnographic glosses. See Glosses, referential, ethnographic Ethnographic present. See Present tense, ethnographic Ethnography ancient Greek, 6, 107– 8, 141n. 17, 234, 235, 240, 241, 242 in Herodotus, 1– 2, 6– 8, 13– 14, 18, 74– 132, 134, 135– 36, 143, 153– 55, 168, 244, 254n. 75 modern, 12n. 31, 13– 14, 35, 42n. 65, 107n. 182, 141n. 18, 145– 46, 146n. 36, 156– 59, 160, 163n. 76 totalizing statements in, 118n. 220, 137, 138 See also Description ευδαιµον ι η (happiness), 101, 177, 184, 185, 186, 190, 269 instability of, 181– 83, 206, 230 Euenius, 70– 73 Euethie (simplemindedness), 142, 210

298

General Index

Euripides, 120 Evaluation (praise or blame), 30, 43, 119, 134– 35, 158– 59 in the ethnography, 135– 72 in the history, 172, 175– 78, 206– 11, 211– 17, 226, 227, 264 Evaluations. See Glosses, referential, evaluations of worth Evaluations of accuracy. See Glosses, self-referential, evaluations of accuracy Events after 479, 190n. 143, 193– 94, 198n. 158, 201– 5, 213, 219, 221, 229, 267– 71 Herodotus’ explicit references to, 45, 190, 191– 93, 204n. 176, 214, 215 Herodotus’ reticence about, 4, 6 See also Peloponnesian War Evidence, 180– 81, 182, 183, 186, 188, 190, 193, 196, 197, 198– 201, 206, 207, 209, 216, 256 See also Glosses, self-referential, of evidence  ι σκειν, 54, 55, 91n. ε ξευρι σκειν, ευρ 143, 105, 120, 137, 138, 268n. 4 Expansionism. See Imperialism and expansionism Explanation, 23, 43, 72– 73, 232 See also Glosses, referential, explanatory First person. See Person, first; Person, first plural Flashback, 39, 189. See also Narratives, analeptic Focalization, 23n. 12, 34n. 39, 144– 45, 148n. 39, 161 Focalizer, embedded, 23 narrator as primary, 23n. 12 Fornara, C., 3– 5, 6 Frame, metanarrative, 22n. 11, 28n. 20 Freedom, 63, 111– 12, 201, 211, 212,

213, 231, 237, 257– 58, 264– 65 Froidefond, C., 241 Funeral oration, 4, 125n. 241, 203n. 169, 220n. 216 Funeral rituals, 95n. 157, 101, 160– 63, 167– 72 γα ρ, 22, 27, 30, 244 Geertz, C., 157, 171 Gelon, 83n. 114, 218– 19 Geloni, 81, 92n. 146 Generalizations (gnomic statements), 178– 83 from a dream, 196 made by characters, 101n. 171, 122n. 232, 180, 184, 211– 16 of the narrator, 7, 41, 94, 100, 165, 168, 180– 81, 182, 185, 186– 87, 186n. 134, 194– 95, 197– 201, 202, 206– 10, 211n. 193, 212, 217, 226– 31 primary referent of, 227 Geography, 73, 74, 82– 87, 237 Getae, 101n. 172, 137, 164 Glaucus, 188– 89, 190 Glory. See Kleos Glosses, 14, 17, 23, 28 self-referential, 32– 38, 41, 42, 44, 55, 82, 182, 228 of corroboration, 35– 36, 38, 39n. 58, 101, 143, 147 evaluations of accuracy, 35– 36, 226 of evidence, 35– 36, 38, 39, 41, 118, 173, 187, 192– 93, 194– 95, 215 of historie (autopsy, hearsay, etc.), 31– 32, 35, 36nn. 49, 51, 37, 43, 70, 119, 143, 241 of identification, 39, 43, 58, 60, 69, 177– 78, 246, 255 of knowledge, 35– 36, 39, 41, 149 of narration, 30, 32– 34, 37, 38n. 56, 42, 93– 94, 226

General Index of noncorroboration, 35– 36, 226, 239, 240, 259 of not knowing (ignorance), 35– 36, 37, 85n. 123, 221, 226, 227, 261 of opinion, 36– 37, 41, 42, 43, 44, 83n. 114, 138, 149, 166n. 87, 173, 244, 254, 261 of refutation, 36, 41n. 62, 142, 143, 223, 260, 261 of rejection, 36, 141, 142, 143, 260, 261, 264 of source, 34– 35, 39, 83n. 115, 116, 141, 142, 143, 148, 241, 252 of testimony, 38– 39, 57n. 42, 64, 239n. 25, 253, 263 referential, 36, 38– 44, 95n. 156, 228 of anticipation of doom, 42n. 64, 120 of celebration, 36, 43– 44, 74, 75, 133, 235, 237– 38, 239, 241– 42, 247n. 55, 248, 253, 254, 256– 57 of comparison (analogy, similarity, difference or uniqueness), 40– 41, 45, 50, 54, 56, 59, 60, 75– 76, 77n. 95, 79, 80, 81, 82, 83n. 118, 89– 100, 100– 101, 118, 133, 137, 166n. 87, 217, 241, 260 ethnographic, 40, 69n. 80, 90, 102, 146– 47, 153n. 51 evaluations of worth, 36, 42– 43, 69, 76n. 93, 134, 136– 40, 150, 178, 206– 7, 208– 10, 226, 228 explanatory, 27, 36n. 47, 39– 41, 64, 69, 120, 244, 248, 252n. 68, 261 historical, 40, 105 historico-ethnographic, 103, 155 of interpretation, 36, 39n. 58, 41– 42, 53, 118, 132, 136n. 3, 137, 151, 165, 168, 173– 231

299

metalinguistic, 40n. 61, 94– 95 See also Conclusions; Introductions

γνωµη, 50n. 28, 59n. 51, 164, 173– 74, 176, 177– 78, 186, 193, 226 Gnomic statements. See Generalizations Gods. See Beliefs, religious; Divine action; Divine communication; Divine principle; Nomoi, religious Gold, 103, 104, 105, 235, 236 Golden Age, 79 “Good,” 42, 43, 93, 148, 155, 163, 209 γρα ␸ω, 33n. 30, 222 Greatness, 38n. 57, 44, 211, 240n. 33 Greeks defining characteristics of, 18, 105, 120 as sources, 223, 224– 28 Gyges, 28n. 20, 40, 42n. 64, 183 Hares, 22– 23, 34, 250 Hartog, 111, 156, 158, 234 Health, 88n. 131 Hearsay, 35, 156n. 63, 164, 239, 242 Hecataeus, 85n. 12, 222n. 220, 239, 240, 267n. 1 Hegemonie (leadership), 214– 17, 218, 225– 26 Hegesistratus, 60, 66– 68, 70 Helen, 142– 44, 185– 86, 193 Hellespont, 11, 135, 193, 200, 204 Hellman, F., 78 Helms, L. V., 157, 160, 161 Helots, 68n. 75, 70n. 81, 113 Heracles, 65, 66, 84n. 121, 94, 124n. 237, 128, 142, 143, 233n. 5, 239– 40 Hermotimus, 47 Herodotus as real author, 31, 257n. 88, 267– 73 See also Histor; Narrator

300

General Index

Heroes and heroic figures, 52, 55, 60, 65– 66, 67, 124, 176, 254 Heroic past. See Mythical past Hesiod, 5, 165 Hipparchus, 196, 246 Hippoclides, 263– 64 Hippocratics, 37n. 55, 84n. 120, 87, 90, 141n. 17, 159 Histor, 8n. 21, 24, 31– 32, 35, 49, 82n. 113, 141n. 20, 206, 209, 217, 221, 225, 257, 260, 267 audience as, 37 Historie, 6– 8, 31– 32, 96, 98, 132, 141, 146– 48, 166, 181, 182, 201, 221– 22, 224, 225, 227, 230, 254, 265 See also Codes of historie; Glosses, self-referential, of historie Histories Book 2 of, 5, 35n. 45, 96 dating, of, 3n. 6, 5, 267 end of, 15n. 40, 57– 58, 256 genesis of, 2n. 1, 5 message of, 4, 5– 6, 8, 16, 18, 24, 70, 93n. 153, 116, 132, 136, 146, 171, 178, 179– 81, 210, 212, 223, 230– 31, 253 moralism of, 18, 44, 88n. 131, 163, 184– 85, 266 obscurity of, 6, 7, 18 open-endedness of, 15n. 40 purpose of, 3, 4– 8, 12– 13 structure of, 15, 181 unity of, 2n. 1, 15– 16 Homer and Homeric poetry, 4, 14n. 37, 15, 24, 39n. 60, 43– 44, 90, 119, 143, 147n. 38, 165, 166, 176, 178, 186, 247 Iliad, 23n. 12, 107, 211, 218n. 211 Odyssey, 33n. 34, 182, 247 Homoioi, 111 Hoplites, 113, 114n. 205 Horses, 17, 22– 23, 34, 244n. 43, 245, 248– 50 Hubris, 6, 69, 155, 189, 190, 199–

200, 207, 210, 213, 215, 220, 248, 249, 250n. 63 Hyperboreans, 79, 108nn. 186, 187 Hypothetical constructions, 41, 169, 173, 175 Illocutionary acts, 16, 17, 179n. 119 Imperialism and expansionism, 13, 49, 50, 85, 188, 212 Athenian, 3, 57, 124, 156, 174n. 110, 204n. 175, 209– 10, 213– 14, 231 Persian, 11, 102, 122n. 232, 151– 53, 155, 187– 88 Index. See Codes, semic India, 235– 36 Indians, 91n. 143 See also Callatians; Padaeans Inquiry, 19, 188, 189, 230 See also Historie Interpretatio Graeca, 91 Interpretation, 13, 27, 28, 29n. 22, 30, 36, 43, 51, 54, 73, 82, 94, 132, 134– 35, 232, 233, 236, 245, 247, 248 in the ethnography, 168– 71 in the history, 172– 231 See also Glosses, referential, of interpretation Introductions, 17, 23, 24, 32, 41, 237 mixed, 27, 30 programmatic, 26, 29n. 22, 30– 32, 36n. 51, 44, 94, 181, 183, 235, 242 prospective, 26, 29, 30, 42, 160 summary, 26n. 16, 27, 29– 30, 95n. 156, 184, 192, 244n. 44 types and rhetorical value of, 24– 32 Inventions, 92n. 149, 136  ι σκειν See also ε ξευρι σκειν, ευρ Ionia, 50, 188n. 138, 233n. 6 Ionians, 52, 53– 54, 56, 57n. 42, 83n. 114, 88n. 131, 92n. 151, 105n. 179, 106, 111– 12, 136n. 3, 138n. 11, 146n. 33, 195,

General Index 204– 5, 209– 10, 212, 216, 221, 267 as sources, 221, 237 Issedones, 91, 108n. 187, 137, 167n. 91 Ister, 75n. 88, 84 Jacoby, F., 5 Jokes, 69n. 79, 112n. 198 κακα (evils), 195, 202– 5, 206, 211– 12, 214– 15, 225, 226, 227, 228– 29 Kings, 102, 176, 184, 185, 241, 262 as founders, 105, 188 as inquirers, 49, 141, 166, 168, 170, 252n. 69 in relation to citizens, 13, 62, 66, 121– 22, 178 three-king period, 203– 5 Kingship, 9– 13, 49– 50, 60– 66, 73, 132, 136, 153– 54, 196, 207, 237, 245 dual, 62, 64– 65 See also Codes of kingship; Monarchical model; Tyranny Kleos, 6, 65, 176– 77, 211 κως, 41, 86– 87, 195 κου, 41n. 62 Labyrinth, 241– 42 Lacedaemonians. See Spartans Laughter and derision, 55, 70n. 81, 169 of the narrator, 85 Leonidas, 65– 66, 69n. 80, 172, 176– 78, 246– 47 Leotychides, 4n. 10, 51n. 27, 65nn. 66, 68, 188– 89 Libya, 84, 86n. 130, 105n. 179, 156n. 63 Libyans, 79, 80, 92n. 151, 147 Linus, 98– 99, 233n. 4, 238– 39 Lions, 58, 171n. 100, 196, 244– 47, 249, 260

301

Literary theory, 2, 8n. 21, 10n. 25, 12n. 30, 15n. 40, 16– 17, 28n. 21, 42n. 64, 248n. 57 See also Narratology; Speech-act theory Lloyd, A. B., 241 Logios, 35, 117, 143n. 24 Logoi reports of sources, 33, 34, 37, 41, 56n. 42, 141– 42, 166, 175, 222– 24, 225– 30, 237, 259 semiautonomous narratives, 2n. 1, 13, 15nn. 39, 40, 20n. 4, 33, 122, 123, 126, 130, 141, 186n. 133, 187, 193, 234n. 9, 256 Logos, of Herodotus, 33– 34, 44, 94, 135, 173, 182– 83, 188, 191, 196, 201, 206, 218, 242, 252, 253, 257, 258 Lovejoy, A. O., 79 Luxury, 69, 79– 80, 250 Lydia, 235, 236– 37, 246 Lydian appendix, 103– 6 Lydians, 25, 28, 40, 80, 123, 245, 246, 249 as ethnographic subjects, 76n. 93, 77, 78, 92n. 151, 97, 102– 7, 136, 150, 153, 167n. 91 as historical agents, 136 Magi, 77n. 95, 166n. 89, 167 Maps, 84– 85, 209 Marathon, 203, 246n. 50, 247n. 55, 258, 260, 264 Mardonius, 135, 146, 147, 153, 233n. 6 Marketplace, 140, 147, 148 Marriage, 68– 69, 127, 139 irregular, 49, 65, 121 Marriage customs. See Nomoi, sexual Massagetae, 11 as ethnographic subjects, 77, 81, 97– 98, 161– 63, 164, 167n. 91 Mataios, 141, 146

302

General Index

Medes, 9, 39, 92n. 147, 109, 136, 152– 53 Media, 63 Medism, 218, 220– 21, 222– 25 Megistias, 66 Melampus, 60– 62, 64– 66, 70, 71 Menalaus, 144, 186 Metanarrative, 14, 16– 17, 20– 44, 156, 254 definition of, 22n. 11 level, as opposed to narrative, 120, 218, 219 levels, 259 referential, 38– 44, 134 self-referential, 20, 30, 32– 38, 82– 83, 181, 186, 187, 191– 92, 194, 195, 197– 98 Metaphors, 50, 56– 57, 61– 63, 66, 73, 83n. 114, 91, 140, 153, 200, 219, 247, 249, 250– 51, 258 Miltiades, 57, 210 Mimesis within the narrative, 52, 54, 55– 56, 60– 61, 68, 89 narrative as, 17n. 44, 21, 261 Monarchical code. See Codes of kingship Monarchical model, 49, 52– 70, 105, 121– 23, 135, 168, 188, 200, 219 Monarchy. See Kingship Motives of characters, 41n. 62, 42, 172, 173, 175  µουνος, 47n. 6, 59 Mules, 23, 63, 64, 242– 44, 247– 50 Muller, ¨ K., 78 muthos, 84n. 121 Mutilations, 49, 69, 104, 110, 121, 141, 153n. 51, 154, 169n. 94 of self, 67, 154, 167 Mycale, 51nn. 25, 27, 70, 194– 95, 197, 204 Mythical past, 6, 61, 80, 142– 44, 185– 86, 218, 219, 220

Myths, 94, 124– 25, 128, 132, 166n. 89, 238, 239n. 24, 253 of resistance, 120 Nagy, G., 5– 8, 14, 174, 227 Narratability, 44, 54, 59, 97, 98, 211, 232, 252, 254 advertisements of, 31, 43– 44, 47, 67, 232, 242 See also Glosses, referential, of celebration Narrated, the, 38 world of, 32, 43– 44, 218, 221, 228, 232, 247, 260 Narratee, 37n. 54 See also Audiences Narration, 26, 33– 34 as a journey, 33, 242 real world of, 45, 228 time of, 21, 39, 48, 57n. 42, 102, 123, 146, 204, 205– 6, 216– 17, 217, 218, 221 verbs denoting, 33nn. 30– 33 See also Codes of narration; Glosses, self-referential, of narration Narrative, 16– 17, 20– 24, 101, 248 boundaries of, 45 definition of, 21 directions for reading, 27, 35, 38, 51, 58, 120, 232 main, 38, 70 “pure,” nonnarrated, or minimally narrated, 16n. 43, 17, 21, 23, 41, 159, 244 recipient of (see Audiences) summary, 27 Narrative levels, 20, 23, 32, 35, 38, 120, 145, 217– 18, 255 Narratives, 15, 33, 39, 40, 180, 225, 261, 263 anachronic, 21n. 6, 252n. 68 analeptic, 21n. 6, 54, 59, 62n. 59, 67 connections among, 21, 46, 51, 54, 252

General Index goal of, 4 historico-ethnographic, 105, 118, 123 inserted, 39, 59 personal (of historie), 35 proleptic, 21n. 6, 59, 62n. 59, 67 See also Logoi Narratology, 19n. 49, 20, 20n. 4, 21nn. 7, 9, 23n. 12, 27n. 17, 37n. 55, 41n. 62, 45n. 1, 241 Narrator, 8n. 21, 14, 16– 17, 23n. 12, 30, 32, 33– 34, 37, 38, 47, 73, 98, 102, 119n. 221, 132, 134, 135, 159– 60, 166, 169, 180– 81, 198, 241– 42, 246, 255– 59, 264– 65 as adviser, 6, 16, 17, 179– 81 ambivalence of, 93 apologies of, 83n. 115, 94, 166n. 89 corrections of, 69, 81, 143, 161 invisible, 23, 218, 221, 225 omniscient, 42n. 63 as prophet, 50, 205– 6 in relation to the speakers, 187, 207 as warner, 6, 16, 179, 266 See also Beliefs; Character; Focalizer; Histor; Laughter and derision; Obligation of the narrator or his logos; Polemics; Relativism; Subjectivity Negation, 41, 74, 93, 129, 146– 49, 150, 151, 155, 159, 161, 162, 188n. 139, 245, 248 in a gloss of narration, 32, 38n. 56, 93– 94, 198 Nemesis, 30, 184 Nile, 75– 76, 83– 84, 237, 239, 241 Nitocris, 9, 12, 51– 52, 241 Nomads, 80, 81, 107, 114, 250 νοµι ζειν, 37, 164 Nomoi (customs), 7, 54n. 35, 55, 88, 99, 142, 149, 229, 232, 244 borrowing of, 80n. 105, 92, 117, 152– 53 dietary, 77, 161, 163

303

Greek, 76, 91, 96– 97, 145– 46, 147n. 37 medical, 139– 40 religious, 77, 92– 95, 120, 137, 139, 149, 168, 169 secular, 119 sexual, 77, 79, 81, 82, 92n. 148, 98, 138– 39, 155n. 59 value of, 168– 72 See also Funeral rituals; Sacrifice Nomos, 85, 88– 100, 111, 120, 168– 71, 208, 236, 252, 253, 259 in relation to phusis, 75, 99– 100, 168– 69 Non-Greeks. See Barbarians Obligation of the narrator or his logos, 34, 94, 174, 242, 258– 59. See also αναγκOcean, 84, 237, 239 Oikoumene, 78, 85– 86, 123 Omens, 22– 23, 41, 175, 195, 201– 6, 245, 247, 249, 251  οψις. See Autopsy; Glosses, selfreferential, of historie Oracles, 41, 46n. 5, 49, 70– 73, 165, 196, 197– 201, 247n. 55 Delphic, 55, 58, 59, 62, 80n. 104, 175– 76, 183– 84, 188, 189, 195n. 153, 197, 198n. 158, 199, 246, 249 Orality in relation to the Histories, 14– 15, 33 Orientalism, 136, 153 ! οστις, 37 Otanes, 49– 50, 64n. 63, 65n. 67, 130n. 235 Other and otherness, 18, 91, 96, 104, 107, 110, 118, 123, 130, 133, 156, 234n. 9, 256 Padaeans, 162– 63, 167n. 91 Panhellenism, 174, 178, 215, 267n. 2 Paradigms, 168, 180, 186, 188, 195, 200, 204, 206, 222 barbarians as (see Barbarians)

304

General Index

Paradoxes, 13, 42, 53n. 34, 65n. 69, 122, 166, 210, 236, 256 Past, 39, 216, 218 tense, 21, 35, 178, 242n. 39 Patterns, 48, 50n. 22, 63, 68, 86, 112, 133, 143, 235. See also Analogy Pausanias, 68– 70, 90, 115, 121– 22, 135, 177, 215, 246n. 50 Peloponnesian War, 3, 116, 146, 193, 201, 203– 5, 213, 220, 221, 270 Performance, 5, 14– 17, 28, 32, 260, 266 Pericles, 58, 146, 202n. 169, 203n. 170, 219, 246n. 48, 262, 264, 267n. 2 Persian army, 11, 82n. 111, 131n. 258, 152, 244– 45 Persians, 1, 13, 25, 28, 29, 66, 69, 80n. 104, 101, 102, 123, 137, 153, 154, 175, 177, 187, 188, 197, 200– 201, 209, 217, 225, 229, 243– 44, 247– 50, 260, 262 as critics, 146– 47, 148 in an ethnographic context, 77, 78, 80n. 104, 85, 90, 97, 140, 149– 56, 167 as sources, 35 Person first, 17, 26, 34, 36n. 51, 149, 241– 42 first plural, 34n. 35, 37n. 54, 241– 42 second, 37, 119, 139 third imperative, 119 Phatic function, 28 Pheretime, 187– 88 Phoenicians, 35, 92n. 151, 98, 147 Phthon- (envy), 79, 174, 271 divine, 184– 85, 187 Phusis, 76, 83, 88n. 131, 95, 99– 100, 142, 143, 169, 236, 247n. 54, 251n. 67

summary of use of in Herodotus, 99n. 169 Pindar, 5– 6, 7, 39n. 58, 44, 72, 170, 271 Pisistratids, 67, 196, 260 Pisistratus, 47, 57, 63, 64, 207, 250n. 61 Plataea, 51n. 25, 59, 60, 61, 66– 70, 174n. 107, 177, 194– 95, 220 Plato, 171, 252 Plutarch, 7, 13, 143, 155, 175, 263, 264, 271 Polarities, 75, 76, 124, 132 Polemics of ethnographic subjects, 145– 46, 149, 151 of the narrator, 35n. 45, 75n. 87, 84, 141, 146– 47, 237, 239, 241 Praise of referent, 218, 222, 264 (see also Glosses, referential, evaluations of worth) of subject (see Celebration; Glosses, referential, of celebration) Present of narration. See Narration, time of Present tense ethnographic, 21, 159 gnomic, 21n. 9, 178, 183 “Primitive opponent,” 49, 78, 250 Primitives, 76– 77, 95, 236n. 16 hard, 79– 80, 107– 8 soft, 79– 80, 107n. 185 Primitivity, 12, 126n. 245

Prodigy, 120. See also τερας Program/Programmatic statements. See Conclusions, programmatic; Glosses, self-referential, of narration; Introductions, programmatic Prolepsis. See Narratives, proleptic Prophecy. See Divine communication; Dreams; Narrator as prophet; Omens

General Index Prospective introductions. See Introductions, prospective Protagoras, 165, 166, 170n. 98, 207, 268– 69  πρωτος, 36n. 49 Proust, M., 15n. 40 Psammetichus, 84n. 121, 141 Puns, 69, 119. See also Jokes Pyramids, 241 Quarrels. See Disputes Queens, 9, 12, 51– 52 Questions, 29, 41, 142n. 21 Raaflaub, K., 6 Redfield, J., 77, 80 Relativism, 169– 71 cognitive, 163– 66 cultural, 167– 72 of the narrator, 18, 91, 94n. 154, 137, 138, 161, 230, 268 Requirements. See Obligation of the narrator or his logos Retribution divine, 6, 181, 183– 94, 206, 212, 223 human, 186– 88, 189– 91, 192– 93 See also τιµωρι η; τι σις Retrospective sentences. See Conclusions, retrospective Rhampsinitus, 63n. 62 Rhodopis, 26, 142n. 21 Rivers, 9– 12, 50, 75, 83– 85, 86, 88n. 131, 103, 131 Romm, J., 79 Rosaldo, R., 145 Rosellini, M., 77 Sacrifice, 77, 91, 135n. 1, 149, 154 human, 142, 144, 154, 157– 58, 159– 63, 165 Sa¨ıd, S., 77 Sais, 35n. 45 Salamis, 43, 51nn. 25, 27, 57n. 44,

305

173, 174, 199– 201, 223– 24, 258, 264 Salmoxis, 137 Sameness, 82, 96, 100, 107, 117, 123– 32, 133, 239, 256 Samian War, 202n. 169, 220 Saying, verbs of, 52nn. 30, 32 Scyles, 118, 119– 23 Scyllias, 254– 55 Scythia, 75, 86, 90, 137, 233n. 4, 239– 40, 242– 43 Scythians, 7, 13, 79, 123– 32, 137, 138, 154n. 58, 188n. 138, 248– 50 as critics, 120– 21, 126n. 33, 233n. 5 as ethnographic subjects, 75, 77, 79, 80, 81, 83n. 114, 91n. 143, 107– 23, 159– 60, 167n. 91, 212 Second person. See Person, second Seers, 59– 62, 65– 66, 66– 73 Self-referential characters. See Character, self-referential Self-referential metanarrative. See Metanarrative, self-referential Ships, 107, 126n. 245, 204, 256. See also Boats Similarity, 76, 82– 87, 88– 93, 97, 98, 100, 133, 235, 238, 239 See also Analogy; Glosses, referential, of comparison Slavery and enslavement, 102, 103, 111– 12, 153, 187, 200– 201,

207. See also δουλοω Solon, 4n. 10, 47, 92n. 149, 101n. 171, 117, 162n. 72, 184– 85, 187, 263 Sophie (wisdom, intelligence, cleverness), 10, 12, 32n. 28, 33n. 32, 51, 63, 71, 105, 115– 16, 117, 118, 138, 139, 185, 210 Sophists, 99, 165, 166, 169– 71, 198– 99, 207 Sophrosune, 12, 115– 16, 117, 118, 131

306

General Index

Sources as characters, 35 ethnographic subjects as, 148 historical agents as, 223– 24, 228 See also Glosses, self-referential, of source See also under specific ethnographic groups Sparta, 3, 48n. 9, 59– 70, 109– 10, 112– 16, 176– 78, 198n. 158, 209 Spartans, 38, 53, 121, 125, 135, 146, 189, 192– 93, 202n. 169, 208, 212– 13, 218, 221n. 218, 225, 227, 233n. 5, 245– 46 as ethnographic subjects, 76, 96– 97 as sources, 114 Spartiates, 113 Speakers, 48, 49– 50, 180, 183– 85, 187, 207 as narrators, 4, 188 Speaking in relation to the Histories. See Orality in relation to the Histories Speech act, 5, 14, 19, 32, 196 Speech-act theory, 16nn. 41, 42, 179, 222 Speeches, 10, 17n. 44, 177, 180, 182n. 124, 224 Sperthias and Boulis, 43, 147n. 37, 175, 192– 93, 246 Stasis, 53, 130, 214– 17, 258, 270 Sthenelaidas, 115n. 211, 174n. 109, 215, 220 Story, 20– 21, 23, 27, 32, 41, 44, 51, 60, 241 Subcodes. See Codes Subjectivity, 166, 170, 171, 181, 225– 30 of the audience, 34 of characters, 35, 207, 145– 46, 213 of ethnographic subjects, 145– 46 of the narrator, 94, 138, 247 of sources, 35, 221– 22, 225, 228 συµβα λλειν, 45n. 1, 83– 84, 85

συµβα λλεσθαι, 83n. 116, 126 Summarization, 23, 24, 25, 27, 30, 31, 32 element of, 25, 26, 28– 29, 41, 235 Summary conclusions. See Conclusions, summary Summary introductions. See Introductions, summary Summary narrative. See Narrative, summary Superlatives, 44n. 70, 47, 67 Symbolic actions, 111, 135 Symbolism, 52, 96, 212, 217, 221, 249– 50, 252, 255, 264 Symbols, 11n. 27, 122, 132, 155, 168, 204n. 176, 245, 256 See also Codes, symbolic Tellus of Athens, 13

τερας, 22, 23, 27, 95, 195– 96, 202, 243n. 41, 247 Text definition of, 20n. 4 types of, 19 Thebans, 189, 193n. 149, 229 Themistocles, 43, 48n. 8, 51n. 27, 57, 67, 180, 187, 207n. 183, 210, 221n. 218 Theognis, 5 Thermopylae, 112n. 198, 145n. 32, 172, 175– 78, 213, 222– 23, 233n. 5, 246– 47, 258 Third person imperative. See Person, third imperative Thoma, 44n. 70, 47n. 6, 177 See also Wonder; Wonders Thracians, 154n. 58 as ethnographic subjects, 74, 77, 79, 81n. 109, 96, 138n. 11, 160– 61, 162, 163, 164, 167n. 91 Thucydides, 17n. 45, 18n. 49, 23n. 12, 44, 130n. 255, 136, 146, 201, 208n. 184, 213, 215, 216, 220, 221, 261, 267n. 1

General Index Thurii, 267– 71 τιµωρι η, 184, 186, 187– 88, 189, 195. See also Retribution; τι σις τις, 37, 173, 174, 184 Tisamenus, 59– 62, 64– 66, 68, 72 τι σις, 183, 188n. 139, 189, 191. See also Retribution; τιµωρι η Tragedy, 48n. 9, 229 Translation, of culture, 82, 91, 114 linguistic, 40n. 61 See also Glosses, referential, metalingustic Troy, 80, 142– 44, 185, 190, 193, 195, 202n. 169 Truth, 35n. 46
See also αληθη ς Tyranny, 6, 13n. 32, 49, 56, 58, 62n. 58, 63, 64, 65n. 67, 69, 188n. 136, 196, 206, 209, 245 See also Kingship Tyrant city, 58, 174, 257, 264 Tyrants, 53, 54– 55, 56, 257, 262, 264 Uniqueness, 48, 59, 74, 75– 76, 77n. 95 War, 108n. 187, 110, 114, 125, 128, 129, 130– 31, 146, 186, 187, 202– 5, 208– 9, 211– 17, 218, 220, 229, 231, 246, 257

307

Women, 65n. 67, 102, 110, 112, 125, 129– 30 in ethnographic descriptions, 74, 78, 79, 81, 104, 129, 132, 139, 150, 160– 61 See also Nomoi, sexual Writing in relation to the Histories, 14, 33 Wonder metanarrative, 19, 26, 31, 44, 98, 177, 232– 65 narrative, 233n. 5, 243n. 41, 246, 247 Wonders ethnographic, 9, 13– 14, 75, 83n. 118, 95, 98– 99, 133, 233, 234– 42 historical, 71, 177– 78, 233, 243– 65 Xanthippus, 58, 262 Xenie, 142– 44 Xenophanes, 166 Xerxes, 11, 22– 24, 39, 41n. 62, 48n. 9, 51, 57, 68, 102, 111, 114, 122, 130, 135, 146, 152n. 47, 171n. 100, 172, 173, 174, 185, 188n. 138, 200, 202, 212n. 194, 213, 221, 224, 226, 233n. 5, 244, 249, 257, 271n. 14

Index of Passages

Aelian Varia Historia 4.1: 161n. 71 Aeschylus Agamemnon 717– 36: 245n. 47 766: 200n. 163; 1231– 32: 125n. 242 Choephorae 161– 63: 110n. 191 Eumenides 625– 28: 125n. 242 681– 706: 109, 125 700– 703: 109– 10 Persians: 133 Prometheus Bound 709– 14: 109n. 188 Seven against Thebes 727– 33: 229n. 237 Supplices 234– 37: 125n. 241 277– 90: 125n. 241 Aesop 266 Perry: 227n. 230 Anacreon frag 11b Page (⫽PMG 356): 114n. 208 Anaxagoras DK 59 B21a: 84n. 120 Anaxandrides frag. 39 Edmunds: 93n. 152 Antiphon DK 87 B44, frag. 2: 99n. 168, 231n. 241 Apollodorus 1.9.1– 12: 60n. 53 2.5.9: 124n. 237, 128n. 252 Aristophanes Acharnians 702– 3: 109n. 188 Ecclesiazusae 613– 35: 79n. 103, 139

Knights: 210 1111– 12: 58n. 49 Lysistrata 672– 80: 124n. 239, 125n. 241, 255n. 81 Thesmophoriazusae 1070– 1175: 109n. 188 Aristophanes the Boeotian FGrHist 379 F 5: 271n. 13 Aristotle Athenaion Politeia 22.1: 53n. 34 Metaphysics 1.2.8– 9: 234n. 8 Politics 2.1.4– 18: 79n. 103 2.5: 268n. 4, 269n. 7 2.7.6: 117n. 216 Rhetoric 1.7.34: 219n. 212 2.21.2: 178– 79 3.9.2: 272n. 16 3.10.7: 219n. 212 Bacchylides frag. 4.50– 51 SM: 60n. 53 Charon of Lampsacus FGrHist 262 F 1: 251n. 66 Critias DK 88 B32– 37: 114n. 208 frag. 6 West: 114n. 208 Demosthenes Epitaphios 60.4– 8: 125n. 241, 132n. 259 Diodorus 2.45.1– 3: 128n. 251 3.53.1– 2: 128n. 251 4.68.4– 5: 60n. 53 12.4: 131n. 258 309

310

Index of Passages

Diodorus (continued) 12.9– 11: 267n. 2 12.10– 12: 269n. 8, 270n. 9 12.35: 270n. 11 17.77.1– 3: 123n. 236 Diogenes Laertius 1.33: 124n. 240 Dissoi Logoi DK 90 2.14: 163n. 74 2.15: 155n. 59 2.18: 170, 228n. 236 2.26: 170n. 98, 228n. 236 Diyllus FGrHist 73 F 3: 15n. 38, 271n. 13 Ephorus FGrHist 70 F 42: 80n. 106, 108, 141n. 17 Euripides Andromache 448: 228n. 234 Bacchae: 120 Heraclides 215– 17: 124n. 237 Hercules Furens 408– 18: 124n. 237 Hippolytus 10: 124n. 237 305– 9: 124n. 237, 128n. 252 Ion 1144– 45: 124n. 237 Supplices 410– 25: 210n. 191 432– 41: 210n. 191 Eusebius Chronica Olympiaca 83.4 (83.3): 15n. 38, 271n. 13 Hecataeus FGrHist 1 F 185: 80n. 105 302: 237n. 19 305: 239n. 127 Hellanicus FGrHist 4 F 106: 124n. 237 323a F 16, F 17: 125n. 243 Heraclitus DK 22 B114: 169n. 97 Herodotus Book 1 First sentence: 13, 14nn. 35, 37, 30– 32, 43, 44, 47, 177 1.1: 31n. 25

1.4.4: 85 5.3– 4: 181– 83 5.3: 31n. 25, 33nn. 29, 31, 48, 61n. 55, 212 5.4: 179 6: 39n. 60 6.2: 103 7.1: 26n. 16, 28n. 20 8– 13: 63, 65n. 67 8.2: 42n. 64 10.3: 40, 76n. 93 12.2: 39n. 58 13.2: 184 14.1: 28n. 20, 215 14.3: 237n. 17 19.1– 21.1: 253n. 73 20: 36n. 49 23– 24: 21n. 6, 252– 55 23: 39n. 60 23.1: 34n .39 24.5: 10n. 10 24.8: 39n. 58 25.2: 47n. 6 26.3: 188n. 138 27.4: 28 29– 33: 117n. 217 29.1: 103, 119n. 221 30– 33: 263n. 99 30.1: 119n. 221 30.3– 5: 13 30.4– 5: 4n. 10 32: 184– 85 32.1: 182n. 124 32.2– 9: 162n. 72 34.1: 30, 184– 88 35.2: 92n. 145 46– 49: 197 46: 212n. 194 46.1: 188n. 138 50.1– 3: 237n. 17 51.1– 3: 237n. 17 51.5: 237n. 17 53.3: 197 55.2: 63n. 62, 80n. 104, 249n. 60 57.3: 81n. 107

Index of Passages 59– 63: 63 59.1: 207n. 183 59.4: 63, 250n. 61 59.5– 6: 210 59.6: 207n. 183 60.3: 12, 47, 55, 210 61.1: 65n. 67 64.2: 66 64.3: 262n. 98 66– 68: 212n. 197 66.4: 38 70.2– 3: 222n. 120 71: 155 71.1: 188n. 138 71.2– 4: 250 71.2: 12 71.4: 80n. 104 72: 103 73– 75.1: 188n. 138 73: 110n. 193 73.1: 11n. 26 74.1: 40n. 61 74.6: 92n. 145, 97 75: 11, 103 75.6: 41n. 62, 46n. 22 79.3: 103 80: 245, 249, 251n. 66 82– 83: 213n. 199 84.3: 246 85.1: 29n. 23 86– 91: 104 86.2: 41n. 62, 154n. 56 86.5: 50n. 24 87.3: 183n. 126 87.3– 4: 211– 12 89.1: 28n. 19 89.2: 99n. 169, 155, 249 91: 184, 212 91.4: 183n. 126 91.5– 6: 63n. 62, 249n. 60 92– 94: 25, 103– 6, 149n. 41 92.1: 27, 50, 237n. 17 92.4: 26 93.1: 235 94.1: 92n. 145, 235n. 11 94.1– 7: 28

311

94.1– 2: 92n. 151 94.2: 150n. 44 94.7: 25, 237n. 18 95.1: 23, 25, 33nn. 30, 32, 33 95.2: 63 96.1: 48n. 8, 63 96.2: 65n. 67 98.5: 90 103.1: 51n. 26 103.3– 106.1: 212n. 196 105.4: 37n. 54 106.2: 16n. 40 107– 8.3: 23 117.2: 33n. 32 125– 26: 155n. 60 131– 39: 149– 56 131.1: 75n. 87, 164, 166n. 89 132.2: 91 136.1: 91 136.2: 250n. 63 137.1: 42, 113, 140 138: 140 139: 61n. 55 140: 154n. 53, 167 140.2: 77n. 95 140.3: 33nn. 29, 30, 164n. 79 141: 251n. 67 142: 88n. 131 143.2: 88n. 131, 138n. 11 143.3: 57n. 42 147.2: 47n. 6 152: 213n. 199 153: 97 153.1: 227n. 232 155.4: 102n. 174 164: 204n. 176 167.1: 243n. 42 168: 204n. 176 169.1: 88n. 131, 208n. 185, 212 169.2: 50 170: 204n. 176, 212 171.4: 92n. 151 172.1: 77n. 95, 92n. 147, 148n. 39 173.4: 77n. 95 177: 211

312

Index of Passages

Herodotus Book 1 (continued) 178– 87: 9 179.1: 34 182.1– 2: 98n. 164 182.3: 98n. 165 183.1– 3: 156n. 63 183.3: 11, 42 184: 16n. 40, 33n. 29 184– 85.1: 51– 52 185.3: 232n. 2, 241 187: 11 188– 89: 9 189: 153n. 51 189.1: 249n. 59 189.2: 250n. 63 190– 92.1: 10 191.3: 51n. 27 192: 10, 17 193: 10 193.1– 2: 86 193.1: 16 193.3: 140n. 16 193.4: 38n. 56 193.5: 91n. 143 194: 1, 9– 14, 233, 250n. 61 194.1, 9, 13– 14, 26 194.2: 90 194.3: 241 195.1: 89 196: 139, 140 196.1– 4: 104 196.1: 10, 26, 37, 98n. 165 196.5: 104 197: 10 198: 98n. 164, 167n. 91 199: 138– 39 199.1: 162, 171 199.4: 37n. 55 199.5: 98n. 165 200: 10 201: 81 202.2: 91n. 143 203.2: 77, 136n. 5 207.2: 183 207.5: 102n. 174

207.6– 7: 155n. 59 208: 11 211: 155n. 59 212– 14: 102n. 174 215.1: 81 215.1– 2: 250n. 62 215.4: 250n. 62 216.1: 81, 98n. 164 216.2– 3: 161– 63 216.2: 167n. 91 Book 2: 13– 14, 57n. 45 2.2: 141 2.3.2: 94, 164– 65, 230, 259n. 91 2.4.1: 143n. 24 2.5: 84n. 121 3.1: 82n. 113 3.2: 33n. 30, 34n. 38, 37 4.1: 92n. 149 4.2: 92n. 149 5.1: 60– 61 5.2: 37n. 55 6– 11: 241n. 35 7.1: 90n. 139 10.1– 3: 75n. 89, 83n. 114, 84 11.3: 68n. 62 11.4: 41n. 62 12.2: 75n. 89 13.2– 14.1: 145n. 31 14.2: 75n. 89 15– 16: 84n. 121 15– 18: 241n. 35 15.2: 41n. 62 16: 86 16.1: 85n. 123 17.1: 86 17.2: 85n. 123 18.3: 85n. 123 19.2– 3: 75n. 89 20– 27: 83n. 118 20.1: 33n. 32, 84n. 121, 237 21: 84n. 121, 237, 259n. 93 22.1: 33n. 32 22.2: 41n. 62 22.3: 88n. 131 22.4: 36n. 47

Index of Passages 23: 36n. 47, 54n. 35, 84nn. 121, 122, 237 24.1: 75n. 89 25: 92n. 148 25.1: 83n. 118, 85n. 123 27: 85n. 123 28.2: 35n. 45 29.2: 90 29.3: 84n. 119 29.5: 37n. 55 30.1: 37n. 55 31: 37n. 54, 85n. 123 33.2: 84n. 120 34: 84n. 120 35: 96, 233 35.1: 33n. 30, 75, 103, 235n. 10 35.2– 36.4: 76, 87– 88 35.2: 148n. 39 36.2: 95 37.3: 147n. 37 37.4– 5: 147n. 37 40.1: 81n. 109 40.2: 85n. 123 41.3: 118n. 220 42.1: 81n. 109 42.3– 4: 94, 166n. 89 42.4: 81n. 107 43.2: 92n. 149 43.4– 44: 82n. 113 44.1: 35n. 45 45: 84n. 121 45.1– 3: 142 45.1: 210 45.2: 41n. 62, 99n. 169, 147n. 37 45.3: 34n. 36, 41n. 62, 166n. 89 46.1– 4: 93– 95 46.1: 147n. 37 47.2: 94n. 154 49: 92n. 149 49.2: 92n. 149, 120 50.1– 51.1: 92n. 151 50.1: 92n. 149 50.3: 92n. 151 52.2: 92n. 149 53: 165

313

53.5: 85n. 123 54– 57: 92n. 149 54: 147n. 37 56: 147n. 37 57.2: 41n. 62 58: 92n. 149 64.1: 196 65– 76: 95 65.2: 34n. 38, 94n. 154, 259n. 91 66.2: 95, 236n. 16 66.3: 95, 247n. 56 67.2: 89n. 136 68.2– 3: 89n. 135 69.3: 83n. 114 70.1: 33n. 30 73.2: 89n. 136 76.1: 89n. 135, 118n. 220 76.3: 89n. 135 77.1: 117n. 217, 142, 143n. 24 77.3: 140n. 16, 179 79.1– 2: 98– 100, 238 79.1: 147n. 3 80: 96n. 161 81.2: 92n. 150 82: 92n. 149 82.2: 153n. 48 91.1: 118n. 220 91.2: 96 91.4: 92n. 148 92.1: 54n. 35 92.2– 4: 89 92.2: 75n. 90 96.1: 89 97.2: 37n. 55 99.1: 33n. 31, 36n. 51 100.2: 51 101.1: 44n. 70 102.5: 102n. 174 104.1: 92n. 146 104.2: 75n. 88 104.4: 52n. 30 105: 81n. 107 108: 92n. 146, 250n. 62 109: 92n. 149 109.3: 92n. 151

314

Index of Passages

Herodotus Book 2 (continued) 110.1– 3: 51n. 26 112– 20: 142– 44, 185– 86 115.3: 33n. 32 117: 33n. 31, 164n. 79 118.1: 143 120.5: 185, 186, 193, 195, 200 121: 63n. 62 121δ: 250n. 61 123.1: 33n. 30, 34n. 38, 37, 228n. 233 123.1– 2: 92n. 149 123.3: 33n. 30, 39n. 58, 239n. 25 124.1: 51n. 26 125.7: 41n. 62 127.1: 51n. 26 129: 186n. 133 129.3– 130.1: 95n. 157 131.3: 39n. 58 133: 186n. 133 135: 92n. 148 135.3: 37n. 54 135.6: 26, 33n. 30 143– 46: 66n. 89 143: 35n. 45 145.3: 34n. 39 146.1: 32n. 28 147.2: 88n. 131, 136n. 3 148– 49: 241– 42 148.1: 44n. 70 150.2– 4: 98n. 166 154.4: 37n. 54 155– 56: 238– 39 156.2: 254n. 75 157: 211 160: 92n. 149 160.2– 4: 146n. 33 161.3, 42n. 64 162.1– 2: 63 164– 68: 268 167.1– 2: 97 170: 90n. 139 171.2– 3: 92n. 149 172.2: 63

175: 241n. 34 177.2: 92n. 149 Book 3 3.3: 169n. 94 6.2: 29n. 22, 37n. 55 8.1: 40n. 61 10.3: 243n. 41 11.3: 236n. 16 12: 83n. 118 12.1: 37n. 55, 236n. 16 16: 57n. 43, 154n. 54, 169n. 94 20.2: 77n. 95 22.3– 4: 145n. 31 23.2: 233n. 5 24: 167n. 91 24.2: 98n. 165 27– 29: 57n. 43, 166n. 87 27.3: 154n. 54 29: 154n. 54 30: 154n. 54 30.1: 169n. 94 31: 155n. 59 31.1: 154n. 54 31.4: 54n. 35 32: 154n. 54, 245n. 47, 250n. 67 32.4: 52n. 30 33: 169n. 94, 186n. 134 34.1– 3: 154n. 58 34.1: 169n. 94 35: 153n. 51, 154n. 54 35.5: 169n. 94 36: 153n. 51 37: 57n. 43 37.1: 169n. 94 37.2– 3: 89 37.2: 38n. 56 38: 118, 168– 72, 236 38.1: 229– 30 38.1– 2: 55 38.2: 36n. 51 38.4: 7, 39n. 58, 167n. 91, 230 39– 60: 21n. 6 40.2: 185n. 132 43.1: 185n. 132 45.3: 36n. 47

Index of Passages 46: 115n. 211 47.2: 232n. 2 50.1: 65n. 67 52.4– 5: 174n. 108 53.4: 65n. 67 55.1: 47n. 6, 112n. 198 55.2: 35n. 45 56: 224n. 226 57.4: 197 59– 63: 63 60.1: 33n. 30 64.3: 51n. 25, 57n. 43, 166n. 87, 186n. 134 65.3: 99n. 169 66.2: 57n. 43 68.1: 29n. 23 69.5: 154n. 54 70: 63n. 61 75.1– 2: 259n. 92 79.1: 154n. 54 79.3: 155n. 59 80– 83: 207 80: 50 80.1: 38n. 56, 101, 233n. 6 80.2: 200n. 164 80.4: 200 80.5: 55n. 37, 65n. 67 80.6: 207n. 182 81.1: 200n. 164 81.2: 210 82– 83.1: 63 83.1: 207n. 182 85– 87: 63 86: 153n. 48 95.2: 33n. 30 98.2: 85n. 123 98.3: 81n. 107 98.4: 91n. 143 99: 162– 63, 167n. 90 100: 89n. 136, 167 101.1: 77, 101 102.1: 92n. 145 102.2– 105: 235n. 12 102.2: 89n. 136 103: 38n. 56 104.2– 3: 75n. 88

315

106.1: 86– 87, 88n. 131, 195, 236n. 15 106.2: 89n. 135 107.1: 75n. 88 108: 95n. 158, 195, 236n. 15, 243 108.2: 41n. 62, 185 108.4: 245n. 46 111.1: 34n. 39, 36n. 51, 236n. 13 111.2: 233n. 4 112: 140n. 16, 233n. 4, 236n. 13 113.1: 75n. 88, 140, 232n. 2, 236n. 13 114: 235n. 12 116.1: 235n. 12 116.2: 83n. 117, 99n. 169 116.3: 37n. 54 118.1: 200n. 164 118.2: 154n. 54 123.2– 3: 92n. 150 125.3: 154n. 54 126.2: 200n. 164 127: 153n. 51 127.3: 200n. 164 134.5: 153n. 48 139.2: 63n. 61 143.2: 136.3 151.2: 249n. 60 152.1: 51n. 27 153: 249n. 60 154.2: 154n. 55 Book 4 1.1: 188n. 136 1.1– 2: 212n. 196 1.3: 110, 113 3.3: 113n. 200 4.2: 110, 113 4.3– 4: 110– 11 5– 10: 113n. 202 5– 12: 116n. 214 8– 10: 240n. 30 8.2: 84n. 122 9.1: 33n. 33 9.2: 233n. 5 10.3– 11.1: 240n. 30

316

Index of Passages

Herodotus Book 4 (continued) 11: 113 11.4: 39n. 58 12.1: 39n. 58 17: 82n. 111 18.3: 80 20.1: 113 22.1: 81 23: 80 23.2– 5: 137n. 7 23.2: 81n. 107 23.3: 89n. 135 23.5: 108n. 186 25.1: 36n. 50, 83n. 117 26: 167n. 91 26.2: 91, 108n. 186 28.1: 37n. 55 28.2: 75n. 88 28.3: 243n. 41 29: 88n. 131 30: 242– 43 30.1: 33n. 33, 242 31.2: 83n. 114, 115 33.5: 98n. 166, 108n. 187 36.1– 2: 84n. 122 36.2– 45: 85– 86, 237n. 20 36.2: 209n. 188 39: 99n. 169 42.1: 237n. 20, 259n. 93 43.7: 33n. 30 45.1: 85n. 123 46.1– 3: 117 46.1– 2: 138n. 9 46.2: 54n. 35, 212 46.3: 41n. 62 47– 59: 240n. 29 47.1: 86, 110 50.1: 75n. 89, 84n. 119 53.1: 75n. 89 53.3: 232n. 2 53.5: 84n. 119 54: 113 56: 85n. 123 59.1: 110, 138n. 9, 140n. 16 61: 110n. 193

61.1: 89, 89n. 135 61.1– 2: 138 62: 135n. 1 62.2: 115 63: 147n. 37 64– 65: 110n. 193 64.1: 114, 129 64.3: 159 65.2: 119, 148n. 39, 167n. 91 66: 114, 129 71– 72: 163n. 74 72.1: 113n. 201 73: 163n. 75 74: 90n. 139, 138n. 9, 159 75.1: 91n. 143 75.3: 138n. 9 76: 118– 19 76.1: 118, 147n. 37 77: 115– 16, 118 78– 80: 118, 119– 23 78.1: 51n. 28, 119 79.1: 42n. 64, 120 79.3: 54n. 35, 146n. 33 79.5: 162n. 72 81.3: 90, 115 81.4: 38n. 56 82: 33n. 31, 233n. 4, 239– 40 84: 153n. 51 93: 101n. 172, 137 94: 137n. 8 96.2: 33n. 31, 164 99.4– 5: 86 99.5: 38n. 56, 83n. 114 103.2: 34n. 40 104: 148n. 39 105: 34n. 39 105.2: 83n. 117 106: 80, 108n. 187, 137n. 6, 146n. 36 107: 80 108.2: 81n. 107 109.1: 81 110– 17: 123– 32, 255– 56 110.2: 113nn. 203, 205 111.1: 233n. 5 117: 81n. 107

Index of Passages 118.3– 5: 188n. 136 119.2– 4: 188n. 136, 212n. 196 120: 113n. 204 120.1: 116n. 215 122: 116n. 215 122.2: 61n. 55 125.1– 5: 249n. 59 127.1: 116n. 215 127.1– 3: 114n. 206 127.3: 212 127.4: 111 129: 248– 51 131– 32: 115 132.2: 250 134.1: 110n. 193, 250 134.3– 135.3: 250n. 64 137– 39: 88n. 131 142: 88n. 131, 112, 146n. 33 147.3: 65n. 66 156.1: 86n. 129 162.3: 188n. 136 163– 64: 188n. 136 165– 66: 187 166.1: 52 166.2: 39n. 58 168.1: 81n. 108 169.2: 81, 188n. 139 170: 92n. 148 172.1: 74 172.2: 98n. 164 173.4: 82n. 111 176: 156n. 63 177: 89n. 135 178: 80n. 105, 82n. 111, 156n. 63 179: 80n. 105 180.1: 82n. 111 180.4: 36n. 50, 92n. 149 180.5: 77n. 95 182: 86n. 129 183.1: 86n. 129 183.3: 89n. 135 183.4: 156n. 63 184.1: 75n. 88 184.3: 156n. 63 186.1: 92n. 145

187.1: 80 187.3: 140n. 16 189: 80n. 105 189.1: 92n. 151 190: 80n. 105, 97, 167n. 91 191.1: 80 191.2– 4: 80 191.4: 83n. 117 192.1: 89n. 136 192.2: 89n. 136 192.3: 89n. 136 195: 98n. 166 195.2: 228n. 233 195.4: 36n. 51 197.2: 81n. 109 198: 140n. 16, 167n. 91 199.1: 232n. 2 202– 3: 187 205: 187 Book 5 3.1: 138n. 11 3.2– 4.1: 81n. 109 4:167 4.1: 82n. 111 5: 160– 61, 162n. 72 6: 74, 97 8: 167 9: 92n. 147 9.3: 186n. 134 10: 88n. 131 12– 13: 153n. 48 13.1: 233n. 5 21.1: 25 22: 36n. 48, 64 22.2: 29 25: 154n. 54 28.1: 204n. 17 30.1: 204n. 173 32: 39n. 60, 69 40.2: 65n. 67 42.2: 65n. 66 44– 45: 222n. 220 45.1: 34n. 39 45.2: 34n. 39, 37, 72n. 83 47: 62n. 58 49.1: 209n. 188

317

318

Index of Passages

Herodotus Book 5 (continued) 49.2– 3: 209n. 189 49.3: 209n. 190 49.3– 6: 209 49.4: 209n. 190 49.5: 103 50: 213n. 199 56.1: 196, 246 58: 92n. 151 58.3: 39n. 58 59: 90n. 139 65.5– 66.1: 52 66– 70.1: 52– 57 67.1: 54, 60 68.1: 250n. 61 69.1: 56, 60 71: 62n. 58 72.1– 2: 53 72.4: 25n. 15 73.3: 262n. 97 75.1: 213n. 199 76: 213n. 199 77: 214n. 202 77.1: 177, 213n. 199 77.3: 38 77.4: 200n. 164 78: 53, 206– 9, 213n. 201 79– 81: 191n. 146 79.1: 189, 199n. 161 81.2: 190 82– 88: 189, 191n. 146 83.2: 190 85– 86: 191n. 145 87.1– 2: 191n. 145 88.1: 92n. 151 89: 191n. 146 89.2– 90.1: 189 89.2: 190, 199n. 161 89.3: 191n. 145 90.2: 66n. 73 91– 93: 213n. 199 92β3: 58, 246 92δ1: 42n. 64 92γ2– 4: 63n. 62 92.η1– 3: 65n. 67

92: 4n. 10 93.2: 66n. 73 97: 214 97.1: 53n. 32, 209n. 187, 210 97.2: 36n. 51, 207– 10 97.3: 204, 210, 211 101.2: 103 118.2: 99n. 169 Book 6 11– 12: 88n. 131 14.1: 34n. 39, 36n. 50, 221 14.3: 39n. 58 19.2: 195n. 153, 199n. 161 21.2: 227n. 232 27: 195 30.1: 154n. 54 32: 50, 154n. 54 35: 39n. 60 36: 62n. 58 39.2: 53n. 32 43.3: 38n. 56, 101, 203n. 170, 207n. 182, 233n. 6, 259 49– 50: 189 57.2: 66n. 73 58.2: 76n. 94, 96n. 161 59: 96n. 161 60: 96n. 161 61.1: 41n. 62 62: 65n. 67 65: 65n. 66 65.2: 65n. 67 66: 197 66.2: 53n. 32 68– 69: 250n. 61 68.2: 64n. 65 69.4: 64n. 65 70: 65n. 68, 67 70.2: 68n. 75 72: 65n. 68 73.1– 2: 189 74.2: 208 75.2– 3: 68n. 75 77: 112, 208 83: 112 86α–δ: 4n. 10, 188– 90 86– 90: 189

Index of Passages 87: 190, 200n. 164 91.1: 190– 91, 206n. 180 92.1: 189 94.1: 67 98: 201– 5, 214 98.1: 41n. 62 98.2: 211, 217, 227n. 231, 229 102: 67 103: 62n. 58 106: 213n. 199 107.1: 67 108: 221n. 218 109.3: 213n. 201 115– 16: 260 116: 194n. 151 117.2: 247n. 55 119.4: 81n. 107 121– 31: 222n. 220, 259– 65 121: 229n. 239 121.1: 260 122: 62n. 58 125– 26: 62n. 58 125.4: 103, 237n. 17 127.3: 200n. 164 131: 39n. 60 131.1– 2: 58 131.1: 207n. 182 131.2: 246n. 48 132– 36: 57n. 43 132: 57, 210 135.3: 42n. 64 136.1: 210 137: 222n. 220 137.4: 219n. 214 139.1: 243n. 42 Book 7 5– 9: 188n. 138, 203n. 170 5.3: 153n. 48 6.2– 5: 67 6.3– 4: 66 6.3– 5: 197 7.3: 67 8α2: 153n. 48 10α– γ: 48n. 8 10δ1: 213n. 198, 257n. 86 10ε: 122, 185

319

10η1: 146n. 34 10θ3: 167n. 90 11.1: 102n. 174 11.2: 212n. 194 11.2– 4: 188n. 138 11.3: 130n. 255 12– 18: 197 16α1: 99n. 169 16α2: 200n. 164 18.2: 122n. 232, 182n. 124, 185 18.3: 213n. 198 20.2: 44n. 69 21.1: 11n. 29, 29n. 22 26– 31: 11n. 29 35: 11, 153 35.2: 135 35.3: 154n. 54 38– 39: 153n. 51 39.3: 154n. 54 40.2– 4: 250n. 63 40.2: 45– 53, 185n. 132 43.1: 11n. 29 44: 152n. 47 49.5: 245n. 45 50.4: 245n. 45 54.3: 41n. 62 57: 51n. 28, 247 57.1: 22– 23, 95n. 156, 250 57.2: 249n. 60 57.2– 58.1: 23 58.3: 11n. 29 61– 95: 82n. 111 62.1: 82n. 111 63.4: 82n. 111 70.1: 81n. 107 74.1: 92n. 145 83.2: 245n. 45 85.3: 81n. 107 99.1: 34n. 38, 102n. 174 101– 4: 208 102.1: 88n. 131 103.3: 145n. 32 103.4: 99n. 169, 257n. 86 104: 114n. 205 104.3: 213 104.4: 111

320

Index of Passages

Herodotus Book 7 (continued) 106– 7: 51 107.1: 47n. 6 107.2: 154n. 55 108.1: 256n. 87 108.2– 109.1: 11n. 29 109.2: 11n. 29 110: 256n. 87 111: 98n. 165 114: 154n. 56 114.2: 51n. 28 118– 20: 245n. 45 119: 155n. 59 125: 27n. 18, 244– 47, 249 127.2: 11n. 29 131: 192 132.2: 218n. 210, 256n. 87 133: 191– 92 133.2: 256n. 85 134– 36: 175 134.2: 99n. 169 135.1: 43, 155n. 59, 246 135.3: 208 136.1: 256n. 87 136.2: 112n. 198, 192 136.3: 115n. 211 137– 38.1: 192– 94, 206, 229n. 238 139.1– 5: 173– 75, 194, 258 139.1: 32n. 28, 33n. 30, 34n. 38, 38, 173– 75, 171n. 14, 198 139.2– 4: 41n. 62 139.3: 256n. 87 139.5: 204, 215, 261 139.6: 197 140– 41: 197 142– 43: 197, 203 142.3– 143.1: 66 143.1: 48n. 8 144: 209n. 186 144.1: 51n. 28 144.2: 42, 191, 204 148– 52: 225– 30 148.3: 197 151: 226, 229n. 239, 261n. 94

152: 226– 30 152.3: 34n. 38 153: 211n. 193, 247n. 54 157– 62: 218– 19 161.3: 132n. 259 162.2: 83n. 114 166: 51n. 25, 194n. 151 167.1: 36n. 51 167.2: 39n. 58 168: 220– 21, 224– 25 171.2: 243n. 42 172.1: 256n. 87 178.2: 39n. 58 180: 154n. 56 181.2– 3: 154n. 55 187: 245n. 45 187.1: 249n. 58 190: 255n. 77 191.2: 41n. 62 194.2: 153n. 51, 154n. 54 204– 5.1: 178 204: 176n. 114, 246 205.3: 229n. 237 208.1– 3: 145n. 32 208.1: 176n. 114 208.3: 233n. 5, 247 209.3: 114n. 205 210.2: 102n. 174 211.3: 246n. 51 213– 14: 222– 23 213.3: 15n. 40 214.3: 17n. 47, 33n. 30, 222, 261 215: 176n. 114 219.2: 175, 208n. 185 220: 175– 78 220.4: 146 222: 229n. 237 223– 25: 176n. 114 223.2– 3: 246n. 51 223.2: 176n. 114 225.2: 246 226.1– 2: 115n. 211 226.2: 33n. 30 233: 193n. 149, 206, 229n. 237 234.2: 112n. 198

Index of Passages 235: 203n. 170 236.1: 174n. 108 237.2– 3: 271n. 15 238.2: 153n. 51, 154n. 54 239.1: 33n. 30 239.2: 36n. 51, 41n. 62 Book 8 2.2: 217 3: 214– 17 3.1: 130n. 256, 227, 227n. 231 3.2: 69, 200n. 164 4– 5: 57n. 43 8: 254– 55 8.3: 32n. 28 15.1: 194n. 150 22.2: 256n. 87 27: 66 37– 38: 247n. 55 38: 99n. 169 41: 116n. 215 42: 51n. 27 56– 64: 221n. 218 57.2: 208n. 185 59– 61: 223 59– 62: 224n. 227 62.2– 63: 57n. 43 66.1: 51n. 27 67– 69: 256n. 87 68α1: 102n. 174 68γ: 112 69.1: 271n. 15 75– 76: 57n. 43 75: 221n. 218 77: 197– 201 78: 220n. 215 79.1: 39n. 60, 43, 261n. 94 83.2: 99n. 169, 146n. 191 84.1– 2: 224n. 227 85: 224n. 227 85.2: 33n. 30 87– 88: 256, 258 87.3: 41n. 62 88.3: 102n. 174 90.3: 154n. 54 92: 224n. 227 94: 223– 24

321

98.1: 54n. 35 98.2: 90– 91 99: 256– 57 105.1: 47 105.2: 76n. 93, 155n. 59 108– 9: 204, 207n. 183 109– 10.1: 57n. 43 109.3: 180, 185, 187 109.5: 57, 67 110.1: 210 111– 12: 57n. 43, 204, 218n. 210, 221n. 217 112: 154n. 54 118: 153n. 51 122.3: 130n. 255 124.2: 43n. 67 129.4: 165n. 84 140– 49.11: 135n. 2, 221n. 218 140α2: 256n. 87 142.3: 207n. 183 143– 44: 208n. 184 221: 66 238.2: 69n. 80 Book 9 2.3: 53n. 32 6– 8: 213n. 199 7: 208n. 184 11: 135n. 2, 208n. 184 11.2: 115n. 211, 218n. 210 17.1: 256n. 87 26– 27: 219– 20 27.2: 200n. 164 27.4: 125n. 241 33– 35: 59– 61 34.1: 60, 83n. 114 35: 59 35.1: 47, 61 35.2: 205n. 178 37– 38.1: 66– 68 46.2: 246 48.3: 177 54.1: 135n. 2 62: 220n. 215 64.1: 69, 177 65.2: 166n. 87, 243– 44 68: 41

322

Index of Passages

Herodotus Book 9 (continued) 73: 205n. 178 73.2: 200n. 164 75: 205n. 178 76: 69 77: 206n. 180 78– 79: 69 78.1: 42 78.1– 3: 177 79.2: 69 82: 69, 155n. 59 90– 91: 115n. 211 92.2: 70 93– 94: 70– 73 95: 70 98.4: 51n. 27 100– 101: 194– 95 100.2: 51n. 25, 155n. 59 105: 205n. 178 106.2– 4: 204 107: 101– 2 108– 13: 65n. 67 109.2: 42n. 64 109.3: 152n. 47 114.2: 58 116.3: 85 117: 58 119– 20: 58n. 46 122.1– 2: 88n. 131 122.2: 153n. 48 Hesiod MW frags 37: 60n. 53 129: 60n. 53 130: 60n. 53 131: 60n. 53 150: 116n. 213 Theogony 27: 254n. 76 Works and Days, 242– 44: 72 378: 142n. 22 Hippocratic Corpus Airs, Waters, Places 13, 15, 16: 87 17– 22: 109n. 188, 141n. 17 17: 129n. 253 On Joints 53: 125n. 243 Homer

Iliad 1.8: 29n. 22 1.146: 119n. 223 2.235: 102n. 173 2.308– 32: 46 2.552: 218n. 211 5.62– 63: 204n. 173 5.703: 29n. 22 6.152– 206: 124n. 237 7.125: 218n. 211 8.163: 102n. 173 11.86: 176n. 114 11.113– 21: 246n. 49 11.170– 78: 246n. 49 11.473– 86: 246n. 51 17.274– 87: 176n. 114 18.503– 4: 72 18.573– 81: 72 22.124– 25: 102n. 173 24.53: 185n. 130 Odyssey 1.325– 52: 4n. 10 6.119– 20: 74n. 84 8.266– 369: 4n. 10 8.471– 541: 4n. 10 9.105– 15: 74n. 84 11.281– 97: 60n. 53 11.333– 476: 4n. 10 12.127– 33: 71 12.439: 176n. 114 15.225– 46: 60n. 53 19.203: 254n. 76 Homeric Hymns Ap. 3.18: 238n. 22 Ap. 3.25– 138: 238n. 21 Ap. 3.48: 239n. 24 Ap. 3.55: 239n. 24 Ap. 3.60: 239n. 24 Ap. 3.72: 239n. 24 Ap. 3.113: 238n. 22 Ap. 388– 544: 253 Dem. 3.3: 239n. 25 Dion. 1.20: 164n. 79 Dion. 7: 253 Isocrates Panegyricus 4.68– 70: 125n. 241, 132n. 259

Index of Passages Lucian Herodotus 1– 2: 15n. 38 Lysias Epitaphios 2.4– 26: 125n. 241 2.4: 132n. 259 2.6 : 123n. 236 2.45: 174n. 107

Marcellinus Life of Thucydides 54: 15n. 38 Myron of Priene FGrHist 106 F 2: 113n. 199

Panyassis frag. 26K: 142n. 22 Pausanias 1.15: 124n. 239, 202n. 169 1.17: 124n. 239 6.2.5: 61n. 57 6.4.5: 62n. 58 6.17.5– 6: 62n. 58 Pherecydes FGrHist 3 F 15: 124n. 237 17: 142n. 22 F 33: 60n. 53 F 114: 60n. 53 F 115: 60n. 53 F 151: 124n. 237, 128n. 252 F 152: 124n. 237, 128n. 252 174: 116n. 213 Phocylides Frag. 4 Diehl: 240n. 32 Pindar Isthm. 6.23: 5n. 91 Nem. 3.36– 39: 124n. 237 Ol. 1.55– 57: 200 6.35– 72: 61n. 57 6: 62n. 58 13.10: 200n. 163 13.63– 92: 124n. 237 Paean 4.24– 26: 60n. 53 SM frags. 33d: 239n. 26 169.1– 4: 170 172: 124n. 237

323

175: 128n. 252 176: 124n. 237 Plato Apology 35b2– 3: 102n. 173 De Exilio 13: 271n. 16 Gorgias 482E– 84D: 171 483E: 171n. 100 491E: 171n. 100 492B: 171n. 100 Laches 190D– E: 114n. 205 191A– C: 114n. 205 Republic 392d– 94c: 17n. 44 454d: 252n. 70, 255n. 77 463c– 65b: 79n. 103 Theaetetus 155d: 234n. 8 Plutarch Lycurgus 12: 114n. 207 22: 108n. 187, 114n. 208 28: 113n. 199 On the Maliciousness of Herodotus, Mor.854E– 74C: 172 857A– B: 143n. 26 857A– F: 7 857C: 155n. 59 857F– 858A: 184 862A– B: 15n. 38, 271n. 13 862E: 261n. 95 862F– 63A: 263n. 101 863A– B: 262n. 98 863A: 224n. 226 863E: 228 864A– B: 175n. 112 864C: 255n. 77 864D: 271n. 13 867B: 263n. 101 868A: 172n. 16 870B– 71B: 224n. 226 871C: 255n. 77 871D: 7 Pericles 3: 58n. 48 6.2: 269n. 6 7.1– 2: 58n. 48 12: 58n. 48 16: 58n. 48 28: 203n. 170 28.7: 202n. 169

324

Index of Passages

Plutarch Pericles (continued) 31.4: 124n. 239 37.5: 264n. 103 Theseus 26– 28: 124n. 237 26: 128n. 252 Polybius 6: 123n. 235 Protagoras B4: 163n. 76 DK 80 A2: 166n. 87 Simonides frag. 11 W2, 42: 61n. 57 Solon frag. 5 Diehl3, 9: 200n. 163 Sophocles Ajax: 65n. 70 189n. 140 Oedipus Tyrannus: 198n. 158 872: 200n. 164 Trachiniae 1071– 72: 102n. 173 Strabo 6.1.13– 15: 267n. 2 6.1.13: 269n. 8 6.1.15: 270n. 11 7.3.9: 80n. 106, 108, 141n. 17 11.5.1: 128 11.5.4: 123n. 236 11.6.2: 80n. 106 11.6.3: 234n. 9 11.11.3: 161n. 71 15.3.17: 154n. 53 Suidas  s.vv. Ηροδοτος and Πανυασσις 267n. 1 Tacitus Germania 22: 154n. 56 Theognis 153– 54: 200n. 163 425– 28: 101n. 172 713: 254n. 76 825– 30: 114n. 208

Thucydides 1.1– 19: 44n. 69 1.22.1: 174n. 108 1.22.4: 99n. 169 1.23.1– 3: 46 1.23.6: 212n. 194, 213n. 200 1.24– 55: 221 1.69.5: 174n. 107, 203n. 170, 258n. 89 1.73.1: 220 1.73.2: 220 1.73.4– 74.4: 74n. 107 1.75: 208n. 184, 212n. 194 1.75.1: 216n. 206 1.76.4: 219n. 214 1.77.6: 216n. 206 1.79.2: 116n. 212 1.80.1: 213n. 200 1.82.1: 229 1.84.2: 116n. 212 1.84.3: 116n. 212 1.86.1: 174n. 109, 215, 220 1.86.3: 146n. 35 1.94: 68– 69 1.98.3: 205n. 178 1.102.4: 229n. 237 1.108.4: 190n. 143 1.118.2: 213 1.118.3: 198n. 158 1.122.3: 58n. 49 1.124.3: 58n. 49 1.131– 34: 68– 69 1.140.2: 146n. 35 1.143.5: 116n. 215 1.146: 146n. 35 2.7.1: 213n. 200, 229n. 237 2.8.2– 3: 198n. 158 2.8.3: 201 2.8.5: 174n. 108 2.17.2: 198n. 158 2.27.1: 190n. 142 2.28.3: 194n. 150 2.42: 189n. 140 2.52– 68: 221n. 218

Index of Passages 2.52– 54: 198n. 158 2.63.2: 58n. 48 2.64.5: 174n. 108 2.65.9: 58n. 48 2.67: 193, 229n. 238 2.70.3: 219n. 214 2.71– 78: 221n. 218 3.20– 24: 221n. 118 3.37.2: 58n. 48 3.50: 219n. 214 3.62.3– 4: 53n. 32 3.82.3: 99n. 169 3.96.1: 198n. 158 4.27.3: 102n. 173 4.36.3: 175n. 112 4.40.1: 175n. 112 4.40.2: 115n. 211 4.50: 229n. 238 5.14.4: 229n. 237 5.26.3: 198n. 158 5.28.1: 229n. 237 6.8– 26: 210 6.85.1: 58n. 49 7.28.3: 194n. 150 7.50: 194n. 150 7.51.1: 194n. 150 7.87: 44n. 69

325

Tibullus 4.1.144– 45: 162n. 72 Tyrtaeus frags. 12.13– 20, 11.31– 34 W: 114n. 205 Tzetzes Chiliades 7.629– 36: 83n. 117 Xenophanes DK 21 A14: 149n. 42 DK 21 A34: 166n. 88 Xenophon Anabasis 5.4.34: 151n. 45 Lakedaimonion Politeia 1.3: 104n. 177 5: 114n. 208 9.4: 114 10.7: 112n. 197 11.3: 102n. 173 13.1: 112n. 197 Oeconomicus 7.23: 130 [Xenophon] Athenaion Politeia 1.6.– 9: 208n. 184 2.2.1: 191n. 146 2.7: 2.14: 191n. 146, 156n. 61 2.20: 208n. 184