The Exemplifying Past: A Philosophy of History 9789048537891

Citation preview

The Exemplifying Past

The Exemplifying Past A Philosophy of History

Chiel van den Akker

Amsterdam University Press

Cover image: Rob Scholte, Vrede van Münster © Rob Scholte / 1998 Photographer: courtesy of Collectie Museum Prinsenhof Delft Cover design: Coördesign, Leiden Lay-out: Crius Group, Hulshout isbn 978 94 6298 661 9 e-isbn 978 90 4853 789 1 (pdf) doi 10.5117/9789462986619 nur 687 © Chiel van den Akker / Amsterdam University Press B.V., Amsterdam 2018 All rights reserved. Without limiting the rights under copyright reserved above, no part of this book may be reproduced, stored in or introduced into a retrieval system, or transmitted, in any form or by any means (electronic, mechanical, photocopying, recording or otherwise) without the written permission of both the copyright owner and the author of the book.

To L.D.B.

Contents Preface 11 1 Introduction: Retroaction, Indeterminacy, and Seeing-in

13

2 Periods and Other Minds

21

3 Narrative Truth

51

4 Resemblance, Substitution, Expression

75

5 Exemplification

99

6 Danto’s End of Art

127

Bibliography 149 Index 155

Das Geschäft des Geschichtschreibers in seiner Letzten, aber einfachsten Auflösung ist Darstellung des Strebens einer Idee, Dasein in der Wirklichkeit zu gewinnen. − Wilhelm von Humboldt

Preface The Swiss historian Jacob Burckhardt (1818-1897) famously claimed that modern man was born in the Italian Renaissance of the fourteenth and fifteenth centuries, and although he wanted his readers to be astonished by this high point in European culture, he also warned his readers that all mid-nineteenth-century modern evils – radical politics, unbridled egotism, corruption, the cheapness of the masses’ aesthetic taste – had their roots in the individualism that defined modern man. Here history estranged its audience from the present world they lived in by raising historical awareness of the way they lived their lives. This is how, I think, it should be. It is not that we should turn to the past to criticize the present – we may equally well turn to the foreseeable future to do that, for we know that in the future, the present has become past, and our criticism of the present, if we have any, is to serve future pasts. It is also not that the past should be used to tell us how we ought to behave and act. It is that historical awareness makes us realize that life is historically conditioned, and in that realization, we estrange ourselves from the present we happen to live in, for what at first appeared to be natural and self-evident, turns out to be mutable and subject to change. It is not always clear how a particular view of the past has us take a particular attitude towards our present, but inasmuch as a view of the past intends to have an audience see the past in a certain light, it is an instrument of rhetoric. Burckhardt not only interpreted and explained fifteenth-century Italian culture and its antecedents in the fourteenth century in his Die Kultur der Renaissance in Italien, he also presented it in a certain way, namely as the birth of modern man. This is how he proposed we should view the past. Such views of or theses on the past interest me philosophically in this book. This interest is motivated by the observation that the historian sees something in the past which was not there according to past witnesses and the evidence that is available today. Think of how Burckhardt called Petrarch ‘one of the first truly modern men’ because of his sense of natural beauty (Petrarch famously climbed Mont Ventoux to enjoy its view and allegedly was the first man since antiquity to do so). This sense of natural beauty was there in the past to witness and there is evidence for it, but it is Burckhardt who saw in Petrarch a modern human being. One explanation of this remarkable feature of historical understanding is that views are never part of what they are a view of: what the historian sees in the past simply was not there to see. Another related but different explanation of it would be to claim that the historian’s view or thesis retroactively became concrete

12 

The Exemplifying Past

in the past. Petrarch became a modern man because of Burckhardt’s work, and with that, a predecessor of him and his contemporaries. There are, in other words, two sides to the views of the past. In this book my interest is not only in the nature of these views; I am equally interested in how the past makes us view it, i.e. in how we see in the past what has become concrete in it, but which was not there for witnesses and their contemporaries to see. At this point readers no doubt will be unconvinced that there really are two sides to these views of or theses on the past. It is the task of this book to convince them. The argument of this book is built up cumulatively, and even though some of its chapters have appeared as articles, they are not to be read as independent essays. Chapter 3 was published as ‘Mink’s Riddle of Narrative Truth’ in the Journal of the Philosophy of History in 2013 in a special issue on history and truth, edited by Professors Frank Ankersmit and Jeff Malpas. Chapter 6 appeared in the same journal in 2018 as ‘Arthur Danto, the End of Art, and the Philosophical View of History’ under its then new Editor-in-Chief Professor Jouni-Matti Kuukkanen. Several minor revisions and additions have been made to these essays. I thank Koninklijke Brill for their permission to republish them. Chapter 4 makes use of some of the material of the essay ‘Ankersmit on historical representation. Resemblance, substitution and exemplification,’ published in the journal Rethinking History in 2011. All of the reused material has been thoroughly rewritten and the argument substantially expanded and improved upon. I thank Editor-in-Chief Professor Alun Munslow for his kind support in having the essay published. An early version of Chapter 5 appeared in the Journal of the Philosophy of History in 2012 at the invitation of its then Editor-in-Chief Professor Ankersmit. The essay has been fully rewritten and updated, and its argument reworked and substantially expanded, making it a rather new chapter. I am grateful to Ankersmit for his support of my work. Rob Scholte I thank for his kind permission to use his wonderful 1998 Vrede van Münster for this book’s cover, which depicts a detail of the work. Museum Prinsenhof Delft provided me with its image. The painting was commissioned by the Nationaal Comité Herdenking Vrede van Münster to commemorate the 1648 Peace of Münster. It presents the European order as the outcome of war. This legacy, the painting tells us, is ours and it is not to be forgotten, as peace is still but a ‘newborn’ in the hands of Time. Chiel van den Akker Amsterdam, 2018

1

Introduction: Retroaction, Indeterminacy, and Seeing-in

1 Seeing-in This book argues that to see the past in a certain light is to have the past exemplify the spirit of its age. The latter term may seem antiquated and sound too Hegelian to some tastes, but the reason to use it here is that it captures the scope of the argument. (Rest assured, the book most of the time makes use of a contemporary idiom instead of this ‘ghostly’ entity.) The book wants to make sense of the claim that the spirit of an age retroactively becomes concrete in what has actually been found to have existed. Take the widely known example used in the preface, and think of how the spirit of modern man became concrete in the behaviour, attitudes, and concerns of the Italians studied by Jacob Burckhardt, in how they treated the states of affairs objectively, or how Francesco Sforza personally earned the credit of his soldiers. Think also of their calculated self-interest, the Venetian statistical accounts of its resources, their modern desire for fame and sense of moral responsibility, their discovery of the aesthetics of the outer world, and their depiction of chivalry as ludicrous. Such behaviour, attitudes, and desires exemplify the birth of modern man. It is something that Burckhardt as an historian sees in the past. Several other and more recent examples of this special and defining feature of historical understanding will be discussed in this book. The emergence of modern man in fourteenth and fifteenth-century Italy can be said to be Burckhardt’s view of or thesis on the past. In this book my interest is not only in the nature of these views;1 I am equally interested in how we retroactively see something in the past which was not there to see at first, but which has become concrete in what has been found to have existed. This claim suggests that the past is indeterminate and mutable, open and subject to change. It is one of the claims that are argued in Chapter 3 of this book. Here I start with the very idea of the indeterminate past and the philosophers who put this topic on the agenda of the philosophy of history. 1 Jouni-Matti Kuukkanen, Postnarrativist Philosophy of Historiography (Basingstoke: Palgrave Macmillan, 2015), 1, notes that only since the last third of the twentieth century these views are being analyzed by what has become known as the narrativist philosophy of history. I discuss the narrativist philosophy of history in Chapters 3 and 5 of this book.

14 

2

The Exemplifying Past

Indeterminacy and Re-description

One reason why the past is indeterminate is that new concepts and newly acquired sensibilities may motivate us to re-describe past actions. Since such re-descriptions change what someone in the past did, the past is indeterminate. Of course, the historian’s intuition to avoid anachronisms is to be taken seriously, as is his objective stance towards his subject matter, but we cannot expect him to simply ignore concepts that help him understand past actions because those concepts were unavailable to the persons he aims to understand. Nor can we expect him to simply shake off all of his moral values: that would be to ask him to deny that the values he has are historically conditioned. I might add that historians have always successfully incorporated concepts from other scholarly and scientific disciplines into their work to improve their understanding of the past, and that is one of the reasons why their work is so interesting. The result of re-describing past actions may surprise. This is what Ian Hacking has to say: If a description did not exist, or was not available, at an earlier time, then at that time one could not act intentionally under that description. Only later did it become true that, at that time, one performed an action under that description. At the very least, we rewrite the past, not because we find out more about it, but because we present actions under new descriptions.2

Hacking gives the example of deserters who were executed during the Great War and who retroactively were said not to have deserted but to have suffered from a post-traumatic stress disorder. Another example concerns the Scottish explorer Alexander Mackenzie who in 1802 married a fourteen-year-old girl, which at that time was not illegal according to Hacking. Retroactively, his consummation of the marriage is labelled as child abuse and thus presented under a new description. A third example, and one that he criticizes, concerns someone who regards Oedipus Rex as 2 Ian Hacking, Rewriting the Soul. Multiple Personality and the Sciences of Memory (Princeton: Princeton University Press, 1995), 243. This aspect of the book is much discussed. Of particular interest is Paul Roth, ‘Ways of Pastmaking,’ History of the Human Sciences (2002), 15(4), 125-143, which, in my opinion, offers the most interesting discussion of Hacking’s claims. Arthur Danto’s Narration and Knowledge (New York: Columbia University Press, 1985) is mentioned by neither Hacking nor Roth. This is surprising since Danto’s book is all about presenting past actions under new descriptions at a later date. Elsewhere, in a somewhat different context but also in relation to Hacking’s emphasis on retroactive re-descriptions, Roth does discuss Danto’s views. See Roth, ‘The Pasts,’ History and Theory, 51 (2012), 313-339.

Introduc tion: Retroac tion, Indeterminac y, and Seeing-in

15

a story of child abuse.3 This appears to be a clear case of an anachronism gone wrong, or perhaps it is simply a case of someone missing the point of this particular story, or of someone misunderstanding the concept of child abuse. The line between warranted and unwarranted anachronisms cannot be drawn beforehand other than by stating that the concepts used to describe the past should clarify rather than obscure it to us. Here the point is that the past is indeterminate in that past actions are presented under new descriptions after a new moral, medical, or other type of concept, has become available. Using new moral and medical concepts are evident examples of re-descriptions that change what someone did, but also new economic, social, political, cultural, and artistic concepts allow us to present past actions under new descriptions. Arthur Danto gives the following example: ‘Monet influenced not a single member of the New York School [of abstract expressionism] but because these men began to paint in a special way Monet became a predecessor in his later works.’4 Or think of the pre-Socratic philosophers who were not intentionally doing this sort of philosophy, as if they knew how philosophy would change after Socrates would enter the scene. The description of their philosophy as being pre-Socratic changed the philosophy they in fact professed. Perhaps someone would object that such re-descriptions do not alter the past itself but merely alter our way of talking about it. But since all action is action under a description,5 newly available descriptions do change what happened in the past. There are not, on the one hand, actions outside any description, and on the other, descriptions and re-descriptions of these actions. The distinction central to this book is that between events under the description of witnesses and their contemporaries and events under the description of historians which were unavailable to witnesses and their contemporaries. Here talk of events is a shorthand for talk about what individuals did and went through, and the attitudes, desires, beliefs, fears, volitions, values, hopes, and dreams we associate with these occurrences and the circumstances that made them possible. To be sure, historians study remains rather than events, but these remains are intelligible as remains insofar as they can be related to the behaviour and inner states of the individuals that brought them about. 3 Hacking, Rewriting the Soul, 241-242. 4 Danto, Narration and Knowledge, 168. 5 The point was first made by Elizabeth Anscombe, ‘Under a Description,’ Noûs, 13(2) (1979), 219-233.

16 

The Exemplifying Past

We often use new concepts to re-describe past actions. It does not follow, or not for the reason suggested by Paul Roth, that ‘we choose, in some important respects, our history as well. For what sense can be made of our past also depends on our stock of descriptions for describing it.’6 We are less free than Roth suggests, because the stock of descriptions that is available, and with that, the sense we can make of our past, is itself historically conditioned and therefore not a matter of choice. Our sense of the past depends on how the present came to be, i.e. on the stock of descriptions that has become available and the reasons for it. The point that Hacking, Roth, and Danto agree upon is that some description available at t-2 was not available at t-1, and the past at t-1 is re-described by using a description that became available at t-2. This makes evident that the past is indeterminate, but it does not explain why the past is this way. The reason is because the future is. This is Danto’s argument. Since later events determine the meaning of the earlier events with which they are connected, events will always be re-described as long as the future is open. These yet unknown future events will allow for new connections to the events that are already known, and those unknown future events will determine the future historian’s interests and the stock of descriptions then available and preferred. The openness of the future thus guarantees the openness of the past.7 Re-describing the past with the help of new concepts or because of new sensibilities is not the only reason why the past can be said to be indeterminate. Another reason is that historical narratives make it clear why some event is significant in terms of its connection with later events and in terms of what historians see in them. Here too past events are presented under new descriptions that were unavailable to witnesses and contemporaries. The description of an event as being historically significant is only available at a later date. Danto brought this feature of historical knowledge to our attention and it forms the heart of his analytical philosophy of history.8 He puts it thus: events are continually being re-described, and their signif icance reevaluated in the light of later information. And because they have this 6 Roth, ‘Ways of Pastmaking,’ 136. 7 Danto, Narration and Knowledge, 17, 181. 8 The field is habitually distinguished from the so-called substantive or speculative philosophy of history, which is concerned with the nature of the historical process, its direction and purpose, rather than with the nature of historical knowledge, with which the analytical philosophy of history is concerned. Substantive philosophy of history is typically associated with Hegel’s and Marx’s grand theories of history. I discuss the distinction between the substantive and analytical philosophy of history in Chapter 6 of this book, and, in a sense, propose to draw that distinction anew.

Introduc tion: Retroac tion, Indeterminac y, and Seeing-in

17

information, historians can say things that witnesses and contemporaries could not justifiably have said.9

Think once again of Burckhardt’s description of Petrarch as the first truly modern man because of his sense of natural beauty.10 This description was not available to Petrarch and his contemporaries – and even if it were available, it could not have meant to them what it does to us, since our understanding of it depends on our knowledge of what has happened after Petrarch, and that, obviously, is something Petrarch and his contemporaries could not have known. The re-description of the past here depends on the narrative that makes evident why Petrarch’s climbing of the Mont Ventoux and enjoyment of its view is historically significant. Such action or event may, because of that dependency, be called an action or event under a narration. In this book I am interested in re-descriptions that attribute a historical meaning to what someone in the past did. Such re-descriptions depend on narrative. The argument is that past behaviour, attitudes, desires, and the objects and events associated with them, exemplify certain properties after being historically understood. As will be explained in Chapters 5 and 6, the historical thesis presented by the historian retroactively becomes concrete in the past he discusses. In virtue of his narrative, the past exemplifies the historical thesis the narrative expresses. The exemplification theory of history as proposed in this book thus explains both how the past acquires its meaning and how it is related to the historian’s narrative. One of the remarkable effects of history writing is that a historical thesis expressed by some narrative may become concrete in events that are not discussed in that narrative. If narratives were theories, this would not surprise us, for theories, after all, treat events as instances of the theory that explains their existence. But narratives are no theories. The events they represent are not instances of the thesis they express; rather those events exemplify the thesis expressed by the narrative. The relation they have to the evidence is different. A theory treats what exists as an instance of the theory and uses evidence of what exists to validate the theory. The evidence either confirms or disconfirms the particular theory or hypothesis. This is not how evidence relates to historical theses. A central though counterintuitive claim of this book, which is argued in Chapters 3, 5, and 6, is that a historical thesis cannot be confirmed or disconfirmed since there is no empirical 9 Danto, Narration and Knowledge, 11. 10 Jacob Burckhardt, The Civilization of the Renaissance in Italy, trans. S.G.C. Middlemore (1878), 119.

18 

The Exemplifying Past

evidence for or against it, yet such thesis has become concrete in the past under consideration, is credible, and can be taken to be true in the sense that it is our best guide to the past present at hand. Historical theses, I hold, are exemplified rather than being justified by the available evidence. I might add that this last claim does not in the least oppose the idea that there are rational criteria such as consistency and scope that enable us to evaluate the merits and plausibility of historical theses, as philosophers of history such as Frank Ankersmit and more recently Jouni-Matti Kuukkanen have convincingly argued.11 Such criteria are not, however, what this book is concerned with.

3 Exemplification The notion of exemplification figures in two of this book’s central claims. One is that past objects and events exemplify the historical thesis of some narrative, even if those objects and events are not mentioned in the narrative expressing that thesis. The other claim is that historical theses are exemplified rather than being justified by the available evidence. The notion of exemplification is well known from Nelson Goodman’s theory of symbols.12 I depart from his notion for reasons that are rather technical and that I discuss in the margins of Chapters 4 and 5. It is not my aim to be true to Goodman when using the term ‘exemplification’. However, I take from Goodman the idea that exemplification is a semantic term; a form of reference that runs in the opposite direction from the direction we are accustomed with: it runs from object or behaviour under consideration to symbol rather than from symbol to object or behaviour; 13 and I take from Goodman the idea that the exemplifying object or behaviour only exemplifies some but not all of its properties. I argue that the past under consideration in some narrative exemplifies the historical thesis expressed 11 See e.g. Frank Ankersmit, Historical Representation (Stanford: Stanford University Press, 2001), 15 and 96-97, and Kuukkanen, Postnarrativist Philosophy, especially 155-158. 12 Nelson Goodman, Languages of Art. An Approach to a Theory of Symbols (Indianapolis and Cambridge: Hackett Publishing, 1976). On the relevance of Goodman for the philosophy of history, see Eugen Zeleňák, ‘Using Goodman to Explore Historical Representation,’ Journal of the Philosophy of History, 7(3) (2013), 371-395. 13 This is the most striking feature of exemplification. Goodman, Language of Art, 65, writes: ‘exemplification is reference running back from denotatum back to label.’ When a tailor uses a swatch to show you the fabric of some suit, the swatch exemplifies the fabric. Similarly, when a historian uses some event to show the reader how some part of the past is to be understood in terms of some thesis, the event (the denotatum) exemplifies that thesis (the label).

Introduc tion: Retroac tion, Indeterminac y, and Seeing-in

19

by that narrative as a result of being historically understood. Past events only retroactively acquire the property of being of a certain kind relative to some narrative. The term ‘exemplification’ has its place in a semantics of narratives rather than in a semantics of statements about the past. This is one of the claims argued in Chapter 3. The concept is further discussed in the context of different theories of representation in Chapter 4. There the term ‘representation’ is used in Goodman’s sense of symbolization and in the sense proposed by Ankersmit in his theory of historical representation. Historical representations, Ankersmit argues, are proposals as to how the past should be viewed.14 The notion of exemplification is subsequently discussed in Chapter 5 in the context of the so-called narrativist philosophy of history. Finally, the concept of exemplification is discussed in relation to what I will call the philosophical view of history and Danto’s somewhat notorious end-of-art thesis in Chapter 6. This last chapter may appear to stand out from the others in that it is as much concerned with the philosophy of art as with the philosophy of history. However, the reader soon will realize that its main concern is the philosophy of the history of art. The themes that are discussed in the other chapters – the indeterminateness of the past, retroactive re-description, narrative coherence, the concept of representation, and the nature of historical 14 Ankersmit, Narrative Logic. A Semantic Analysis of the Historian’s Language (The Hague: Martinus Nijhoff Publishers, 1983); Ankersmit, Historical Representation; and Ankersmit Meaning, Truth, and Reference in Historical Representation (Ithaca and New York: Cornell University Press, 2012). The term ‘representation’ is somewhat unfortunate in that it suggests a representational view on knowledge and language (Ankersmit does not hold such a view: historical representations are not pictures or (mirror) images of the past). In this book I agree with the anti-representationalists Donald Davidson (Chapter 2), Willard Van Orman Quine (Chapter 3), Nelson Goodman (Chapter 4), and Richard Rorty that it is misleading to talk about a ready-made world that our representations should fit, match, mirror, or correspond to. This is not to deny that historians should accurately represent the past. In his essay ‘John Searle on Realism and Relativism’ in his Truth and Progress. Philosophical Papers, Volume 3 (Cambridge: Cambridge University Press, 1998), 63-83, Rorty urges us to distinguish between what he refers to as the philosophical view of accurate representation (which has to do with the view that language should represent reality as it is and hence correspond to it) and the non-philosophical view of accurate representation (which has to do with the norms of the historical profession, such as being unbiased, doing research free from state and church, using methods and being transparent about them, not intentionally disregarding evidence, not letting prejudice cloud one’s judgment, and so on). The best introduction to anti-representationalism to my knowledge is Giancarlo Marchetti, ‘Davidson and the Demise of Representationalism,’ in Jeff Malpas ed., Dialogues with Davidson. Acting, Interpreting, Understanding (Cambridge MA and London: The MIT Press, 2011), 113-128.

20 

The Exemplifying Past

theses – all come together in this chapter and the philosophical view of history that it argues for. Danto’s end-of-art thesis is, I argue, a historical thesis, and the artworks that he discusses exemplify rather than justify his thesis. Not only historians are capable of seeing a historical thesis in the course of past events. Anyone who turns to the past in order to have his or her audience take a particular attitude towards the present has to take recourse to a historical thesis at some point. The book starts in Chapter 2 with separating the problem of other minds (how do we understand others?) from the problem of other periods (how do we understand times other than our own?). The first problem is concerned with understanding utterances and inscriptions of witnesses and their contemporaries. The second with the historian’s language that is outside the grasp of witnesses and their contemporaries. As said, the distinction between events under the description of witnesses and their contemporaries and events under the description of historians is central to this book.

4

A Motto

The question whether history is a science, an art, or both, and whether that is how it should be, has shaped the field we know today as the philosophy of history.15 The question itself has become rather tedious for some time now, and it is not one I incline to address in this book. I do, however, agree with Danto, when he concludes that: A certain autonomy then attaches to history, indeed to narrative history, which cannot become more ‘scientific’ without losing its defining human importance since it is human interests, after all, which determine which events are important and under what sort of descriptions.16

If I were to give this book a motto, it would be this.

15 On the f irst and def ining days of this f ield, see Frederick C. Beiser, After Hegel. German Philosophy 1840-1900 (Princeton and Oxford: Princeton University Press, 2014), 133-157. The question was again defining the field in the twentieth century in the aftermath of the publication of Carl Hempel’s, ‘The Function of General Laws in History,’ The Journal of Philosophy, 39(2) (1942), 35-48. 16 Danto, Narration and Knowledge, xii.

2

Periods and Other Minds And new philosophy calls all in doubt, The element of fire is quite put out, The sun is lost, and th’earth, and no man’s wit Can well direct him where to look for it. And freely men confess that this world’s spent, When in the planets and the firmament They seek so many new; they see that this Is crumbled out again to his atomies. ’Tis all in pieces, all coherence gone, All just supply, and all relation; Prince, subject, father, son, are things forgot, For every man alone thinks he hath got To be a phoenix, and that then can be None of that kind, of which he is, but he. This is the world’s condition now, − John Donne, excerpt from An Anatomy of the World (1611)

1 Introduction Between September 1610 and October 1612, Ludovico Cigoli painted the fresco Assumption of the Virgin in the Borghese Chapel at the Santa Maria Maggiore in Rome, depicting Mary ascending towards the opened heavens while standing on the moon, which was painted exactly as it was revealed through the telescope of Cigoli’s long-time friend Galileo Galilei. Two worlds meet in this fresco: the traditional world of the Catholic Church which commissioned the fresco and the world of science – of modern science, based on observation and instruments that enhance and extend the human senses. It makes us wonder: How was Cigoli able to not depict the moon according to custom, as smooth and transparent, pure and unmarred, as the Madonna herself? The explanation of historians is straightforward. One only has to read those theological treatises that explain why the moon is imperfect. One expert theologian, Andrea Vittorelli of Basano del Grappa, wrote: ‘The Moon not only signifies the flaw of corruption, but lunacy […]. All Ignorance [is] under the feet of Mary.’1 In 1612, Cigoli’s marred 1 Massimo Bucciantini, Michele Camerota, and Franco Giudice, Galileo’s Telescope. A European Story (Cambridge MA and London: Harvard University Press, 2015), trans. Catherine Bolton, 223.

22 

The Exemplifying Past

moon thus confirmed the church’s teachings to one, and the path-breaking discoveries of Galileo and his new instrument om verre te sien to another. The theologians and the Galileonians not only disagree as to what Cigoli’s moon signifies, they also disagree as to what the moon is. They think differently about the moon. If we have the required skills, we should be able to understand what the moon is to each of them. It appears that the distance in time, and the differences in beliefs that this distance testifies to, is at best a practical obstacle here. Not only is the example just given proof of that. Historians have provided an overwhelming amount of evidence that to our standard as to what counts as understanding others, we are able to understand minds living in times and places that are remote from our own. It does not follow that there is no such thing as the problem of other periods. It does follow that the problem of other minds (how do we understand others?) is a problem distinct from the problem of other periods (how do we understand times other than our own?). Why this is so will prove to be an important insight into the nature of historical knowledge and a key datum of this book. There are two overall claims I want to make in this chapter. Understanding others requires that the interpreter possesses the concept of truth and the problem of other minds is a problem separate from the problem of other periods. These claims have to do with what understanding others is and what the problem of other periods consists of, not with methods of understanding others and other periods. To substantiate these overall claims I will make use of several of Donald Davidson’s core ideas.2 Davidson only incidentally pondered the question whether understanding a past belief differs from understanding a contemporary one, but he did not find that these are different activities. It confirms that the problem of other minds is not hindered by something like the problem of other periods.3 I will also discuss Robin Collingwood’s well-known and influential doctrine of the re-enactment of past thought, which no philosopher of history can ignore 2 Davidson’s views have only recently gained attention in the field of philosophy of history. Rüdiger Graf, ‘Interpretation, Truth, and Past Reality. Donald Davidson meets history,’ Rethinking History 7(3) (2003), 387-402, shows the relevance of those views in the context of a discussion of claims made by postmodernists. This is not what I am after in this chapter. Guiseppina D’Oro, ‘Re-enactment and Radical Translation,’ History and Theory (2004), 198-208, misinterprets Davidson’s core ideas. Some of those misinterpretations are carried over in Serge Grigoriev, ‘Beyond Radical Interpretation: Individuality as the Basis of Historical Understanding,’ European Journal of Philosophy, 17(4) (2008), 489-503. I turn to these misinterpretations below. 3 Davidson wrote his dissertation on Plato’s Philebus. I mention this because it makes evident that Davidson was well acquainted with the idea of other periods.

Periods and Other Minds

23

when discussing the understanding of others. Collingwood’s doctrine has been opposed to Davidson’s views, but this, I argue, is unwarranted and rests on a misinterpretation of Davidson’s views. Rather than opposing, I will argue that their views are complementary. The claim that the problem of other minds is a problem separate from the problem of other periods follows from the discussion of the conditions of understanding others. I will return to the subjects discussed in Galileo’s Telescope. A European Story time and again to help the reader grasp the philosophical considerations that other minds and other periods give rise to. I start with a series of abstract concepts – truth, truth-conditions, meaning, and truth value – that are needed to elucidate the idea of understanding others.

2

Truth-conditions, Evidence, and Meaning

I discern three conditions that allow us to understand the beliefs of others: (i) we have to know the truth-conditions of the utterance, sentence, or inscription expressing the belief; (ii) we should assume that the speaker holds his or her beliefs to be true; and (iii) we should be able to relate the belief to other presupposed but unmentioned beliefs. If the conditions of understanding do not obtain, we either have to put in an effort to try to understand the linguistic behaviour in order that those conditions obtain, or there simply is no linguistic behaviour to interpret to begin with. There are several related points that I want to make in this section. Two should be made at the start. First, the conditions of understanding do not imply that the interpreter only understands beliefs that are true. Second, the assumption that speakers hold their beliefs to be true implies that speakers are able to justify their beliefs and offer reasons for them. The meaning of an utterance is, however, given by its truth-conditions rather than by what would count as justification of the belief expressed by that utterance. To know the truth-condition of an utterance is to know what the words the utterance makes use of mean, and to know in what circumstances the utterance is made. If we do not know the language in which the utterance, sentence, or inscription is stated, then interpretation has to start from scratch. In that case, the first condition does not obtain, but we may still observe that the speaker holds particular utterances to be true in particular circumstances. The assumption that the speaker holds his beliefs to be true is an assumption that the interpreter is allowed to make and which is the basis of all interpretation. This assumption does not imply knowledge of the

24 

The Exemplifying Past

language of the speaker. It is why all interpretation is radical interpretation.4 If interpretation has to start from scratch, the task of the interpreter is to ascertain in what circumstances the speaker utters what words for what reason. This is, obviously, not an easy task. The interpreter has to device a theory that explains what circumstances lead to what verbal response, starting with predicates, singular terms, connectives and quantifiers, and subsequently with indexicals, in order to interpret those utterances that are true in one circumstance but false in another. The method followed here is that of assigning truth-conditions to alien utterances.5 The effort 4 Davidson, ‘Radical Interpretation,’ in Davidson, Inquiries into Truth and Interpretation (Oxford: Clarendon Pres, 2001 [1984]), 125-139. A related though different notion is what Quine refers to as radical translation. On the difference, see Davidson, ‘Radical Interpretation,’ 136 n.16 and Davidson, ‘Meaning, Truth, and Evidence,’ in his Truth, Language and History (Oxford: Clarendon Press, 2005), 47-62. The difference is relevant for the issues discussed in this chapter in the following sense. In Quine’s view, an interpreter understands an observation sentence (i.e. a sentence which the speaker and the interpreter would agree upon outright on witnessing the occasion) because he experiences the same stimulus as the speaker. Davidson has the following to say on this: ‘Stimulus synonymy is not of much direct help in translating non-observation sentences, but it does about as well as can be done for observation sentences, and supplies the basis for all translation.’ This idea is clear. If someone utters the words ‘Dit is een afbeelding van Maria’ while pointing at the fresco of the Madonna at the Santa Maria Maggiore, and I have no knowledge of the language used, I may still arrive at an approximately correct translation of the utterance. Davidson’s own view, which he refers to as the distal theory of meaning, is this. ‘The distal theory, on the other hand, depends primarily on shared causes which are salient for speaker and interpreter, learner and teacher. Meanings are shared when identical events, objects or situations cause or would cause assent and dissent’ (‘Meaning, Truth, and Evidence,’ 54). Both the speaker of the utterance ‘Dit is een afbeelding van Maria’ and I would agree that this utterance is true, in the circumstance we are both in, even if I do not know the language that the speaker uses. The question is how Davidson’s distal theory differs from Quine’s proximal theory of meaning and why that difference is relevant. Davidson argues that the distal theory takes truth as primary to meaning whereas the proximal theory takes evidence as primary to meaning (58). He emphasizes that a relevant cause of a belief is salient to both speaker and interpreter, and salience ‘is defined in terms of similarity of response’. He concludes: ‘What makes communication possible is the sharing, inherited and acquired, of similarity of responses. The interpreter’s verbal responses class together or identify the same objects and events that the speaker’s verbal responses class together’ (61). I will return to the distal theory of meaning and this quotation in the body of the text of this chapter. Here I note that the proximal theory is a justificationist theory of meaning since the meaning of the utterance and the belief it expresses depends on the evidence we use to justify that utterance (and evidence need not be and is not identified with sensory stimuli by Quine), while the distal theory may be called a truth-conditional theory of meaning since the meaning of the utterance and the belief it expresses depends on the conditions for it to be true. The latter is Davidson’s position (58). The relevance of the distinction between a justificationist and truth-conditional theory of meaning will become clear in what follows. I will follow Davidson and take truth rather than evidence as primary to meaning. 5 Davidson, ‘Radical Interpretation,’ 136-137.

Periods and Other Minds

25

is made to reason from the second condition of understanding to the first, and this is supported by the third condition. It does so by focusing on the circumstances in which utterances are made rather than by focusing on the meaning of the words uttered (which is obvious since the language is unknown to the interpreter). The former enables the interpreter to decide what beliefs the speaker has, the latter what words are used to express those beliefs. This is relevant in that the method followed cannot guarantee that the interpreter knows for certain what the words uttered mean, but this indeterminacy of interpretation is constrained by the fact that he can know in what circumstances the speaker has what beliefs.6 Historians are usually not confronted with situations in which the language in which an inscription or sentence is stated is unknown to them because they study known natural languages, some of which are no longer spoken. The first and third conditions of understanding need further elaboration. The second condition is rather straightforward. It is part of the concept of beliefs that beliefs are held to be true. I cannot have a belief and not hold that belief to be true. This is why we may assume that the speaker holds his belief to be true. Obviously, holding a belief to be true does not make it true. The question how beliefs are tied to the world is a constant concern in this chapter. The difficulty here is to relate the notion of holding a belief to be true to the notion of the truth of the belief. As a semantic concept, truth, like meaning, is primitive, according to Davidson. Truth, therefore, cannot be defined, for that would imply that there are more basic concepts than truth itself, and there are no such concepts. All we can do is relate truth and meaning to one another, and this we do with the proposition that the meaning of an utterance is given by its truthconditions.7 As to the question how holding a belief to be true is related to the truth of the belief, the following observations suffice. If I believe that s and hold s to be true, I also believe that s was true yesterday and will be true tomorrow. Galileo believed that he saw mountains and valleys on the moon and he held this belief to be true, and this entails that he also held 6 Davidson discusses the indeterminacy of interpretation somewhat differently. Ibid., 153154. Just like radical interpretation is contrasted with Quine’s radical translation, so is the indeterminacy of interpretation to be contrasted with Quine’s indeterminacy of translation. The difference that is relevant in the context of this chapter is that Quine associates his indeterminacy of translation with a justif icationist rather than a truth-conditional theory of meaning. On Quine’s indeterminacy of translation, see Paul Roth, ‘Paradox and Indeterminacy,’ The Journal of Philosophy, 75(7) (1978), 347-367. 7 These two related claims are argued by Davidson in his ‘The Folly of Trying to Define Truth,’ The Journal of Philosophy, 93 (1996), 263-278.

26 

The Exemplifying Past

that before he made his observations, the moon had mountains and valleys, and that after he is dead and gone, the moon will still have mountains and valleys. When we hold a belief to be true, we assume that the truth of the belief is independent from our acquisition of the belief. Beliefs may change over time, but if we have a belief, we take that belief to be true absolutely,8 until something or someone makes us believe otherwise. I now return to the first condition of understanding. This is what we have said. Understanding an utterance by knowing its truth-conditions entails knowledge of the meaning of the words that the utterance makes use of, and knowledge of the circumstances in which the utterance is made. It does not imply that we only understand utterances that are true: it implies that we only understand utterances of which the elements are there on the basis of which we are able to determine whether the utterance is true or not. The latter is not something that we have to do: knowing what the words mean and knowing how the speaker came to utter those words is usually sufficient to determine whether what is said is true or not. It would be a misunderstanding to presume that only utterances that are true have truth-conditions, just as it is a misunderstanding to presume that only utterances that are believable are understandable. Utterances have a truth value: they are either true or false, i.e. there are conditions under which they are true or false; but knowing whether those conditions obtain or not, and therefore, knowing whether the utterance is true or not, is to be distinguished from understanding the utterance by knowing its truth-conditions.9 Understanding an utterance has to do with what the utterance says, not with what we should do to establish whether the utterance is true or not.10 As we read Galileo’s Sidereus Nuncius, the Starry Messenger published in March 1610, and marvel at its images of lunar mountains and valleys and the day-to-day depiction of four of 8 Gottlob Frege was, to my knowledge, the f irst to notice this. See Frege, ‘The Thought: A Logical Inquiry,’ in P. Strawson ed., Philosophical Logic. Oxford Readings in Philosophy (Oxford: Oxford University Press, 1967), 17-38, at 35. 9 It was Frege who first argued that understanding an utterance is knowing its truth-conditions. His second big idea was that only in the context of an utterance (sentence), words refer and mean something (some utterances consist of one word: think of utterances such as ‘halt!’ or ‘help!’). The reference of an utterance (as distinguished from the reference of the words in the utterance) is its truth value, and the truth value of an utterance is the circumstance in which the utterance (or, to be more precise, the thought or belief the utterance expresses) is true or not. See Frege ‘On Sense and Reference’ [1892] in: A.W. Moor ed., Meaning and Reference (Oxford: Oxford University Press, 1993), 23-42. 10 The distinction is Michael Dummett’s, Truth and the Past (New York: Columbia University Press, 2004), 44-55.

Periods and Other Minds

27

Jupiter’s satellites, we may understand what is written and shown, but such understanding does not imply that we ought to make the observations ourselves to establish their truth. I know that the evidence is there, if I would need it, but my understanding of what is said is not hindered at all by not being able to make the observations myself. To be sure, if we disagree with what is said, and want to answer the question whether what is said and what the images allow us to say is true or not, the question arises as to how to establish whether the utterances are true or not, and to Galileo’s frustration, his standard of what counts as evidence for justifying and holding a belief to be true was not shared by many of his peers, of whom he bitterly complained in a letter to Johannes Kepler, that they thought that ‘philosophy [science] is a book like the Aeneid and the Odyssey, and that the truth must be sought not in the real world or nature, but (using their words) by comparing texts.’11 Justification is not only relative to the available evidence, but also to an audience and their standards of justification, i.e. to the type of reasons they believe are appropriate to adopt in a particular circumstance. Truth is not relative that way, for regardless of the different standards of justification invoked by Galileo and his colleagues, the moon either has or has no mountains or valleys. We do not know the truth value of all utterances. We do know, however, of all utterances that we understand, that they have a truth value, even if we do not know what that truth value is.12 If someone for instance tells me that I will like the frescos in the Santa Maria Maggiore, I know that the utterance has a truth value, without knowing what that value is, until I actually visit and inspect the frescos. When, then, do we take an utterance of some speaker to be true? The following seems warranted. We hold an utterance, sentence, or inscription s to be true if we would use s ourselves if we would find ourselves in the circumstance in which s is made, or accept that using such utterance is appropriate and useful. An utterance which we understand but do not hold true, is an utterance which would not be 11 The letter is quoted in Bucciantini, Camerota, and Giudice, Galileo’s Telescope, 102. 12 Do utterances in theatrical performances or in novels have a truth value? If the meaning of an utterance is given by its truth-condition, then clearly, those utterances have truth-conditions. But what about their truth value? We assume that the actor holds the beliefs he expresses to be true in the context of the play, just like we assume that characters in novels hold their beliefs to be true relative to the novel itself, and a plot hole emerges when this is not the case. This shows that Aristotle was right when he claimed in his Poetics that actors on a stage and characters in tragedies and epics imitate the actions and thoughts of human beings. We do not, however, assume that the person playing a role holds the beliefs of the character he plays to be true. It is not the purpose of fiction to be true. Wondering what the truth value of sentences in plays and novels is, therefore, is taking those plays and novels to be something which they are not (see Chapter 3).

28 

The Exemplifying Past

uttered or accepted by us if we would find ourselves in the circumstance in which the utterance is made. This does not entail that in order to accept that some utterance is true we should be in the circumstance that led to that utterance ourselves. I have never visited the Santa Maria Maggiore (or perhaps I did, but I do not recall having visited it), but if I would visit it, I would use the utterance, or would think that such use is appropriate, that I see, among other things, a fresco depicting Mary standing on a moon. Not one moment have I doubted that the statements I read in Galileo’s Telescope on Cigoli’s fresco are true. I know that the evidence is there, if I would need it, but my understanding of what is said about the fresco is not hindered at all by not actually being in the circumstance myself that led to what is said. If we do not know whether there is evidence to establish the truth or falsity of some utterance, we may still understand the utterance, given the conditions of understanding. A ‘statement about the past could be true if someone at the relevant time could have verified it, even though all reason for asserting it may have blown away.’13 Here the point becomes clear that the meaning of an utterance is given by its truth-conditions rather than by what would count as evidence for the belief expressed by that utterance. The distinction between what an utterance says or means and what we need to establish its truth or falsity, is missed by the justificationist about meaning. All of this does not preclude misunderstanding and lying. On the contrary, there is misunderstanding when we know what the words mean but we do not know in what circumstances the speaker came to utter these words (if we would not know what the words mean, those words would not be words but mere sounds, drops of ink, pixels on a screen, or mysterious symbols carved in stone, and interpretation would have to start from scratch). Suppose your companion, with whom you visit the Santa Maria Maggiore, points at Cigoli’s Assumption of the Virgin and says: ‘This is how John saw it.’ Your reply ‘No, this is how Galileo saw it,’ is quite appropriate, given your lack of knowledge 13 Dummett, Truth and the Past, 45. The issue is of importance to Dummett, for he is well known for his anti-realism about the past and a justificationist theory of meaning supports his anti-realism: only the remains present today can be used to justify beliefs about the past; there is no reality of the past outside these justified beliefs. A supporter of the truth-conditional theory is not a realist about meaning, but, for lack of a better term, an anti-anti-realist. In his Thought and Reality (Oxford University Press: Oxford, 2006), Dummett argued that a justificationist need not be an anti-realist about the past, but this I doubt. I suggest that we take his views in Truth and the Past as his f inal say on the matter. As for the term anti-anti-realist, a perhaps more adequate term is ‘everyday realism,’ which is typical of Davidson, and which has little if anything to do with the longstanding realism vs. anti-realism dispute. On this, see Jeff Malpas, ‘On Not Giving Up the World – Davidson and the Grounds of Belief,’ International Journal of Philosophical Studies, 16(2) (2008), 201-215.

Periods and Other Minds

29

and your perplexity as to who John is, but it betrays the misunderstanding. You know what the words your companion used mean, but you do not know how she came to utter those words in that circumstance. If she in response to your reply explains that Cigoli’s instruction was to paint John’s vision of ‘a woman clothed with the sun, and the moon under her feet, and upon her head a crown of twelve stars,’ as it can be found in Chapter 12 of the Book of Revelation,14 the use of the utterance ‘This is how John saw it’ becomes clear to you, and your initial misunderstanding resolves. A second example is this: When someone tells you that she is angry with you and you do not know why (you do know what the words she is using mean, but you do not know how she came to utter those words), then that person is to explain to you how she came to utter those words, and perhaps you are able to make things right by pointing out that the use of the utterance in that circumstance is unwarranted, or you realize that the use of the utterance is warranted, which makes you understand her anger and apologize. These are rather simple examples that show that misunderstanding is not concerned with knowledge of what the words used mean, but with knowledge of how the speaker came to utter those words. This also holds for the occasion when we do not understand what a particular word in a particular utterance means. When the misunderstanding concerns a single word, the speaker will be asked to explain that particular word, and that comes down to explaining how – in what circumstances – he came to use that word in the utterance made.15 But what if the speaker is no longer with us, as is the case when we aim to understand the past? Then we need to re-enact the thought process of the speaker. This we turn to below. What if what the speaker says is not held true by him and he is telling a lie? If the interpreter assumes that the speaker holds his utterance to be true, as the second condition states, then clearly, the utterance is not a lie in the eyes of the interpreter. Knowing that some utterance is a lie implies knowing what the words mean and knowing in what circumstances the words came to be uttered. The latter will make the interpreter realize that the speaker does not hold his utterance to be true and is lying (he is not simply mistaken). Deceiving someone is abusing the habit that interpreters assume that what speakers say is held true by them. Irony and jokes also make use of this habit (without abusing it), as is well understood by the interpreters who get the irony or the joke. 14 Bucciantini, Camerota, and Giudice, Galileo’s Telescope, 221. 15 If none of the words are understood, there is no misunderstanding but no understanding at all and interpretation has to start from scratch.

30 

3

The Exemplifying Past

Understanding False and Irrational Beliefs

We understand an utterance if (i) we know its truth-conditions, (ii) we take the speaker to hold his utterance to be true, and (iii) the belief expressed in the utterance is related to other beliefs. The first two conditions have been discussed, the third one not yet. There may be other conditions that allow for the interpretation of utterances and the beliefs expressed by them, but these three conditions are basic, and therefore I limit myself to them. It seems rather trivial to observe that the belief that is interpreted is related to other beliefs. However, this observation is far from trivial and brings us to the core of understanding false beliefs. From Davidson we learn that we can only qualify a belief as incorrect or false against the background of many correct and true beliefs. Evidently, the meaning of the words used and the circumstances in which those words are uttered (the truth-conditions) are central to such effort. Davidson writes: But of course it cannot be assumed that speakers never have false beliefs. […] To take an example, how clear are we that the ancients – some ancients – believed that the earth was flat? This earth? Well, this earth of ours is part of the solar system, a system partly identified by the fact that it is a gaggle of large, cool, solid bodies circling around a very large, hot star. If someone believes none of this about the earth, is it certain that it is the earth that he is thinking about? An answer is not called for. The point is made if this kind of consideration of related beliefs can shake one’s conf idence that the ancients believed the earth was flat. It isn’t that any one false belief necessarily destroys our ability to identify further beliefs, but that the intelligibility of such identifications must depend on a background of largely unmentioned and unquestioned true beliefs. To put it another way: the more things a believer is right about, the sharper his errors are. Too much mistake simply blurs the focus.16

The statement that the earth to some of the ancients was different from what the earth is to us only makes sense if it means that some of the beliefs that the ancients had about the earth are different from some of the beliefs we have about the earth. Identifying these differences in belief requires a shared background of unquestioned and unmentioned beliefs. When we say that the belief that the earth was flat is incorrect or false, it is far from clear that such statement concerns the belief of some of the ancients. First 16 Davidson, ‘Thought and Talk,’ Inquiries, 155-170, at 168.

Periods and Other Minds

31

we would have to know what correct beliefs these ancients had about the earth, and only then we may be able to qualify some beliefs as false. If we do not know what someone is talking about, there is no ground to qualify his utterance as incorrect and false. It follows that most of our beliefs are true. If that was not so, how would we be able to prove that? There would be nothing to go on. More questions arise. Why would we possess a limited number of truths to determine that most of our beliefs are false? Why then not start with those truths and build upon them and disregard the false ones? The main lesson to be learned here is that identifying a belief as false is possible only against the backdrop of many true beliefs. It follows that the notions of truth and falsity are not of the same order. The latter needs the former as a background if it is to make sense, whereas the reverse does not.17 The same holds for irrational beliefs. A belief can only be qualified as irrational against the backdrop of many rational beliefs. This is how the principle of humanity guides the interpretation of past beliefs.18 It is, I think, a more appropriate term in the context of understanding than Davidson’s own principle of charity, which refers to the assumptions we make – and are allowed to make – as interpreters, and which are suggested by the conditions of understanding.19 The principle of humanity emphasizes that interpreters are to take speakers to be fellow sentient and sapient creatures. Davidson would fully agree with this. In understanding others, we attribute beliefs, desires, and intentions to what they do and say, and these constitute reasons for what they do and say. It does not follow, as is sometimes thought, that in this view, speakers cannot have irrational and incoherent beliefs. Let me explain. I take it that the minimal condition for a belief to be rational is that it is coherent with other beliefs. A particular belief is irrational if the belief is 17 I would like to add a remark on method. When we think about binary oppositions, for example oppositions such as truths and lies or truths and falsehoods, we should always ask ourselves which of the two terms cannot be thought of without the other. If one of the terms cannot be thought of without the other, then the alleged binary terms are not of the same order, and one is the condition of the other. In this sense truth is a condition of lies, whereas the reverse is not. I think that for many binary terms it holds that one is the condition of the other and not the other way around, and much confusion results from taking those opposite terms to be of the same order. Here I am thinking of binary oppositions such as reality–appearance, fact–fiction, good–evil, order–chaos, and diachronic–synchronic. Of each of these terms, the first is a necessary condition of the second. 18 The term is coined by Richard Grandy. I came across it in Bernard Williams, The Sense of the Past. Essays in the History of Philosophy, ed. and intro. by Myles Burnyeat (Princeton: Princeton University Press, 2006), 350. 19 This principle is central to the essays collected in Davidson’s Inquiries.

32 

The Exemplifying Past

incoherent with other beliefs. Now, those who have irrational beliefs usually do not qualify their beliefs as such. They do, however, erroneously relate those irrational beliefs to other beliefs that are rational and coherent. Only against the backdrop of many rational beliefs are we able to identify an irrational belief as such: ‘Too much mistake simply blurs the focus.’ Some examples are helpful here. When the future is read in the intestines of some animal, there is nothing preventing us from waiting to see whether the forecast turns out to be true or not. If someone offers you a cup allegedly carved from the horn of a unicorn after being poisoned, you may still drink from it, out of courtesy or curiosity, even though you know that there are no unicorns, and probably are in hurry to see a doctor to receive an antidote that works. The irrational beliefs in these examples – the future can be read in the intestines of animals and a cup made out of the horn of an unicorn detoxifies – do not hinder the interpreter’s understanding of those beliefs and his response towards them. This is so because the irrational beliefs first become apparent against the backdrop of the many shared and coherent beliefs assumed by both speaker and interpreter: beliefs about animals, intestines, forecasts, cups, poison, and so on. A third and final example is this: If someone in the past refuses to cross mountains because he fears devils, we again may identify the belief as false and irrational against the backdrop of many unmentioned and unquestioned true beliefs: beliefs about mountains, devils, fear, crossing, and so one. I mention this last example, which is from Collingwood,20 because some – notably Guiseppina D’Oro and Serge Grigoriev – think that Davidson’s interpreter would be unable to deal with it.21 Grigoriev, while leaning on D’Oro, writes: ‘Davidson’s interpreter […] has to either dismiss the belief in devils as unintelligible or interpret it as a belief about something else.’22 This misrepresents Davidson’s ideas about interpretation. There is nothing ‘unintelligible’ about the belief in devils and that being a reason to refuse to cross mountains. The three basic conditions of understanding others guarantee that the belief is understood. The point is that the irrational belief that there are devils is only understood relative to a large set of unmentioned related and rational beliefs. The root of this misrepresentation of Davidson’s 20 R.G. Collingwood, The Idea of History (Oxford University Press: Oxford, 1978 [1946]), 317. 21 D’Oro, ‘Re-enactment and Radical Translation,’ 201-202. Central to D’Oro misunderstanding of Davidson is this: ‘Since [Davidson’s] radical translation [sic] rests on truth-conditional semantics, it rules out the possibility that there may be statements that are intelligible even though based on false beliefs.’ Ibid., 198. Davidson does not rule out that there are intelligible statements based on false beliefs. 22 Grigoriev, ‘Beyond Radical Interpretation,’ 491.

Periods and Other Minds

33

ideas is, I think, his claim that interpretation is directed at maximizing agreement.23 How to maximize agreement with someone who refuses to cross mountains because he fears devils? An answer is not called for. The agreement has not to do with this particular belief, but with the unmentioned but related beliefs that allow us to interpret his refusal. The past beliefs studied by historians were held true by those having them. This obviously does not imply that historians too should have those beliefs and hold them to be true. A historian may understand past beliefs, even if he thinks those beliefs are false and unbelievable. Martin Horky, a critic of Galileo, wrote in one of his letters: ‘Galileo is under the threat of Jupiter itself, because when he was born, the planet was in conjunction with the evil Saturn in the Twelfth House.’24 I understand the belief expressed here, I know what the words mean and in what circumstances they were uttered, and I know how to locate it in what system of beliefs, which requires rudimentary philological skill and some basic historical knowledge, but I do not think that the statement is true. We are now in a position to answer the question: What ties language and beliefs to the world? Davidson approaches this question by asking: What makes communication possible? His answer is: ‘What makes communication possible is the sharing, inherited and acquired, of similarity of response. The interpreter’s verbal responses class together or identify the same objects and events that the speaker’s verbal responses class together.’25 For a proper understanding of this answer, and of Davidson’s views on interpretation, we should realize that the similarity of response does not exclude differences of response, for the latter only become apparent in light of the former. This is the argument I have been making throughout this section. The theologians and Galileonians respond differently when seeing Cigoli’s moon or the moon itself, or when reading about the moon – different from one another and different from our responses – and they disagree with one another, and we may disagree with them, but insofar as we are able to understand them and they are able to understand and communicate with another, the differences between them, and them and us, only become apparent against the background of the many shared responses they and we have in the circumstances leading to the different responses – e.g. that it is a fresco we are witnessing, that it is a depiction of the moon, that the moon is a celestial body, seen in different shapes at different but recurrent times, 23 Davidson, ‘Radical Interpretation,’ 137. 24 Bucciantini, Camerota, and Giudice, Galileo’s Telescope, 98. 25 Davidson, ‘Meaning, Truth, and Evidence,’ 61.

34 

The Exemplifying Past

and so on. We have to know, in other words, in what circumstances the theologians and the Galileonians came to have certain beliefs, how their beliefs are related to other beliefs, and how those beliefs form reasons for what it is that they do and say. One wonders how all of this relates to Collingwood’s influential doctrine of the re-enactment of past thought. His doctrine takes us back in the history of philosophy prior to Davidson’s important interventions. In what follows, I will discuss thoughts rather than beliefs. Beliefs are central to thoughts. If I think, notice, read, or recall that Cigoli depicted the moon to pay tribute to Galileo, I also believe that he did. I also believe that he is a painter, a human being, and possibly a friend of Galileo, next to beliefs I have about telescopes, the moon, observation, frescos, and so one. A thought requires a background of beliefs which identifies that thought by ‘locating it in a logical and epistemic space,’ as Davidson puts it.26 Here it already becomes clear how the doctrine of re-enactment relates to Davidson’s views. Trying to rethink a thought by asking questions may makes us aware of the beliefs presupposed by that thought.

4

Re-enactment of Past Thought

Collingwood famously claimed that historical knowledge depends on our ability to rethink the thoughts of past individuals.27 A mere description of past remains as they still exist today – the Sidereus Nuncius, a shopping list on the back of some letter, Galileo’s telescopes on display at the Florentine museum, Cigoli’s fresco – does not give us historical knowledge. To acquire such knowledge, the historian has to know the thoughts ‘in’ or ‘behind’ the remains he studies, and he starts to discern those thoughts by asking questions. Explaining why Cigoli depicted the moon as he did and why he 26 Davidson, ‘Thought and Talk,’ 157. 27 Collingwood, The Idea of History, see in particular pp. 282 ff. The doctrine is extensively discussed in the literature. For the doctrine in the context of Collingwood’s philosophy of history, see especially Jan van der Dussen, History as a Science: Collingwood’s Philosophy of History (Krips Repro: Meppel, 1980). For the doctrine itself, see especially William Dray, History as re-enactment. R.G. Collingwood’s Idea of History (Clarendon Press: Oxford, 1995). For its Italian connection, see Rik Peters, History as Thought and Action. The philosophies of Croce, Gentile, de Ruggiero and Collingwood (Exeter: Imprint Academics, 2013). For the philosophy of Collingwood in general, see the essay on him by Williams in his Sense of the Past, 341-358. Williams notes (341) that Collingwood is virtually obliterated from the history of Oxford philosophy, and that is a ‘genuine injustice’. In circles of philosophers of history and historians, Collingwood probably is the best known Oxford philosopher.

Periods and Other Minds

35

was able to do that requires rethinking the thoughts of him, and of the pontiff and his theologians who commissioned the fresco. To be sure, re-enacting past thoughts is not the only thing that historians do, Collingwood does not suggest that, and one does not find explicit re-enactments by the dozens while reading a historical monograph.28 Let us start with an example of re-enactment. In their book Galileo’s Telescope, the historians Massimo Bucciantini, Michele Camerota, and Franco Giudice, discuss a shopping list on the back of a letter which Galileo received at the end of November 1609. Apparently, Galileo needed, among other things, slippers, an ivory comb for his mistress, and a hat for his son Vincenzo. But soon the historians realized that it was no ordinary shopping list they were reading, for it also included artillery balls, pieces of mirror, rock crystal, a tin organ pipe, iron bowls, and many other such non-ordinary items. They started to ask questions. For what purpose did Galileo make this list? To what problem does this list present a solution? The historians tell us: Galileo was setting up an optical workshop to improve the spyglass.29 Now, such re-enactment of past thought in one’s own mind is not, as Collingwood argues, a matter of copying that past thought in one’s own mind. Nor is it a matter of identifying oneself with the person whose thought one is rethinking – Collingwood is not arguing for the nineteenth-century’s hermeneutic method of empathic understanding or Einfühlen.30 The doctrine of re-enactment is at odds with the idea that thinking is first and foremost a subjective phenomenon, and therefore, it is far removed from something that requires Einfühlen or empathy on the side of the interpreter. Collingwood’s central idea is that thoughts have a subjective and an objective side and only the latter is and can be re-enacted. This is a distinction between 28 In Chapter 3 of Galileo’s Telescope, there are several fine examples of explicit re-enactments. This makes one speculate about how the three historians divided up their work, whether each worked mainly on one of its chapters, and who is most inclined to show his or her re-enactments of past thought in his or her writings. 29 Bucciantini, Camerota, and Giudice, Galileo’s Telescope, 59-62. This is a rather simple case of re-enactment. It is rather curious that of the many theoretical accounts about the re-enactment of past thought that I have read, I only rarely found an author discussing re-enactments of actual historians. 30 Nineteenth-century hermeneutics conceived of understanding – Verstehen and Einfühlen (emphatic understanding) – as a psychological category, whereas twentieth-century hermeneutics – e.g. Collingwood’s re-enactment of past thought and Davidson’s theory of interpretation – conceived of understanding as a semantic category. On this distinction, see Georg Hendrik von Wright, Explanation and Understanding (Ithaca and London: Cornell University Press, 2004 [1971]), 30.

36 

The Exemplifying Past

the act or state of thinking (noesis), which is by definition relative to the time and place of the person having that thought, and the propositional content of that thought (noema), which is not relative to the time and place of the person having it.31 The ingenuity of the distinction between the subjective and objective sides of thoughts becomes clear when we relate it to the distinction between thoughts and sensations. At some point Collingwood writes, We shall never know how the flowers smelt in the garden of Epicurus, or how Nietzsche felt the wind in his hair as he walked on the mountains; we cannot relive the triumph of Archimedes or the bitterness of Marius; but the evidence of what these men thought is in our hands; and in re-creating these thoughts in our own minds by interpretation of that evidence we can know, so far as there is any knowledge, that the thoughts we create were theirs.32

Here, as elsewhere in his posthumous published The Idea of History, Collingwood distinguishes thoughts from sensations such as smells, feelings, and emotions. We may know what people in the past thought if the evidence allows us to re-create their thoughts, but we cannot know how they personally experienced the world (as they experienced it at the moment they were having that experience). To be sure, if we had evidence about the flowers in Epicurus’ garden and how they smelt, and evidence about how Nietzsche felt the wind in his hair, we would be able to have knowledge of the thoughts on those smells and feeling of the wind. Clearly, historians are not only interested in past thoughts, they are also interested in sensations, feelings, smells, and emotions in the past, and there is as much evidence about the latter as about the former.33 31 Van der Dussen, History as a Science, 402, too emphasizes the centrality of this distinction in relation to the doctrine of re-enactment, as do Giuseppina D’Oro and James Connelly in the introduction of a recent special issue on Collingwood’s philosophy of history. D’Oro and Connelly, ‘Collingwood, Scientism and Historicism,’ Journal of the Philosophy of History, 11(3) (2017), 275-288, at 286. 32 Collingwood, Idea of History, 296. 33 Suppose we would know what flowers there were in Epicurus’ garden and we would grow them ourselves, carefully recreating his garden. Would we then not know how the flowers smelt in Epicurus’ garden? Or suppose we would know on what mountains Nietzsche walked at what time of the year, and for how long, and how long his hair was, and we would walk those mountains ourselves, having grown our hair in the right length, would we then not feel how Nietzsche felt the wind in his hair, and know that? No, the smell of flowers and feeling of the wind would be our experiences. These literal re-enactments at best create the illusion of having

Periods and Other Minds

37

The passage quoted is therefore misleading in that it suggests that smells, feelings, and emotions cannot be known, whereas thoughts can. Both, however, can be known. The real opposition is not between sensation and thought, but between the subjective and objective sides of thought, between the act of thinking and the content of the thought.34 We know what past individuals thought, given the evidence we have for that, because of the objective side of thought. We also know what past individuals felt and what other sensations they had, given the evidence we have for that, because of the objective side of the thoughts related to those sensations. Smells, feelings, and emotions are subjective in that a person smells, feels, and has emotions at particular points in time. They are objective inasmuch as that person is able to give content to and communicate what he smelled, felt, and what emotions he had – even if only to himself in the internal dialogue he is entertaining – and these are thoughts, and only as thoughts do smells, feelings, and emotions have a history. All of this should not surprise us. Usually when we attribute sensations, feelings, and emotions to someone on the basis of the available evidence, we do not presume to have those sensations, feelings, and emotions ourselves, even if, in particular cases, we empathize with the person we understand, because we know what it is like to have them given our own life experience. The past experiences of others, as all experiences of others, can only be known, not re-experienced. The first condition of understanding we identified above is this: Understanding an utterance requires that we know the truth-conditions of the utterance: we have to know what the words mean and in what circumstances the person uttering those words came to use those words. Knowing what the words mean constitutes what we may call in Collingwood’s idiom the philological stage of understanding. This stage, Collingwood argues, does not lead to historical understanding of the words uttered. Knowing in what circumstances some person came to utter his words is the re-enactment stage the experience that past individuals had. The tendency in today’s museum to create such experiences is therefore not supported by historical theory. Moreover, it is morally wrong to suppose to have an experience as people in the past had them, for it discards and even erases those experiences and replaces them with experiences of one’s own. 34 The distinction therefore is central for a proper understanding of the doctrine of reenactment. Cf. Dray, History as re-enactment, 123-132, who discusses the issues I am discussing without distinguishing between the objective and subjective sides of thought. Although I agree with Dray’s conclusion that ‘the complaint that it [the doctrine of re-enactment] rules out other forms of human experience, such as perception, appetite, or emotion, needs at any rate to be taken with caution’ (150), the stronger point is that the complaint itself is unwarranted since the real opposition is not between sensations (perception, appetite, or emotion) and thoughts, but between the subjective and objective sides of thought.

38 

The Exemplifying Past

of understanding. Here the thought entertained is placed in the context of the beliefs presupposed by that thought and their reasons for having them. This not only shows how the doctrine of re-enactment agrees with what we have said about understanding others, but it also shows that re-enactment helps to determine in what circumstances the person came to utter his words if it is unclear what that circumstance was. At this point, Davidson’s theory of interpretation is silent, and Collingwood’s doctrine complements his views.35 Davidson assumes that when interpretation has to start from scratch, the interpreter is present and able to assign truth-conditions to the utterances in the specific circumstance the speaker is in, and that, obviously, is not the case when it is unclear what the circumstance was of the past utterances studied by the historian. In that case, re-enacting past thoughts by asking questions is all the historian can do. This being said, our conclusion is that re-enacting past thoughts and assigning truth-conditions to alien utterances are complementary rather than contradictory activities.

5

Conceptual Schemes

The claim that we can understand our fellow sentient sapients in the past on the basis of the three conditions of understanding suggests that all human beings throughout history share one conceptual scheme with which they interpret a common world. Against that view it can be argued that different periods are characterized by different conceptual schemes. In this view, the pontiff and the theologians were a period apart from Galileo and Cigoli because they possessed a different conceptual scheme. They lived, as it were, in different worlds: the one characterized by a geocentric view of the universe and a system of beliefs resting on the authority of church and scripture; the other characterized by a heliocentric view of the universe and a system of beliefs resting on the authority of observation and instruments that extend and enhance the human senses. Cigoli’s moon agrees with the teachings of scripture and church according to one scheme, whereas it agrees with the new science according to another, which is why these men are living in different periods. There thus appears to be a conflict between the understanding of others as we discussed it and the idea of periods being different. The first suggests that there is one universal conceptual scheme 35 This argument can also be found in Grigoriev, ‘Beyond Radical Interpretation,’ 496, but with his conclusion about Davidson I disagree, as I argued.

Periods and Other Minds

39

whereas the second suggests that periods differ inasmuch as conceptual schemes differ. A way to resolve this (apparent) conflict is to reject the very idea of conceptual schemes. If the idea of a conceptual scheme is rejected, there is neither one nor are there different conceptual schemes. This is what Davidson does in his seminal essay ‘On the Very Idea of a Conceptual Scheme’. Rather than living in different worlds, Davidson would say that the theologians and Galileonians are only words apart.36 Below, I will make this claim plausible. Some arguments may appear to be farfetched to some readers, others difficult to grasp, but they will support both the overall claims of this chapter. Understanding others requires the interpreter to possess the concept of truth, and the problem of other minds is a problem distinct from the problem of other periods. Davidson starts his essay by telling us what conceptual schemes supposedly are: Philosophers of many persuasions are prone to talk of conceptual schemes. Conceptual schemes, we are told, are ways of organizing experience; they are systems of categories that give form to the data of sensation; they are points of view from which individuals, cultures, or periods survey the passing scene. There may be no translating from one scheme to another, in which case the beliefs, desires, hopes, and bits of knowledge that characterize one person have no true counterparts for the subscriber to another scheme. Reality itself is relative to a scheme: what counts as real in one system may not in another.37

Davidson rejects the very notion of a conceptual scheme. He rejects both the idea of one conceptual scheme which is shared by every human being in every age, and the idea of multiple conceptual schemes, as in the case when different cultures and periods are characterized by different points of view 36 Davidson, ‘On the Very Idea of a Conceptual Scheme,’ Inquiries, 183-198, at 189. In his essay, Davidson had in particular Thomas Kuhn and his famous idea of incommensurable scientific paradigms in mind. One paradigm or conceptual scheme is incommensurable with the other and therefore, scientists adhering to different paradigms work in different worlds and are unable to communicate with one another. See Kuhn, The Structure of Scientific Revolution (Chicago and London: The University of Chicago Press, 3th edition 1996 [1962]), in particular 119, 150, and 198-220. The idea of living in different worlds is also ubiquitous in the constructivist literature. For this idea and the criticism of it, see Paul Boghossian, Fear of Knowledge. Against Relativism and Constructivism (Oxford: Clarendon Press, 2006), especially the last chapter, where he discusses Kuhn’s Structure of Scientific Revolution. 37 Davidson, ‘Conceptual Scheme,’ 183.

40 

The Exemplifying Past

or worldviews with which they ‘survey the passing scene’. One consequence is that there is no basis for conceptual relativism, for conceptual relativism holds that reality and truth are relative to some conceptual scheme, but if the very idea of a conceptual scheme is rejected, there is nothing for reality and truth to be relative to. Davidson also claims that there is no such thing as ‘uninterpreted reality, something outside all schemes and science.’38 Perhaps a conceptual relativist would immediately object that his position is precisely that there is no such thing as uninterpreted reality, the data of sensation, or the passing scene. He claims after all that reality is relative to some conceptual scheme and what is real according to one culture’s or epoch’s scheme is not in another. It is rather easy to see why this objection fails. When the notion of uninterpreted reality is the necessary counterpart of the notion of conceptual scheme, and the conceptual relativist rejects the former, then he cannot hold on to the latter either, and with that to the idea that reality and truth are relative to it. If we give up on the idea that there is something like uninterpreted reality, the data of sensation, or the passing scene, we also give up on the idea of conceptual schemes and vice versa. The dualism of scheme and empirical content, of an organizing system and something waiting to be organized ‘cannot be made intelligible and defensible.’39 This is what Davidson sets out to make clear in his essay. Davidson criticizes the notions ‘conceptual scheme’ and ‘uninterpreted reality’ by focusing on the terms that substantiate these notions. He argues thus: conceptual schemes (languages) either organize something, or they fit it. […] As for the entities that get organized, or which the scheme must fit, I think again we may detect two main ideas: either it is reality (the universe, the world, nature), or it is experience (the passing show, surface irritations, sensory promptings, sense-data, the given). We cannot attach a clear meaning to the notion of organizing a single object (the world, nature etc.) unless that object is understood to contain or consist in other objects). 40

The notion of ‘organizing’ only applies to multiple objects. When someone says that he organizes or systematizes the world or the past, or brings order 38 Ibid., 198. 39 Ibid., 189. 40 Ibid., 191-192.

Periods and Other Minds

41

to it, or any other such formula, he means that he organizes a particular set of objects. Davidson gives the example of a closet. You do not organize your closet, but you organize the clothing inside. Another example of Davidson is this: We do not organize the ocean by rearranging its islands, but we organize the fish and shellfish by dividing them into groups.41 Similarly, the historian does not organize the past; he organizes kingdoms, events, inventions, artworks and so on. Why is it relevant that the notion of ‘organizing’ only applies to multiple objects? Because it makes clear that we first have to identify those objects before we can start organizing them, and if that is so, then there is no ‘uninterpreted reality’ getting organized. Consequently, there cannot be a dualism of uninterpreted reality and conceptual scheme. If we cannot substantiate the notion of uninterpreted reality, we cannot substantiate the notion of conceptual scheme either, let alone their dualism. There is no such thing as ‘empirical content’ or ‘uninterpreted reality’ waiting to be organized. Next to the notion of ‘organizing’ there is the notion of ‘fitting’ and related notions such as ‘facing’. Davidson rejects the idea that sentences (or sets of sentences such as theories) fit or correspond to reality and are related to one another as conceptual schemes and uninterpreted reality supposedly are. His argument in support of this claim proceeds in two steps. First he observes that the very idea that some conceptual scheme (or theory) ‘fits’ or ‘corresponds’ to reality comes down to the claim that this conceptual scheme (or theory) is true. This is a serious and well-known flaw of the correspondence theory of truth: stating that a sentence s corresponds to reality comes down to stating that s is true and vice versa. It makes correspondence synonymous with truth and therefore incapable of elucidating it. 42 This is the first step of the argument. Conceptual schemes that fit or face reality are true. It follows that the idea of different conceptual schemes implies that these schemes and the sort of sentences they allow are true but untranslatable into one another. There is no such thing as another scheme that is true and translatable, for if one scheme is translatable into another and if what is true according to one scheme is also true according to another, we do not have different schemes but one and the same scheme. The possibility that some other scheme is false is not in question:43 the idea we are investigating 41 Ibid., 192. 42 This is one of the arguments against the correspondence theory of truth. Davidson discusses this and other arguments in his ‘The Structure and Content of Truth,’ Journal of Philosophy, 87(6) (1990), 279-328. 43 It appears that the idea of different conceptual schemes best agrees with the idea that the other scheme is false but translatable: what is false in one scheme may be true in another, as

42 

The Exemplifying Past

here is after all the idea that a conceptual scheme fits or faces reality and that comes down to the idea that the scheme is true. Therefore, our focus is on the idea that some person, culture, or period possesses a different scheme which is true (by their light) but untranslatable into our own scheme. This is what the idea of conceptual schemes supposes. Here we arrive at the second part of the argument. If another scheme is true but untranslatable, then truth and translatability are separable, and, according to Davidson, that is not the case. 44 Therefore, there is no such thing as a different true but untranslatable conceptual scheme. Davidson argues that truth and translatability cannot be separated with the help of Tarski’s Convention T. Here we arrive at the hard part of the argument. Starting point is the trivial observation that the sentence ‘Snow is white’ is true if snow is white. According to Alfred Tarksi, every sentence in a language (or conceptual scheme) L satisfies the condition that s in L is true if and only if p, where we should substitute s with a description of s (‘Snow is white’) and p with s itself (snow is white). The latter (substituting s for p) is something we do in case the language L is the language in which the sentence is made, which in this case is English. Is L however some other language, Dutch for example, then we substitute p with a translation of s: ‘Sneeuw is wit’ is true in Dutch if and only if snow is white. The point is that Tarski’s Convention T (s in L is true if and only if p) makes ‘essential use of the notion of translation into a language we know.’45 Of all sentences ever made and to be made in a language L, we can state that the truth predicate minimally has to satisfy Convention T. This is also the case when we neither speak nor understand L. It follows that the idea that there is some different scheme which is true and untranslatable into a scheme we know is self-contradictory, for it suggests that we know what condition the truth predicate minimally has to satisfy in that language or conceptual scheme L for example Galileo’s sentences on the moon are false in the scheme of the theologians but true according to his own scheme. Translatability is required here, for otherwise we would not be able to say that the other scheme is false. Since the other scheme is simply false in Galileo’s eyes, it follows that schemes only sometimes fit and face reality (it does in the case of Galileo’s scheme, and it fails to do so in the case of the theologians), and therefore, there is only one true scheme rather than there being different schemes (and that one true scheme would probably be the scheme of science). But if there is only one true scheme, we are back at the point we made earlier: it suggests a dualism of conceptual scheme and uninterpreted reality but that dualism cannot be made intelligible. The argument we are making now is that the idea of another conceptual scheme being false is incoherent with the very idea of a conceptual scheme: conceptual schemes are true in that they fit, face, organize or systematize reality. 44 Davidson, ‘Conceptual Scheme,’ 193-195. 45 Ibid., 195.

Periods and Other Minds

43

(we know that s is true in L if and only if p), but, and this is the contradiction, we do not know what p stands for since we cannot translate s. We cannot translate s because we do not speak the language L in which s is made. Not knowing the language L and being unable to translate s is not what is crucial here. Crucial is that when we do not know L and therefore are unable to translate s in L, we also cannot state that s is true in L in accordance with Convention T. The idea of their being a conceptual scheme or language L in which s is true but untranslatable, therefore, is incoherent. We cannot separate truth and translatability, and therefore, we cannot say that some different scheme is true but untranslatable. By rejecting the very idea of conceptual schemes, we no longer can hold that there is one scheme, shared by all human beings throughout history, to interpret uninterpreted reality with. We can also no longer hold that there are different (true but untranslatable) conceptual schemes. Therefore, what separates one period from the other is not the introduction of some other and rival conceptual scheme. If it were a matter of introducing a new scheme, it would imply that the former scheme is false, but this suggestion is incoherent with the idea of conceptual schemes being true for fitting or facing reality. The alternative suggestion would be that since both the former and new schemes are true, truth and reality must be relative to these schemes, as the conceptual relativist would have it: what is real and true to one person or period is not to another. This suggestion, however, depends on the ability to divorce truth and translatability, for when truth and reality are relative to schemes, different schemes would be true and untranslatable into one another; but we cannot, we argued, divorce truth and translatability. With the rejection of conceptual schemes, conceptual relativism no longer makes sense. Truth and reality are not relative to some conceptual scheme. The idea of different conceptual schemes is suggestive of the idea that the theologians and the Galileonians lived in different worlds. If we give up on the very idea of conceptual schemes, this no longer makes sense, but since we can make it clear how the theologians and the Galileonians differ on the earth and the moon in a language they share, these groups are at best words apart.

6

Points of View and Conceptual Relativism

We may reject the very idea of a conceptual scheme, but we do want to make sense of differing points of view, of theologians and Galileonians seeing the

44 

The Exemplifying Past

moon differently, whether they look at it in nature or in the Borghese Chapel at the Santa Maria Maggiore. This is easier said than done. One reason is that the term ‘seeing’ is ambiguous, as is the term ‘view’. Another reason is that points of views are not conceptual schemes and therefore not something with which we survey the passing scene, organize experience, or interpret some sensory input. But what are points of view then? Let us start with the observation that the theologians, the Galileonians, and we have the same stimulus as light hits our retina. They and we will, however, respond to that stimulus differently, not because of the different conceptual schemes with which they and we purportedly interpret that stimulus, but because those stimuli are used as reasons to hold different beliefs to be true. These differences in beliefs only become apparent against the backdrop of the many related and shared but unmentioned beliefs, as we argued earlier in this chapter. This is what different points of views are: different sets of beliefs and our reasons for having them. But is this really that different from the idea of conceptual schemes that we have criticized in the previous section? The problem is, of course, that making sense of differing points of view easily leads to conceptual relativism. Therefore, let us return to Davidson’s criticism of conceptual relativism one more time. He writes: The dominant metaphor of conceptual relativism, that of differing points of view, seems to betray an underlying paradox. Different points of view make sense, but only if there is a common coordinate system on which to plot them; yet the existence of a common system belies the claim of dramatic incomparability. 46

An example is helpful here. According to one point of view, the earth is in the centre of the universe and the celestial bodies circle around it. According to another point of view, the sun is in the centre of our solar system and the earth and the other planets circle around it. Or, according to one point of view the moon is imperfect as scripture and the teachings of the church tell us, whereas according to another point of view, the moon is imperfect because it has mountains and valleys. As long as we talk about different points of view instead of different conceptual schemes, it makes sense to state that what the earth and the moon are to some persons is not what they are to others: sets of beliefs and the reasons associated with them differ and this difference becomes apparent against the backdrop of many shared beliefs. 46 Ibid., 184.

Periods and Other Minds

45

The two rather simple examples show that some common coordinate system is a necessary condition of understanding the differences between the points of view: we have to know that both points of view are concerned with the earth and the moon respectively: they form the common coordinate systems that allow us to even talk of their differences. Now, it makes sense to say that, given the common coordinate system, the points of views are different in that points of views are sets of beliefs and distinguishing between sets of beliefs and acknowledging their difference requires a common coordinate system. However, we cannot say that, given the coordinate system, the conceptual schemes are different, for the ‘existence of a common system belies the claim of dramatic incomparability.’ We cannot say that the schemes are the same either, because, as Davidson drily observes, ‘if we cannot intelligibly say that schemes are different, neither can we intelligibly say that they are one.’47 We also cannot say that reality is relative to the schemes, as the conceptual relativist would have it, for then there would not be a common coordinate system that allows us to make clear that the schemes are different. The paradox of conceptual relativism is that when reality is relative to some scheme, there is no common coordinate system that allows us to make clear that schemes differ. Again it becomes clear that the very idea of conceptual schemes organizing or fitting some uninterpreted reality is incoherent. Someone may respond by saying that I have argued that there is one common coordinate system, namely reality itself: the phenomena we know as the earth and the celestial bodies including the moon that are witnessed by man throughout the ages, and this reality is not relative to some conceptual scheme, as the conceptual relativist erroneously suggests. This shows, as our respondent continues, that reality (or nature, or the world) is independent from our beliefs and that science progresses when it obtains better descriptions of reality. This response would, however, miss the point. It takes us back to a point earlier in our discussion. Our respon­ dent erroneously suggests that there is an uninterpreted reality outside all schemes (and we know that there is no such thing). To be sure, in a causal sense, our respondent has a point when he states that reality is independent from our beliefs. As said, the theologians, the Galileonians, and we have the same stimulus when light hits our retina.48 Our respondent is, however, mistaken in identifying this causal relation with our rational relation with 47 Ibid., 198. 48 One could argue that differences in lenses make for differences in the way light hits the retina, but such difference is of no interest for the argument made here.

46 

The Exemplifying Past

our surroundings. Stimuli, we said, are used as reasons for holding beliefs to be true and not as something in need of ordering, organizing, matching or fitting by some interpretative apparatus. It makes no sense to talk about different schemes. It does, however, make sense to talk about different points of view. Different points of views are different sets of beliefs which only become apparent against the backdrop of many shared but unmentioned and unquestioned beliefs. In other words, different point of views can only be identified as such on the condition that the world is shared, and sharing the world is interpreting the beliefs of others, not having a common uninterpreted reality available. Here we are back at the basic conditions of understanding the beliefs of others with which we started our discussion. The problem of other minds (how do we understand others?) is, I claimed, a problem distinct from the problem of other periods (how do we understand times that are at a distant from our own?). We still have to identify the latter problem and explain why it is distinct from the problem of other minds. Up till now we have only said what the problem of other periods does not amount to: it has not to do with the conditions of understanding others, nor does it have to do with the idea of different conceptual schemes and truth and reality being relative to such scheme.

7

Other Minds and Other Periods

Arthur Danto did not, I think, mean it as a provocation when he wrote that ‘the whole point of history is not to know about actions as witnesses might, but as historians do, in connection with later events and as part of temporal wholes.’49 Supporters of Collingwood no doubt would be provoked. After all, Collingwood claimed that to have historical knowledge, we have to rethink the thoughts of past witnesses. Are we not interested in how Cigoli came to paint the fresco as he did? Or in why Pope Paul V commissioned this fresco? And do we not want to know what steps Galileo took to improve the spyglass, and his reasons for doing so? Danto does not deny that historians often do give descriptions of events that witnesses of the events could have made, but he doubts that such descriptions are the most important ones. As a consequence, the philosopher of history should focus on how historians understand events by connecting them to later events and how 49 Danto, Narration and Knowledge (New York: Columbia University Press, 1985), 183. Danto has never shown any interest in the work of Collingwood.

Periods and Other Minds

47

they place them into a temporal whole, i.e. in a narrative. This confirms that the problem of other minds and the problem of other periods are separate problems. The problem of other periods has to do with connections between events, with how the course of events unfolded, and with the significance of events. Not at the moment they happened, but afterwards, when a period is over and its contours are apparent, and the historian wonders what it was that made that period a period. The problem of other periods is concerned with what other minds could not envision, believe, intend, and know. Danto gives a nice example of what he has in mind when he claims that the whole point of history is not to know about events as witnesses might. Petrarch’s brother witnessed Petrarch’s ascent of Mt Ventoux. Historians might say that when he climbed Mt Ventoux, he opened the Renaissance. But his brother could not have witnessed Petrarch opening the Renaissance. He could hardly have seen the event under that description, not because his senses were defective, but because he could not have understood the description at the time. Not unless he knew what was going to happen in the future, and knew, in addition, what historians were later going to say was the significance of what he saw.50

We may very well understand Petrarch and his reasons for climbing Mont Ventoux – apparently to enjoy its view. But such is not the only thing historians are interested in, and perhaps it is not even the most important thing. The historian is interested in his sense of natural beauty, in his reason for climbing this peak as being modern, and that Petrarch was ‘the first man, perhaps, since the days of antiquity who did so.’51 But, we may wonder, does not this affect what we said about understanding linguistic behaviour? After all, I presented a general theory about the meaning and truth of utterances, and if what Danto says has any bearing on what I said about understanding linguistic behaviour, the problem of other periods and the problem of other minds are related problems, contrary to what I have been arguing. Let us consider how Danto continues the passage I just quoted in order to address this issue. What experiences would verify for him [Petrarch’s brother Gherardo], at that time, the sentence ‘Petrarch is opening the Renaissance’? I would 50 Ibid., 61. 51 Jacob Burckhardt, The Civilization of the Renaissance in Italy, 119.

48 

The Exemplifying Past

hardly dare say. I should like to say that however meaningful such sentence is now, in its appropriate past tense, it would have been on the verge of meaninglessness when the event referred to by it was happening. For strictly speaking, there are no experiences which verify that sentence, if, by verification here, we mean experiencing what the sentence is about under the description given by the sentence. Verifiability, then, is not an adequate criterion of meaningfulness so far as these historical sentences are concerned.52

First we should observe that we did not state that verifiability is the criterion of meaningfulness for utterances. But that is not what is important here. The point here is that there is no evidence whatsoever of statements such as ‘Petrarch opened the Renaissance.’ We should also note that the belief expressed by the sentence ‘Petrarch opened the Renaissance’ has nothing to do with understanding others since it was not a belief that Petrarch or his brother had. The statement has to do with the retrospective view that informs how a historian talks about the past. This reaffirms the claim that the problem of other minds and the problem of other periods are separate problems. However, the conditions of understanding beliefs expressed in utterances apply to all utterances. Those conditions, therefore, also should hold for the historical utterances we are considering here (if the conditions are to hold for all utterances). Recall that in the context of other minds it was stated that we hold an utterance, sentence, or inscription s to be true if we would use that utterance s ourselves if we would find ourselves in the circumstances in which s is or was made, or if we would accept that such use is appropriate and convenient. The utterance is held to be false if we would not use it in that circumstance. It seems that this view is incompatible with the view of Danto in the passage quoted. How are we to apply the general conditions of understanding to the utterance ‘Petrarch opened the Renaissance’? (We know that the utterance would be meaningless to Petrarch and his brother in 1336, given the circumstances they were in.) The answer to the question raised offers a general and important insight into the nature of historical knowledge. My argument is that one of the truth-conditions of the statement ‘Petrarch opened the Renaissance’ consists of the reference to a narrative about the past that gives this statement its meaning. There are, we said, two parts to truth-conditions. To know the truth-condition of some utterance, we have to know what the words the utterance makes 52 Danto, Narration and Knowledge, 61.

Periods and Other Minds

49

use of mean and we have to know in what circumstances those words are uttered. Understanding the utterance ‘Petrarch opened the Renaissance’ requires, besides knowing what the words mean, knowledge of a narrative which gives this utterances its meaning. Knowing in what circumstances some historical utterance is made is knowing what narrative that utterance depends on (and this knowledge is unavailable to past witnesses and their contemporaries). This is how the general conditions of understanding apply to both utterances made by witnesses and their contemporaries in the past and utterances made by historians. The distinction between events under the description of witnesses and their contemporaries and events under the description of historians is central to the rest of this book.

8 Conclusion Galileo and his work exemplify a new vision of mankind and its place in the universe, and according to the historians Bucciantini, Camerota, and Giudice, establishing such a new vision was even Galileo’s intention.53 His Starry Messenger had a huge impact, which is perhaps best illustrated by the anguish and dismay of the poet John Donne, to whom the new astronomy called ‘all in doubt’. It undermined the existing social relations and the order those relations upheld: ‘Prince, subject, father, son, are things forgot.’ According to his poem, a world already complete and orderly had come to an end.54 This is what the Starry Messenger meant to some of Galileo’s contemporaries. This it cannot mean to us, for not only do we live in other times than they did, we also know what happened afterwards, which they could not. Even if Galileo had the intention to change the worldview of his fellow men, he had no knowledge of how the course of events would unfold, and if and how that worldview would come into existence. Such knowledge is only available to historians. With Galileo, one period came to a close and another began. It requires the work of a historian to tell us how this came to be. The concept of truth turned out to be central in this chapter. Understanding others requires that we possess the concept of truth. I wonder: Are historical narratives true? This is the question we turn to in the next chapter.

53 Bucciantini, Camerota, and Giudice, Galileo’s Telescope, 225. 54 Ibid., 146-147.

3

Narrative Truth

1 Introduction Aristotle’s Poetics is somewhat notorious for stating that poetry is more philosophical than history because it is concerned with the universal whereas history is concerned with the particular. Tragedies, Aristotle’s main concern, deal with what types of persons will probably or necessarily do or say, whereas histories deal with what particular persons actually have done or said.1 Aristotle also maintains that tragedy ideally is concerned with the imitation (mimesis) of one action, that is, with one plot that encompasses the beginning, middle, and end of the story, whereas history is concerned with one period, narrating all that has happened during that period, even if the events are not related. Although this distinction explains why history does not display the unity of tragedy,2 it is still a distinction in subject matter, not a distinction in narrative structure. It also explains why Aristotle praises Homer for not taking the entire Trojan War from beginning to end as the subject for his epic. Because Homer chose to narrate a part of it, complemented with related episodes, his epic forms a comprehensible whole.3 This reference to Homer is made after Aristotle’s always neglected remark that most epic poets ‘make the structure like that of history.’4 So after distinguishing tragedy from history in terms of their content, at the risk of reducing history to a chronicle of events, Aristotle observes, without further elaboration, that epic and history have a similar narrative form. This remark need not surprise us; as Aristotle was familiar with the work of Herodotus, how could he think otherwise? It follows that although poetry and history differ in content, they share the same narrative form. If then, we want to discuss the truth of historical narratives we should discuss their content rather than their form. History is about what persons actually have said and done in a certain period, and what historians say about them should be justified on the basis of past remains and the appropriate methods of studying those remains. Who would argue with this? 1 Aristotle, Poetics, 51b1-51b10. 2 Ibid., 59a17-59a29. 3 Ibid., 59a30-59b6. 4 Ibid., 59a29. For this translation, which is absent in several other translations I have consulted, see D.W. Lucas, Aristotle Poetics. Introduction, Commentary and Appendix (Oxford: Clarendon Press, 1980 [1968]), 216. The phrase is also translated thus in the Dutch translation by N. van der Ben and J.M. Bremer (Amsterdam: Athenaeum, Polak & Van Gennep, 1995).

52 

2

The Exemplifying Past

History and Fiction

Louis Mink maintains that history shares its form with fiction and agrees with Aristotle’s def inition of narrative structure as the representation (mimesis) of an action with a beginning, middle, and end.5 We may assume that he also agrees with Aristotle’s distinctions between history and fiction, as does his good friend and colleague Hayden White. He too, with reference to Aristotle, holds that history and fiction differ in content and share the same narrative form.6 Not all historical narratives have a clear beginning, middle, and end. Some are open-ended; others are all middle; but all narratives are emplotted one way or the other.7 5 Mink, ‘The Divergence of History and Sociology in Recent Philosophy of History’ [1973], in Historical Understanding (Ithaca and London: Cornell University Press, 1987), 163-181, at 180-181. 6 H. White, ‘The Fictions of Factual Representation’ [1976], in Tropics of Discourse. Essays in Cultural Criticism (Baltimore and London: The Johns Hopkins University Press, 1978), 121-134, at 121-122. Previously, in his ‘The Historical Text as Literary Artifact’ [1974], in Tropics of Discourse, 81-100, at 98, White wrote: ‘The older distinction between fiction and history, in which fiction is conceived as the representation of the imaginable and history as the representation of the actual, must give place to the recognition that we can only know the actual by contrasting it with or likening it to the imaginable.’ It is not clear why this replaces Aristotle’s distinction, to which White refers; nor do I believe it to be true. It is important that we distinguish the concept of reality and the concept of fiction from their concrete instances. White seems to conflate both. We do not take a concrete instance of fiction for reality and vice versa (we do not, for example, prevent the villain from stealing by rushing on stage, nor do we applaud when we see a grocery store being robbed). We know what these instances are when confronted with them without any need of contrasting reality and fiction. However, and if this is White’s point then he is right, our concept of reality requires knowledge of things that contrast with reality, such as fiction and make-believe. However, I do not see how this is supposed to replace Aristotle’s distinction. Also when White speaks about ‘the content of the form’, his aim, I think, is not to oppose Aristotle’s distinction. White’s overall central claim is that the (literary) plot structures or forms (romance, tragedy, comedy, and satire) that govern history writing entail ‘choices with distinct ideological and even specifically political implications’ (The Content of the Form. Narrative Discourse and Historical Representation (Baltimore and London: The Johns Hopkins University Press, 1987), ix). In this specific sense, a form already has content prior to its actualization. See The Content of the Form, xi. In this chapter, I cannot do justice to this central claim of White. White’s views about the literary form of historical narrative are well known. It can be argued that those views are at best a means to the very different end of underlining the inherent politics of history writing. See for this in particular White, ‘The Politics of Historical Interpretation: Discipline and De-Sublimation,’ Critical Inquiry, 9:1 (1982), 113-137. Herman Paul has convincingly argued that White’s aims were always moral and political rather than epistemological. See Paul, Hayden White. The Historical Imagination (Polity Press: Cambridge, 2011). 7 Or one of the four ways, according to White: as tragedy, romance, satire, or comedy. See White, ‘Interpretation in History’, in Tropics of Discourse, 51-80, at 69-70, and his major work Metahistory. The Historical Imagination in Nineteenth-Century Europe (Baltimore and London: Johns Hopkins University Press, 1973). In his ‘The Structure of Historical Narrative’, in The

Narr ative Truth

53

This is not to say that the comparison between history and f iction (poetry) is unproblematic; far from it, for the distinction in terms of their content reduces history to a chronicle of events, while the acknowledgment of their similarity in form moves history in the direction of fiction. Thus there are good reasons to doubt whether the comparison, if we want to make it, improves our understanding of historical narratives. This is not to suggest that history does not share its form with fiction, for obviously both history and fiction relate actions and events using a narrative structure. It is to say that the nature of history is not in any way affected by it, just as the nature of fiction is not affected by the fact that it shares its form with history. (Sometimes it is believed that important things for our understanding of history follow from the observation that history shares its form with f iction. Strikingly, no one seems to believe the opposite, thinking that important things for our understanding of f iction follow from the observation that it shares its form with history. It is often taken for granted that history borrows its form from fiction; but that I take to be an unwarranted assumption.) Many times we have been told that history provides a true account whereas fiction does not. However, this is not entirely correct. We indeed take histories to be true, even though we doubt some of them and others have turned out to be false, but the whole point of fiction is that it is not supposed to be true or false. The question of its truth simply does not arise, nor is it supposed to. Of course, if the question of the truth of a particular fictional story does arise, we would most likely decide that the story is not true; but the moment the question of its truth arises is the moment that the story is no longer taken as fictional. The point is not that we should focus on the content of the historical narrative if we are to discuss its truth, but that the difference between narratives that purport to be true (and can be doubted and turn out to be false) and narratives that are not supposed to be true (and therefore are neither supposed to be doubted nor turn out to be false) cannot be a difference in content only, for it is at least also a difference in purpose. We may add that discussing truth in terms of the content of the narrative is somewhat unsatisfying in that it suggests that narrative form is Fiction of Narrative. Essays on History, Literature, and Theory, 1957-2007, ed. and introduction by R. Doran (Baltimore: The Johns Hopkins University Press, 2010), 112-125, at 113 and 118, White states that Leopold von Ranke’s History of Germany during the Age of the Reformation has a beginning, middle, and end; Alexis de Tocqueville’s Democracy in America has a beginning and a middle but no ending; and Jacob Burckhardt’s Culture of the Renaissance in Italy only has a middle.

54 

The Exemplifying Past

immaterial to its truth. Is it not so that when historians claim that their narratives are true, that they claim their narratives to be true as a whole, not just their parts? Now, historians may well be mistaken about the correct philosophical underpinning of the word ‘true’, but they cannot be mistaken about what it is that they claim. It is therefore of interest to know what this claim entails and what historians should mean by it. With this, we will move into the direction of pragmatism, away from the vocabulary of literary theory. However, it is not until the conclusion of this chapter that we arrive at a pragmatist understanding of narrative truth. I will not discuss different semantic and pragmatic theories here, but present the minimal outlines of what I take to be an adequate account of narrative truth. The aim of this chapter is to deal with what I will refer to as the riddle of narrative truth. As we will see, pragmatism in the philosophy of language insists that a theory of meaning (semantics) answers to a theory of use (pragmatics).

3

Mink’s Riddle

The problem how to ascertain the truth about the past, or the truths, for there are all too many, is as old as history itself, but until Mink’s work no clear distinction was made between questions concerning the truth of (sets of) statements on the past and questions concerning the truth of historical narratives as a whole. Mink is the first philosopher of history, as far as I know, to formulate the problem of narrative truth. In his ‘Narrative Form as a Cognitive Instrument’ he writes: One can regard any text in direct discourse as a logical conjunction of assertions. The truth-value of the text is then simply a logical function of the truth or falsity of the individual assertions taken separately: the conjunction is true if and only if each of the individual propositions is true. Narrative has in fact been analyzed, especially by philosophers intent on comparing the form of narrative with the form of theories, as if it were nothing but a logical conjunction of past referring statements; and on such an analysis there is no problem of narrative truth. The difficulty with the model of logical conjunction, however, is that it is not a model of narrative form at all. It is rather a model of chronicle.8 8 L.O. Mink, ‘Narrative Form as a Cognitive Instrument’ (1978), in Historical Understanding, 182-203, at 197-198.

Narr ative Truth

55

Indeed, if the narrative is not simply a conjunction of statements on the past, then its truth cannot be a function of the truth of its individual statements.9 The opposition between chronicles and narratives invoked here by Mink cannot, however, sufficiently make clear why the narrative is not a conjunction of statements, for there are conjunctions of statements that are not chronicles. All pieces of discourse are conjunctions of statements that are related by anaphoric references. Such cross-textual referencing is precisely what makes it possible to identify the conjunction as a piece of discourse rather than as a random list of unrelated statements. Therefore we not only have to distinguish narratives from chronicles; we also have to distinguish narratives from other pieces of discourse. The notion of plot may be of help here, for only a narrative has a plot with a beginning, middle, and ending.10 The question then is whether the notion of plot can support a satisfactory account of narrative truth. Someone may argue as follows. A statement on the past is true if what the statement says actually occurred and if it is justified by the available evidence and the appropriate methods of studying it,11 whereas a narrative as a whole is true if it is coherent. Narrative truth then functions on the level of the narrative itself, its linguistic space so to speak, whereas the truth of its statements depends on the past and its remains. This seems promising, for it posits narratives as linguistic entities distinct from statements, in conformity with Mink. It is also in line with the notion of plot, for it is the plot that establishes narrative coherence. However, if we identify narrative truth with narrative coherence, then we cannot but conclude 9 This argument is also made by F.R. Ankersmit in his Narrative Logic. A Semantic Analysis of the Historian’s Language (The Hague: Martinus Nijhoff Publishers, 1983), 58-61. 10 Non-narrative pieces of discourse and chronicles do have beginnings and endings, but not a beginning initiating the plot and an ending bringing closure to it. Some narratives are all middle, as said earlier, but such narratives still have a plot, which non-narrative pieces of discourse and chronicles lack. 11 To be sure, there is a difference between taking a statement to be true with reference to reality and taking a statement to be true with reference to evidence, and the correct philosophical semantics involved here is rather complex. The philosophical crucial issue is, I think, whether we opt for a truth-conditional (pragmatist) theory of meaning or a justificationist theory of meaning. This issue is central in Michael Dummett’s Dewey Lectures published as Truth and the Past (New York: Columbia University Press, 2004). There, contrary to some of his other writings and rather hesitantly, he opts for the truth-conditional theory of meaning, and rightly so. The justificationist theory of meaning collapses the difference between what a statement means or says and what is needed to establish it as true, whereas the truth-conditional theory of meaning does not, carefully distinguishing between these two. This I extensively discussed in the previous chapter. We may add that the truth-conditional theory of meaning is a pragmatist theory inasmuch as it explains meaning in terms of use.

56 

The Exemplifying Past

that history and fiction are both true in the sense of being coherent, and that would be a rather awkward conclusion, for it robs the word ‘truth’ of its very meaning. The notions of narrative coherence and plot thus return us whence we came: since history and fiction share their form, the relevant question of the truth of history is to be found on the level of its content, since the truth of its content depends on the past or what we may infer from its remains. All of this, however, is not what Mink has in mind, although, as said, he holds that history and fiction share their narrative form. We seem to be dealing with a riddle rather than a problem here. Mink writes: narrative form in history, as in f iction, is an artif ice, the product of individual imagination. Yet at the same time it is accepted as claiming truth – that is, as representing a real ensemble of interrelationships in past actuality. Nor can we say that narrative form is like a hypothesis in science, which is the product of individual imagination but once suggested leads to research that can confirm or disconfirm it. The crucial difference is that the narrative combination of relations is simply not subject to confirmation or disconfirmation, as any one of them taken separately might be.12

This, then, is the riddle of narrative truth: historical narratives are, as in fiction, products of individual imagination, and they represent, as fiction does not, ‘a real ensemble of interrelationships in past actuality’ and thus are claimed to be true. For a proper understanding of this problem we need to distinguish between relations between events and the narrative combination or ensemble of these relations. Narratives contain indefinitely many ordering relations and all statements asserting relations between events must be justified, if they are to be true.13 These narrative relations do not seem to cause a particular problem since, as we could argue, each statement or set of statements asserting a relation between events is to be evaluated in the same way as any other statement in the narrative is to be evaluated. Stated this way, a narrative is a conjunction of statements. Its statements not only assert what happened 12 Mink, ‘Narrative Form,’ 199. 13 Arthur Danto’s ‘narrative sentences’ are typical of such relations. Narrative sentences relate two temporally separated events by describing the first event referred to in terms of the second, later event. See Danto, Narration and Knowledge (New York: Columbia University Press, 1985), 143-181. Think of such sentences as ‘The Thirty Years War began in 1618’. In Chapter 5 I have more to say about these narrative sentences.

Narr ative Truth

57

in the past, keeping in mind their chronological order, as chronicles do; they also assert what relations obtain between those events other than their chronological order, as chronicles do not do. Clearly this is not what Mink has in mind, for narratives are not a conjunction of assertions. The point is that although each separate relation between events is subject to confirmation and disconfirmation, the combination of interrelations is not, even though such combination of interrelations represents a real combination in past reality and is claimed to be true. It is this difference which explains why the narrative is not to be taken as a conjunction of statements and I take it to be the central datum of narrativist philosophy of history.14 Three and a half decades after Mink formulated the problem of narrative truth it has not been dealt with in a satisfactory manner. Now, at this moment we do not know why Mink believes that statements in historical narratives – including those asserting a relation between events – can be conf irmed or disconf irmed, whereas the combination of interrelations cannot, and we will see that much, if not all, depends on a correct understanding of it. What we do know is that it has to do with the distinction between content and form and the contention that history and fiction differ in content while they share their form, and this goes back to Aristotle’s Poetics. Furthermore it is clear that the notion of combination is crucial, which is not to be identified with the notion of plot. For if the narrative represents a real combination of interrelations between events in past reality, even though that combination cannot be confirmed or disconfirmed, then it cannot be identified with the notion of plot, for the plot belongs solely to the level of the narrative. It thus seems that the vocabulary of literary theory is of no help in understanding the problem of narrative truth. Although Mink clearly stated what the problem of narrative truth is, and no one has done a better job in doing that, he neither solves nor dissolves the problem he posed. That will be our task in this chapter. This will move us away from the vocabulary of literary theory into the direction of pragmatism. Truth is a term that belongs to our semantic vocabulary. Separating our literary, semantic, and pragmatist vocabularies, as we will see, is an important step towards coming to grips with narrative truth. 14 It has been argued that there is simply no difference at all, or only a difference in degree rather than in kind, between statements on the past and whole narratives. See Chris Lorenz, ‘Can Histories by True? Narrativism, Positivism, and the “Metaphorical Turn,”’ History and Theory, 37(3) (1998), 309-329. One of my aims in this chapter is to show that Mink is right in distinguishing between (conjunctions of) statements and narratives.

58 

4

The Exemplifying Past

Narratives as Literary Artifacts and as Cognitive Instruments

Mink writes in the passage quoted above that ‘narrative form in history, as in fiction, is an artifice.’ White even calls these ‘literary artifacts’ ‘verbal fictions’.15 Although historians and writers of fiction are interested in different kinds of events, so White tells us, ‘the techniques or strategies that they use in the composition of their discourses can be shown to be substantially the same.’16 The writing of history is therefore according to White a literary and fiction-making activity.17 This line of thinking however need not be followed, for instead of emphasizing that the historical narrative is a literary artifact, we could emphasize that it is a cognitive instrument with which we make ‘the flux of experience comprehensible,’ as Mink also does.18 He writes: It is the narrative history itself which claims to be a contribution to knowledge, not something else which the narrative history merely popularizes or organizes. The claim of a narrative history is that its structure is a contribution to knowledge, not just a literary artifice for the presentation of a series of factual descriptions.19

According to Mink, and this seems to oppose the views of White, the narrative is ‘a way of thinking rather than merely a literary device employed for arbitrary or traditional reasons.’20 Rather than emphasizing that history shares its structure (form) with fiction, we should emphasize that 15 White, ‘The Historical Text as Literary Artifact,’ 82. He adds: ‘the contents of which are as much invented as found and the forms of which have more in common with their counterparts in literature than they have with those in the sciences.’ White means, if I understand him correctly, that since ‘events are made into a story’ (84), which is ‘essentially a literary, that is to say f iction-making, operation’ (85), the found events of the past are turned into invented elements of a plot. 16 White ‘The Fictions of Factual Representation,’ 121. 17 White ‘The Historical Text as Literary Artifact,’ 85. 18 Mink, ‘Narrative Form,’ 185. 19 Mink, ‘The Divergence of History and Sociology,’ 168. 20 Ibid., 176. The plot structures romance, tragedy, comedy, and satire are transmitted by the Western tradition. Since the same event in White’s view may be emplotted differently, there is a certain arbitrariness of the plot with regard to the past. One could argue that these plot structures, or more precisely, the tropes that White associates with those structures (metaphor, metonymy, synecdoche, irony respectively), are also ways of thinking. I would happily agree with that, but that is not the issue here. The point is that different things follow from and are presupposed by taking historical narratives to be literary artifacts and taking them to be cognitive instruments. For White’s tropes as ‘ways of thinking,’ see F. Ankersmit, ‘White’s “New

Narr ative Truth

59

its structure is different from theory and generalization, and the crucial difference, according to Mink, is that the ensemble of relationships cannot be confirmed or disconfirmed as can hypotheses in science or (sets of) individual statements in narratives, as we noted above. Why this is so still needs to be explained. The difference in emphasis however should be clear. The distinction between the narrative as a literary device and the narrative as a cognitive instrument is a distinction between the narrative as resulting from the literary act of composition, of what White refers to as the ‘making [of] stories out of mere chronicles,’21 and the narrative as resulting from the retroactive alignment of events. This distinction may seem negligible, merely a matter of choice of words, for both composition and retroactive alignment, so one could argue, emphasize, amongst other things, that past events acquire historical significance by being related to later events, and this significance only becomes apparent in a story, as Arthur Danto argued.22 In terms of the notion of plot we may formulate this as follows. The chronologically first and last events of narratives are not to be identified with their beginning and end. Narratives may begin in the middle, followed by how it all came to be and how it all turned out. The beginning of a narrative initiates the plot and the ending brings closure to it. This is not to deny that historians should get the chronology straight. The point is that the understanding of a succession of events is not arrived at by following their chronological order: we understand a succession of events via the plot or combination of interrelations of the narrative. However, there is a distinction between taking narratives to be literary artifacts and taking narratives to be cognitive instruments. Plots are composed whereas the narrative combination of interrelations results from retrospective understanding, and this retrospective understanding already has a narrative structure inasmuch as it is a comprehensive ensemble of aligned events. The notion of composition as defined by White makes use of the chronicle-story distinction, whereas the notion of retroactive alignment does not (or not necessarily). This is important because the distinction between chronicle and story is modelled on the distinction between content Neo-Kantianism”: Aesthetics, Ethics, and Politics,’ in F. Ankersmit, E. Domańska, and Hans Kellner eds., Re-Figuring Hayden White (Stanford: Stanford University Press, 2009), 34-53. 21 White, ‘The Historical Text as Literary Artifact,’ 83. 22 Danto, Narration and Knowledge, 11. It is the central theme of his book. Mink puts it thus: ‘The cognitive function of narrative form, then, is not just to relate a succession of events but to body forth an ensemble of interrelationships of many different kinds as a single whole.’ Mink, ‘Narrative Form,’ 198.

60 

The Exemplifying Past

and form on the basis of which we distinguish history and fiction and argue for their similarity. The content of history is taken to be a chronicle of events given form by the story, and this obliges us to discuss truth in terms of the content of the narrative only, leaving no room for narrative truth. Now let us elaborate on the distinction between narratives as literary artifacts and narratives as cognitive instruments. Rather than taking developments in narratives as the unfolding of a plot with a beginning, middle, and end, we could maintain that the object of all historical understanding is social change – the prototypical social changes being the Renaissance, the Reformation, the French Revolution, and Industrialization. All history is concerned with processes of change in the societies that individuals have found themselves to be living in. Now, social changes begin and end too. The beginning and ending of these social changes, however, are not poetic inventions, as are the beginning and ending of plots;23 they are the result of retrospective understanding. The retroactively aligned events that individuals brought about or underwent illustrate or exemplify the social change that the historical narrative is about.24 These exemplified social changes in their entirety are historical conclusions or theses on the past: they are what Mink refers to as the combination of interrelations between events. Rather than contributing to the development of the plot, what actual persons do and say contributes in retrospect to the social change they exemplify, even if they did not intend to bring that change about. An example is illuminative here. When Margaret Jacob wrote the history of freemasonry, she did not simply relate the Masonic lodges to a chronicle of events; freemasonry is a phenomenon that is first understood by writing its history. The social change that Jacob wrote about basically is the eighteenth-century Western dream of a different social order based on merit instead of birth and wealth. The lodges, their members, what they did and went through, all exemplify this social change. Throughout her wonderful book, therefore, she constantly emphasizes the governmental nature of the lodges. The development towards a more modern society is the historical thesis of her work: that is the social change which her book is about and with 23 White believes, mistakenly I think, that the beginnings and endings of historical narratives are poetic inventions. See White, ‘The Historical Text as Literary Artifact,’ 98. Elsewhere White too emphasized that historical narratives are about processes of change. Some narratives have a sequential or diachronic structure while others have a conditional synchronic structure, and some narratives are about changes-in-continuities while others are about continuity-in-changes. See White, ‘The Structure,’ 115. 24 This is further argued in Chapter 5.

Narr ative Truth

61

which she understands freemasonry historically.25 So rather than taking the moral of the story to be the central message of historical narratives, as White does, we should take the historical thesis to be its central message, as Mink does.26 Again we distinguish between narratives as literary artifacts and narratives as cognitive instruments and underline that it is the latter that enables us to understand the nature of historical narratives. If we emphasize that the historical narrative is the result of the historian’s retrospective understanding of social change, then the uniqueness of single past actions and events is replaced by the uniqueness of the single narrative as a whole, as Mink observes.27 Nonetheless, it remains the case that history is about what actual persons have done or said, whereas fiction is about what types of persons would do or say. However, it does not follow, as Aristotle maintained, that history is about the particular and fiction (poetry) is about the universal. The reason is this: If the action of a particular person mentioned in a narrative illustrates or exemplifies the social change that the narrative is about, then that person is, as an individual, a particular; but as an individual whose action exemplifies social change, that person is a historical type. Nothing is an example in itself and to be an example is to be something particular and general at the same time. What exemplifies and what is exemplified is mutually dependent. We should add further that exemplifying a historical thesis is different from being an instance of a type, for a person’s action only exemplifies a historical thesis relative to a particular narrative, whereas the instances of character types are found across many narratives. Historical narratives define historical types, whereas fictional narratives make use of types. So rather than taking what persons actually have said and done to be unique, the historical thesis those actions exemplify is unique, for it signals a unique social change in the history of mankind. If a similar thesis is exemplified in another narrative, then those narratives may be said to intersect, and it may even be possible to weave those narratives together. However, we should not forget that Jacob 25 Margaret C. Jacob, The Origins of Freemasonry. Facts & Fictions (Philadelphia: University of Pennsylvania Press, 2006). 26 The moral of the story is central in White’s ‘The Value of Narrativity in the Representation of Reality,’ The Content of the Form, 1-25, especially 21-25. There (21) White writes such things as ‘The demand for closure in the historical story is a demand, I suggest, for moral meaning, a demand that sequences of real events be assessed as to their significance as elements of a moral drama.’ Ankersmit too, takes the historical thesis to be the narrative’s central message. In his Narrative Logic he uses the term ‘narrative substance’ for such a historical thesis. I discuss the nature of these narrative substances in Chapter 5. 27 Mink, ‘The Divergence of History and Sociology,’ 180.

62 

The Exemplifying Past

understands Masonic lodges historically, and these lodges exemplify the historical thesis of her work. What exemplifies and what is exemplified is mutually dependent. Therefore there is no such thing as a detachable general historical thesis that is exemplified in different narratives. At best there are related or intersecting historical theses on for instance the development of Western societies towards modernity. What exemplifies a thesis in one narrative may become part of the thesis proposed in another narrative, but what exemplifies and what is exemplified is mutually dependent. With all of this, we distance ourselves from the vocabulary of literary theory. Moreover, the distinction between content and form on the basis of which we distinguish history from fiction (they have different contents) and argue for their similarity (they share the same narrative form) may itself turn out to be a rather unfortunate distinction to begin with. To this I will now turn.

5

The Indeterminate Past

The distinction between content and form suggests that the past is determined and structured by the narrative. The amorphous and unstructured past provides the content which is subsequently given form or structured by the narrative. The undeniable truth that we cannot directly observe the past is not a complicating factor here, for even though the past is gone for good and only its remains are here for our inspection, we may still believe that what we can infer from those remains is the past as it was: a chronicle of events that is as complete as the remains and our methods of inference allow it to be. If we get the historical record straight by making the right inferences, then we have the elements that need to be structured by means of a narrative. The distinction between content and form gives rise to this suggestion. Think once again of White’s definition of composition as the making of stories out of mere chronicles. This picture of an amorphous and unstructured given past providing the content on the one hand, and narrative form and structure on the other is, however, mistaken. We begin to understand why this is so if we realize that there are no rules that enable the historian to translate or project the past in a narrative medium. To be sure, there are many rules for telling historical stories – be consistent, make sure to have the plot unravel in a surprising way even if the end is known, as is usually the case with history, annotate as required, do not quote out of context, and so one. A rule tells us how something ought to be done and we either follow the rule or depart

Narr ative Truth

63

from it. Mink, as does his fellow narrativist philosopher Frank Ankersmit, does not believe that there are translation rules that enable the historian to project the past in a narrative medium: When it comes to the narrative treatment of an ensemble of relationships, we credit the imagination or the sensibility or the insight of the individual historian. This must be so, since there are no rules for the construction of a narrative as there are for the analysis and interpretation of evidence.28

This observation is itself not a rejection of the picture of an amorphous and unstructured past on the one hand and narrative form and structure on the other, for one may still hold that, even if there are no rules to project the former into the latter, it is the narrative that structures or gives form to the past. This, then, would be the much rehearsed argument that narrative structures are constructions of the historian that cannot be found in the past since the past lacks such narrative structures. Although this is certainly a claim with which I agree, it upholds and even strengthens the distinction between content and form, between an amorphous and unstructured past on the one hand and narrative form and structure on the other, a distinction I believe to be misleading if not mistaken. The point is this: Mink not only rejects translation or projection rules; he also rejects the idea that the past is given and determined: the past is no ‘determinate realm of unchanging reality.’29 If the past is not given and determined, if there is no content out there, then one cannot maintain that this content is amorphous and unstructured and given form or structured by the narrative. Thus the idea of an amorphous and unstructured past is rejected, and with it, the distinction between content and form collapses. Mink argues that we should abandon ‘the idea that there is a determinate historical actuality, the complex referent for all our narratives of “what actually happened”, the untold story to which narrative histories approximate.’ To be sure, as Mink maintains: it does not imply that there is nothing determinate about the past, since individual statements of fact, of the sort to which so much historical research 28 Mink, ‘Narrative Form,’ 199. Ankersmit rejects translation rules in his Narrative Logic, 76, and also in combination with a rejection of a given past. This combination is essential, as we will see. Ankersmit rightly points out that White’s tropes are suggestive of such ‘translation rules’ and therefore he rejects them. See his Narrative Logic, 78-79. 29 Mink, ‘Narrative Form,’ 194.

64 

The Exemplifying Past

is dedicated, remain unaffected. But it does mean that the significance of the past is determinate only by virtue of our own disciplined imagination. Insofar as the significance of past occurrences is understandable only as they are locatable in the ensemble of interrelationships that can be grasped only in the construction of narrative form, it is we who make the past determinate in that respect.30

The indeterminate past thus is determinate in retrospect, i.e. as a result of and in relation to some historical narrative. Narratives, as David Weberman puts it, as such have ontological force.31 The argument may still be said to lead to the conclusion that the past provides the content which is given form or structured by the narrative in that individual statements of fact remain unaffected. We therefore have to expand our argument. I will elaborate on the passage in Mink’s essay just quoted by distinguishing between two issues. The first concerns the distinction between historical research and historical writing; the second concerns the historical event as being significant in the context of a story only. Both points will make it clear why the distinction between content and form is misleading insofar as it is suggestive of a determined past providing the content that is structured or given form by the narrative. However, we should reject the idea of a determinate realm of unchanging reality providing the content of history. Rejecting this idea also has us reject the idea of the narrative giving form to and structuring this unchanging, amorphous, and unstructured reality. We cannot reject the one while holding on to the other. 1. It should be noted that Mink’s distinction between research and writing in the passage quoted above is not modelled on the distinction between content and form. This is important because the distinction between content and form suggests the idea that the past is a determined unchanging reality providing the content which is subsequently structured (or given form) by the narrative, while Mink rejects the determinateness of the past. There is a sense in which we should reject the distinction between historical research and historical writing and a sense in which we are allowed to make that distinction, and it is illuminating to understand why this is so. One can reason thus. The distinction between historical research and historical writing is to be modelled on the distinction between content and form, and this helps us to distinguish history from fiction in terms of their 30 Ibid., 202. 31 David Weberman, ‘The Nonfixity of the Historical Past,’ The Review of Metaphysics 50(4) (1997), 749-768, at 753.

Narr ative Truth

65

content and argue for their similarity in terms of their form. The phase of historical research, a phase which fiction lacks (even if we admit that literary writers do research, they still cannot be said to do historical research), is complemented with a literary phase – the phase of composition. It follows that problems in the philosophy of history can be divided into problems concerning historical research and problems concerning historical writing. The former is about narrative content, concerning itself with interpretation and explanation resulting in justified statements about the past, whereas the latter is about narrative form, concerning itself with composition, i.e. the making of a story out of a mere chronicle. This distinction between a research and a composition phase however is misleading, if not simply mistaken. Although source criticism, methods of analysis, and explanation are activities distinct from selection and retroactive alignment, the point is that history writing is not a matter of presenting research results. The historical narrative, after all, is a way of thinking, a cognitive instrument, and it contributes to our knowledge of the past by virtue of its structure, as Mink maintained. This is where the writing of history differs from the sciences, where writing does come down to the presentation of research results, and why distinguishing between a research phase and a composition phase is misleading, for that distinction suggests that history writing, understood as the literary activity of composition, is a matter of presenting research results, albeit in a literary manner. This is not to say that we cannot make a distinction between historical research and history writing. We may make a distinction between research as the establishment of fact – straightforwardly asserting what there is, thereby making up an inventory of the world – and writing as the establishment of a historical thesis, as Mink does, between source criticism and methods of analysis on the one hand, and selection, retroactive alignment, and comprehensive understanding on the other; but there are no historical theses independent of historical narratives. There is not first a research phase leading to a historical thesis and subsequently a composition phase presenting that thesis. Modelling the distinction between historical research and historical writing on the distinction between content and form therefore is mistaken. History’s content and form are of a piece. We now come to understand what has been misleading us all along. The distinction between content and form assumes that the content of a narrative consists of what particular persons actually have said and done: the events they brought about and went through. If we take content to be this, then we are likely to take the content of the narrative to be a chronicle of actions and events and the form of the narrative to be the story made

66 

The Exemplifying Past

out of this chronicle, to use White’s unfortunate phrase again. What, then, should we take for the content of the narrative? I believe Ankersmit is right when he states that all sentences together determine what the content of a particular narrative is, though some sentences will be more important than others.32 The narrative is not a conjunction of statements, as Mink maintained, but establishes a combination of interrelations between events. If history’s content and form are of a piece, then we can no longer hold that a determined realm of an unchanging past reality provides the content which is given form by the narrative. And it can no longer come as a surprise that historians claim their narratives to be true as a whole; after all, history’s content and form are of a piece. 2. With the emphasis on the significance of the past in the passage quoted above, Mink follows Danto, who, as we noted, argued that past events acquire historical significance by means of retroactive alignment. The significance of an event can only be grasped in the context of a story, i.e. in its relation to later events. This is a clear idea. However, it can lead to a misunderstanding. According to Danto, the same event has a different significance in accor­ dance with the story in which it is located. What is problematic about this, and only if we misunderstand Danto, is that the idea that we can tell different stories about the same event suggests the idea that the determinate past is out there, for in what other sense could we speak of the same event about which different stories are told? Now, two different event descriptions may refer to the same event if the event referred to happened to have occurred at the same time and place. A singular term (or a definite description used as a singular term) may be said to refer only if the concrete particular referred to is identified, i.e. if the hearer or reader knows which or what object is in question. If this test of identification fails, then reference fails, and there is no object to which to apply the general term.33 But here we are not interested in discussing the identification of events and what is predicated of them; at issue is the significance of events in terms of what they exemplify. By contrast, if we want to know what historical event is in question, i.e. if we want to know the historical significance of an event, we have to know what that event exemplifies, and to know that, we have to read the entire narrative, for only then is it clear to us what historical thesis the selected 32 Ankersmit, Narrative Logic, 61. See also Chapter 5 for an elaboration of this claim. 33 P.F. Strawson, ‘Singular Terms and Predication’ [1961], in Strawson ed., Philosophical Logic (Oxford: Oxford University Press, 1973 [1967]), 69-88, at 74. This is not to say that the object of reference and its properties are given. It is to say that the object of reference is located in a logical and epistemic space. This means that we can endorse what is asserted for ourselves, dispute it, go and look for it if we want to, be amazed about it, and so on.

Narr ative Truth

67

and retroactively aligned events exemplify. Now let us explain further why the idea of telling different stories about the same event is misleading. Suppose we have a particular set of events, then, one could argue, we could represent those events either in a chronicle, listing the events in their chronological order, or in a narrative, which may make use of a different order, for example by starting in the middle, followed by how it all came to be and how it all turned out. This picture is misleading inasmuch as it suggests that the same set of events can be either ordered chronologically or narratively. However, we do not order events with either the chronicle or the narrative: we list event descriptions in the case of the chronicle and narrate events in the case of the narrative. Now, one could argue that narrating events simply comes down to listing event descriptions, albeit with an extra set of cross-referencing terms linking those descriptions. One may further argue that one could extract a chronicle from a narrative, just as one could narrate the events listed in the chronicle. It follows, if these arguments are correct, that events or event descriptions are organized by either chronicles or narratives. However, this is a mistake, as Danto argues. For the chronicle cannot make use of terms that relate events to later events – terms such as anticipates, began, and ended – whereas those terms are typical of the event descriptions used in narratives.34 Such terms are not simply used for cross-referencing in the text; they refer to relations in the past itself. It is therefore a mistake to think that we organize events in either the chronicle or the narrative, and it is a mistake to think that the events in chronicles and narratives are interchangeable, even if they contain the same event name. Although a chronicle and a narrative may mention the same event, even its description may be graphically or sonically identical, containing the same marks in ink or pixels or producing the same noises in the same order; the truth-conditions, hence the meaning of these statements will differ, for as part of the truth-conditions of statements in a narrative there belongs a reference to the historical thesis of that narrative, and such reference obviously is lacking in a chronicle consisting of event descriptions. We now see what is misleading about the distinction between history and fiction in terms of their content, while acknowledging their similarity of form. It assumes that the content of history is a chronicle of events which is given form by the narrative. It suggests that on the one hand there are events or event descriptions, and on the other narratives of those events. From this it follows that the same event can be told differently by different 34 Danto, Narration and Knowledge, 353-356.

68 

The Exemplifying Past

historians. We now know that we should give up this idea. What counts as the same event is precisely what is disputed in each single narrative.35 Mink puts it thus: ‘Events’ (or more precisely, descriptions of events) are not the raw material out of which narratives are constructed; rather an event is an abstraction from a narrative. An event may take five seconds or five months, but in either case whether it is one event or many depends not on a definition of ‘event’ but on a particular narrative construction which generates the event’s appropriate description.36

Now we can finish the argument. If historical events are abstractions from narratives and therefore it is the narrative that determines their description, then we cannot maintain that ‘the actuality of the past is an untold story.’37 We also cannot say that the same event (or event description) is found either in a chronicle or in a narrative, nor can we maintain that the same event is narrated in different stories. Finally, if historical events are abstractions from narratives, then the definition of composition as the making of a story out of a mere chronicle is misleading, if not simply mistaken. Events in a narrative exemplify the social change that the historical thesis of the narrative is about. We cannot say that the same event exemplifies thesis 1 in narrative 1 and thesis 2 in narrative 2, for what exemplifies cannot be separated from what is exemplified. Today, many theorists and philosophers maintain that the past has no narrative structure so that narrative structures cannot be found in the past. The past is no untold story, as Mink puts it. Instead, so these theorists and philosophers argue, narrative structures are made by historians; they are constructions, or even fictions, the product of their imagination. However, there is a difference between stating that the past has no narrative structure assuming that it is determinate, and stating that the past has no narrative structure assuming that it is indeterminate. The f irst gives rise to the content-form distinction which, I argued, is misleading, if not mistaken. The second does not. If we give up the idea of a determinate past, then we not only give up the idea that it is the purpose of a narrative to represent the past as it was, we also give up the idea that narratives structure or 35 While discussing Mink, Paul Roth too emphasizes the importance of this point. See his ‘How Narratives Explain,’ Social Research, 56(2) (1989), 449-478, at 455. 36 Mink, ‘Narrative Form,’ 201. 37 Ibid., 201.

Narr ative Truth

69

give form to this past, for there is no given content to structure or give form to, to begin with. We now see what is misleading about such phrases as ‘narratives are organizations of events,’ ‘narratives give coherence to the past,’ ‘narrative structures are being imposed or pressed onto the past,’ ‘narratives give unity to a manifold of disparate and heterogeneous events,’ ‘narratives organize the past into temporal wholes,’ ‘narratives make concordant what is discordant’, and ‘narratives are a distortion of or deviation from the past.’ These phrases, all of which can be found in texts on historical narratives, are misleading insofar as they are suggestive of a determinate, amorphous, unstructured past, waiting to be structured or given form by a narrative. The past is no untold story, Mink wrote, and therefore there is no right way to tell it. This claim can be interpreted as a rejection of the empiricist notion of representation.38 After all, the upshot of Mink’s rejection of a determinate historical actuality is that there is neither a given past nor a narrative about this past, organizing or matching it in a certain way. The past is not something given or determined, it is not some uninterpreted reality waiting to be given form by the historian’s narrative. The dualism of scheme and empirical content, ‘of organizing system and something waiting to be organized,’ as Donald Davidson famously argued, ‘cannot be made intelligible and defensible.’39 Although Mink’s rejection of the indeterminateness of the past is in agreement with the criticism of empiricism, it should be 38 While discussing the arguments against the correspondence theory of truth in his Dewey Lectures published as ‘The Structure and Content of Truth,’ Journal of Philosophy, 87(6) (1990), 279-328, at 303-304, Donald Davidson points out that if there are no entities that make our sentences true, then there is nothing for them to represent either. As for the term representation, we have to realize that when philosophers of history like Ankersmit use that term, they have representing-as in mind, not representing-that, and it is only the latter which is criticized by Davidson cum suis. Ankersmit’s philosophy of history, just like Mink’s, is in agreement with the criticism of empiricism, but should also be distinguished from it carefully in order to avoid the obliteration of the distinction between description (representing-that) and narration (representing-as). I will discuss Ankersmit’s views on historical representation in Chapters 4 and 5. The distinction between representing-that and representing-as can also be referred to as a distinction between seeing-that and seeing-as. 39 Davidson, ‘On the Very Idea of a Conceptual Scheme,’ Inquiries into Truth and Interpretation (Oxford: Clarendon Press, 2001 [1984]), 183-198, at 189. For Davidson, content is either ‘reality (the universe, the world, nature), or it is experience (the passing show, surface irritations, sensory promptings, sense–data, the given)’ (192). Conceptual schemes are ‘ways of organizing experience; they are systems of categories that give form to the data of sensation; they are points of view from which individuals, cultures, or periods survey the passing scene’ (183). I discussed Davidson’s criticism of the idea of conceptual schemes extensively in the previous chapter of this book.

70 

The Exemplifying Past

distinguished from it carefully. The reason is that it tends to obliterate the distinction between description and narration. For past actuality is not only no untold story; it is no un-described description either. Mink’s aim is not to reject the empiricist notion of representation, although his narrativist philosophy is in agreement with that criticism. Mink’s argument is that historical events are abstractions from narratives. That is why the past is not to be taken as a determinate actuality, an untold story which historians can seek to represent. The distinction between description and narration is central for Mink. He observes: there is something incompatible about our concept of ‘event’ and our concept of ‘narrative,’ which might be put as follows: the concept of event is primarily linked to the conceptual structure of science (and to that part of common sense that has adopted the language and methods of science); but in that conceptual structure it is purged of all narrative connections, and refers to something that can be identified and described without any necessary reference to its location in some process of development – a process which only narrative form can represent. 40

I would formulate the distinction between description and narration thus: while an event description leads us to reality if it happens to identify the event referred to, an event narration draws us out of it if the event happens to exemplify the historical thesis of that narrative. Narration moves in the opposite direction of description. 41 This distinction between description and narration is central to the exemplification theory of history proposed in this book.

40 Mink, ‘Narrative Form,’ 200-201. 41 It might be objected that this distinction, confusingly, understands description in terms of where it leads and narration in terms of where it comes from, whereas what should be addressed is where description and narration come from and where they lead. We should distinguish, so the objection goes, between the ground and consequent of description and narration. However, the consequences of particular descriptions and narrations are to be mapped on their grounds and vice versa: and it is the pattern that such mapping makes that gives us their meaning. In this sense, the objection is no objection at all. A simple example is this: I intend to order a drink, which is the ground of me ordering a drink and which has, as a consequence, me getting the drink. The meaning of my utterance is precisely this relation between ground and consequence. Similarly, Jacob intended to understand freemasonry historically, therefore she studied it historically and wrote a narrative about it, which had as a consequence that she and we as her readers know what freemasonry is.

Narr ative Truth

71

6 Conclusion Mink stated that the ‘narrative combination of relations is simply not subject to confirmation or disconfirmation, as any one of them taken separately might be,’ even though it is ‘accepted as claiming truth – i.e. as representing a real ensemble of interrelationships in past actuality.’ We can now provide a minimal account that explains why these claims are acceptable and do not contradict each other. This account must acknowledge the distinction between statements on the past (asserting a relation between events or not) and the narrative combination of relations between events, and the semantic and pragmatist vocabularies to address that distinction. We speak about statements semantically using such terms as reference, meaning, propositional content, that which is said, and truth (and the meaning of a statement is given by its truth-conditions). We speak about statements pragmatically, using such terms as identification, use, asserting, and claiming and taking to be true. We speak about narration semantically using the terms exemplification, historical significance, historical thesis, and narrative truth; we do so pragmatically using such terms as exemplifying, signifying, and claiming to be true. In terms of pragmatics, confirming and disconfirming are to be understood as inferential rather than as representational practices. The first to argue this, as far as I know, is Quine, who argued that the verification theory of meaning is unintelligible because we do not compare bits of language with bits of experience or sense datum to see whether they match or not. Not only should we get rid of the dogma that ‘each statement, taken in isolation from its fellows, can admit of confirmation or infirmation [disconfirmation],’ as if the meaning of a particular statement can be determined independently from its (actual or potential) relation to antecedent and subsequent statements. We should also get rid of the dogma that ‘the truth of a statement is somehow analyzable into a linguistic component and a factual component,’42 as if some statements are true by virtue of their meaning alone while other are true by virtue of reality. Thus with the verification theory of meaning, the analytic-synthetic distinction is thrown overboard as well. The crucial insight is that the meaning of utterances – i.e. statements uttered, inscribed, or read – is understood in terms of what informs and follows from them. Taking confirmation and disconfirmation to be inferential practices, then, is to see the confirmation of a statement as having the statement function 42 Willard Van Orman Quine, ‘Two dogmas of empiricism,’ From a Logical Point of View (Cambridge and London: Harvard University Press, 2003 [1953]), 20-46, at 41.

72 

The Exemplifying Past

as premises and conclusions of correct inferences, and the disconfirmation of a statement as not being able to perform such inferences.43 For example, the assertion that American freemasonry descended from the medieval Knights Templar is disconfirmed if Benjamin Franklin or the British lodges from which he imported freemasonry had no ties with the Knights Templar whatsoever. 44 In terms of semantics, confirmation and disconfirmation comes down to the evaluation of the truth-conditions, hence the meaning of the statements under consideration. We determine the truth value of statements by evaluating their truth-conditions, confirming or disconfirming what it is that the statement says, although most of the time we feel no need to establish whether what s says is true or not, for knowing that the elements are there to determine whether s is true or not usually suffices for the purpose at hand – and in the case of fiction, we not only feel no such need, it is the purpose of fiction that no such need is felt. Confirming or disconfirming what it is that the statement says is in terms of what we do an inferential practice. If a statement functions as a premise or conclusion in correct inferences, then it is confirmed. If it does not lead to correct inferences, then the statement is disconfirmed. In terms of pragmatics, taking a sentence to be true is committing oneself to that particular use of words on that particular occasion of its utterance and the inferential relations entailed by it. Here semantics answers to pragmatics, for we understand the content of our words in terms of what we do with them and use them for. Pragmatics (the theory of use) and semantics (the theory of meaning) are complementary, though it is semantics that answers to pragmatics in the pragmatists’ philosophy of language. 45 In terms of semantics, the past as referred to in a historical narrative derives its meaning from the historical thesis it exemplifies. Historical theses are not true or false: they are exemplified or not (if we were to say that historical theses are true when they are exemplified and false when they are not exemplified, then truth would be a spell we are under rather 43 On inferentialism, see Robert Brandom’s important and elaborate account in Making It Explicit. Reasoning, Representing, and Discursive Commitment (Cambridge MA and London: Harvard University Press, 1998 [1994]). 44 This example is taken from Jacob, Freemasonry, 1-4. 45 Pragmatism comes in different flavours. In its broad sense, pragmatism aims to explain what we do with words and what we use them for. On a narrow and broad sense of pragmatism and its relation to semantics, see Brandom, Perspectives on Pragmatism. Classical, Recent, & Contemporary (Cambridge MA and London: Harvard University Press, 2011), especially Chapter 2. Semantics answering to pragmatics is also one of the themes of Brandom’s Making It Explicit.

Narr ative Truth

73

than fulfilling its semantic role). This is another way of saying that historical theses cannot be confirmed or disconfirmed. What would we do if someone were to ask us to confirm or disconfirm (the theses of) a particular narrative? We would either start retelling it or start telling a different story, as Mink observes:46 but this would obviously not be considered to be confirmation or disconfirmation in the proper sense of the terms. In terms of pragmatics, to claim a historical thesis to be true is to claim that the narrative is the best guide present at hand as long as it is not displaced by another narrative. Here semantics again answers to pragmatics, for as long as the past is understood in terms of the narrative, i.e. as long as its actions, events, and objects exemplify the historical thesis, then the historical thesis is rightfully claimed to be true. So even though historical theses are not true or false semantically (in terms of semantics they are exemplified or not), they are claimed to be true or false. A historical thesis does not claim that the past is thus-and-so: it claims (if we can properly call it that) that the past is thus-and-so in light of and inseparable from the thesis itself. 47 Only after reading Jacob’s book do we understand freemasonry historically, i.e. as exemplifying the eighteenthcentury dream of a society based on merit instead of birth and wealth. Narratives do not assert; they show. As a whole, narratives cannot be confirmed or disconfirmed, but have to be taken at face value. Only by virtue of the narrative do events exemplify the historical thesis of that narrative. What exemplifies cannot be separated from what is exemplified. Mink wrote that the narrative combination of relations, what we have called the historical thesis, represents ‘a real ensemble of interrelationships in past actuality.’ This is indeed so. Not because narratives represent a given combination of interrelations between past events, but because with the narrative at hand, the interrelationships are there, and the past exemplifies the historical thesis. Narratives do not reflect past reality: past reality reflects the narrative.

46 Mink, ‘Narrative Form,’ 198. 47 This is, I think, also the claim argued by Ankersmit in his Meaning, Truth, and Reference in Historical Representation (Ithaca and New York: Cornell University Press, 2012), Chapters 4 and 6.

4

Resemblance, Substitution, Expression The Renaissance stands by itself like a painting on an easel.1 − Felix Gilbert

1 Introduction At the Piazza del Campidoglio in Rome stands the equestrian statue of the Roman emperor Marcus Aurelius (121-180). It is a weatherproof replica of the original, which is on display at the Palazzo dei Conservatori. It survived because the medieval Romans believed it to represent the first Christian emperor Constantine (272-337). The conquered barbarian king who according to medieval testimonies kneeled underneath the horse’s front right leg, begging for mercy, is missing. The statue represents Aurelius as he had just returned from one of his successful conquests. With his gesture he may be greeting the Roman people or show clemency for the conquered king. Striking is his somewhat distant but patient facial expression. I take it that the statue expresses his grandeur and dignity as a stoic harbinger of peace. This is what the statue, under this interpretation, means. We may wonder whether the statue looks like the emperor. It probably does. Of course, never having met him in person, I do not know what Aurelius in real life looked like, but I do know that this particular statue resembles other portraits of him. I also have no reason to doubt the sculptor’s skill to imitate Aurelius’ physique, although he perhaps represented Aurelius better looking than he in fact was – or, what is more likely, rather than imitating Aurelius’ physique, the sculptor had the skill to represent Aurelius in such a way that we believe that he skilfully imitated Aurelius’ physique. Clearly, the sculptured horse resembles a real horse. We may also observe that the statue was meant to take the place of the real Aurelius and function as a substitute for him in the sense that the attitude towards the real Aurelius should be the same as the attitude towards his statue: an attitude of awe and submissiveness. The feeling or thought expressed by a representation need not be aroused in the viewer. A statue expressing grandeur and dignity may arouse a sense of awe and submissiveness in the viewer. The latter is, I think, what the statue was meant for. 1 Felix Gilbert, History: Politics or Culture? Reflections on Ranke and Burckhardt (Princeton and Oxford: Princeton University Press, 1990), 57.

76 

The Exemplifying Past

I have casually introduced three terms that are commonly used to elucidate the relation between a representation and what it represents: resemblance, substitution, and expression. These terms are not always helpful. Many representations do not resemble what they represent, and only some representations function as a substitute for or express something about what they represent. None of these terms is however completely helpless when it comes to elucidating the equestrian statue of Aurelius. Resemblance, we may argue, helps us determine what the statue represents: a victorious Roman on a horse, and if we would know what Aurelius looked like, we might even have recognized him. Here resemblance supports the interpretation of the representation. The substitution theory helps us explain the raison d’être of the statue: because the statue functions as a substitute for the real Aurelius, it has the effect desired by the Roman authorities, who no doubt had particular motives to commission and exhibit statues such as these. Perhaps the expression theory is the most helpful theory in this case, for if we would not know what the statue expresses, we would not know why it functions as a substitute for the real Aurelius. But then if we would not know that the statue portrays Aurelius, and would know nothing about this emperor, we would be in the dark as to what the statue expresses. Or we would interpret the statue differently, and think that it expresses something else. Or we would think that the statue does not express anything at all, and think that it is a mere statue of a man on a horse, raising his hand for no apparent reason at all. The resemblance theory has been much criticized, and to my knowledge, there are no supporters of this theory. Its main criticism is that anything is like anything else in at least some respect. Therefore, the resemblance found between a representation and what it represents is hardly capable of elucidating that relation. Nevertheless, we often say that a representation resembles what it represents: a good portrait is to resemble the person it is a portrait of. The substitution theory of representation was first formulated in relation to two different contexts. One was that of the effigy that took the place of the dead king on his funeral bier. The other that of the person who in a public or legal function represents another person who is absent himself.2 The expression theory of representation has its home in the philosophy of art in its broad sense as the philosophy of symbols,3 and emphasizes that apart 2 Roger Chartier, On the Edge of the Cliff. History, Language, and Practices, trans. Lydia G. Cochrane (Baltimore and London: The Johns Hopkins University Press, 1997), 92. 3 A highly influential and still seminal work in this regard is Nelson Goodman’s, Languages of Art. An Approach to a Theory of Symbols (Hackett Publishing: Indianapolis and Cambridge, 1976). Important in this context is also Arthur Danto’s, The Transfiguration of the Commonplace. A Philosophy of Art (Cambridge: Harvard University Press, 1981).

Resembl ance, Substitution, Expression

77

from representing something, artistic representations express something about what they represent: they have a semantic content. The concept of representation also helps us to better understand the nature of historical narratives. This is what Frank Ankersmit proposes. 4 Like such things as paintings and statues, narratives are representations. His views on historical representations will be discussed closely in this chapter. It argues that: (1) although all representations stand for what they represent, some do more than that; (2) some representations take the place of what they represent and function as their substitute; (3) some representations express something about what they represent, and some do both, as for instance the equestrian statue of Aurelius does; (4) historical representations too express something about what they represent; and (5) representations are instruments of rhetoric inasmuch as they want their interpreters to see what is represented in a certain light and take a certain attitude towards it.

2

Ankersmit on Historical Representation

According to Ankersmit, a historical representation takes the place of the absent past by functioning as a substitute for it. His notion of historical representation has been scrutinized,5 but his substitution view of representation has not. This is surprising since the substitution view goes to the heart of his view on historical representation. In ‘Danto on Representation, Identity, and Indiscernibles,’ Ankersmit supports his claim by comparing the substitution theory of representation to its rival, the resemblance theory of representation.6 Seven arguments can 4 Ankersmit, Historical Representation (Stanford, CA: Stanford University Press, 2001). Eugen Zeleňák argues that the plausibility of the proposal is strengthened when it is related to the work of Goodman in his ‘Using Goodman to Explore Historical Representation,’ Journal of the Philosophy of History, 7(3) (2013), 371-395. 5 For example by John Zammito, ‘Ankersmit and Historical Representation,’ History and Theory 44 (2005), 155-181, and Eugen Zeleňák, ‘Exploring holism in Frank Ankersmit’s historical representation,’ Rethinking History (2009), 13(3), 357-369. 6 Ankersmit, ‘Danto on Representation, Identity, and Indiscernibles,’ History and Theory (1998), 37(1), 44-70, reprinted in his Historical Representation, 218-248. I will refer to this reprint. Ankersmit first proposed to use the term historical representation for historical narratives in his De navel van de geschiedenis. Over interpretatie, representatie en historische realiteit (Historische Uitgeverij Groningen, 1990), 150-181. There he already pleaded for the substitution view on representation. In his latest book, Meaning, Truth, and Reference in Historical Representation (Cornell University Press, 2012), 56-57, 161, he again states that representations should function as a substitute for what they represent and take their place.

78 

The Exemplifying Past

be discerned that purport to substantiate the claim. Several of them can be found elsewhere in his work. Each argument will turn out to be unconvincing. They do, however, help us grasp what is at stake when something is taken to be a representation of something else. Closely discussing Ankersmit’s arguments is therefore far from a scholastic exercise: there is much to be learned by carefully considering them. There is something odd about my extensive and critical discussion of his arguments. This is so because the conclusion following from that discussion endorses rather than opposes what Ankersmit is after when he introduces the concept of representation in his philosophy of history: a representation, he claims, is a proposal as to how a certain part of the past is best viewed. Since I accept this important suggestion in this chapter and the next, the discussion that follows is not to be read as an indictment of what Ankersmit has to say about historical representations. It aims to shows how the terms ‘resemblance’ and ‘substitution’ ought to be used in relation to the concept of representation and why ‘expression’ is the most helpful term in the context of historical representation. Rather than resembling the past or functioning as a substitute for it, historical representations express something about the past in order to make us see the past in a certain light. In virtue of that they are instruments of rhetoric. 2.1

Resemblance and Substitution

Ankersmit’s first argument to prefer the substitution theory to the resemblance theory is that words do not resemble what they represent.7 This is, of course, true, but a moment’s thought makes us realize that this argument cannot be used to favour the substitution view, for words are no substitutes for what they represent either. Not even when the word is a name. My name may stand for me, in the case of my signature on a legal document, or when I was called in for supper when I was a child, but it does not take my place in any relevant sense of the term – except, perhaps, in such extraordinary cases as a memorial service when a name can be said to take the place of its absent bearer. I return 7 Ankersmit, Historical Representation, 222. Perhaps someone with a preference for early modern philosophy would suggest that words do resemble what they represent and argue that the image or idea that a word invokes resembles what that word represents. The word ‘apple,’ for example, invokes the idea ‘apple’ in my mind and that idea resembles real apples. Similarly, the image that strokes of paint invoke resembles what those strokes of paint represent, which explains why a painting of an apple resembles real apples. This way of putting things is, however, misleading. It is grounded in some variant of empiricism and the discredited view that it is the task of language to represent reality as it is.

Resembl ance, Substitution, Expression

79

to this and similar cases below. For now I observe that this first argument gives us no reason to choose the substitution theory over the resemblance theory. Words neither resemble nor are a substitute for what they represent. The argument is flawed for another reason. Like words, strokes of paint do not, as paint, resemble what is being represented with it.8 Neither are strokes of paint substitutes for what is being represented with them. We are clearly on the wrong track here, and using the notions resemblance and substitution in a way that obscures rather than clarifies the concept of representation. To avoid any further confusion, we should start by distinguishing representations from mere ordinary things, and with that, between using the notions of resemblance and substitution in relation to the concept of representation and using the notions of resemblance and substitution apart from the concept of representation. Obviously, the resemblance and substitution theories of representation are concerned with the former, not with the latter. The question whether words, dots of paint, carves in wood, pixels on a screen, celluloid, bodily movements, and pieces of marble resemble or are a substitute for what is being represented with them, is asking the wrong question. According to Ankersmit’s second argument in favour of the substitution theory of representation, the substitution theory is compatible with the resemblance theory while the reverse is not. He reasons thus: there is no inconsistency in saying (1) that all representation is substitution and (2) that in pictorial representation resemblance is sometimes (or perhaps even customarily) relied upon in order to achieve a believable substitute for the represented. On the other hand, one cannot do the opposite and assert (1) that the resemblance theory is correct and (2) that texts, nonfigurational art, or parliaments ‘represent’ a represented.9

Indeed, parliaments do not resemble the people they represent, and members of parliament stand for the people they represent and take their place. But is this the resemblance that is at stake here? Are not the laws of parliaments in representative democracies supposed to resemble the laws that the people would make? If so, the resemblance and substitution theories are compatible in the case of parliaments, contrary to what Ankersmit thinks.10 8 This Ankersmit realizes himself, without noting the contradiction with his first argument. See Ankersmit, Historical Representation, 310 n.8. 9 Ibid., 223. 10 Elsewhere, in the context of political representation, Ankersmit agrees with Jean-Jacques Rousseau’s criticism of the resemblance theory. Rousseau criticized the resemblance theory for making representation redundant: if it is known what the opinion of the people is, then there is

80 

The Exemplifying Past

Ankersmit not only points at parliaments, but also at non-figurative art and texts. If one agrees that non-figurative art does not represent anything, and therefore, there is nothing for it to resemble, then there is nothing for it to be a substitute for either. Thus, in this case the question of the compatibility of the resemblance and substitution theories does not even arise. This leaves us with the question of whether resemblance and substitution are compatible in the case of texts. This is the first argument again. Agreed, words do not resemble what they represent, but they are no substitutes for what they represent either. This critique of Ankersmit’s first two arguments only purports to show that those arguments cannot be used to show that the substitution theory is to be preferred to the resemblance theory of representation. I do not favour either of these theories in the context of historical representation. This I will make clear, but first we turn to Ankersmit’s third argument. According to the third argument, the disadvantage of the resemblance theory is its tendency ‘to reduce problems of representation to those traditionally investigated by epistemology.’ If resemblance is to account for the relation between the representation and the represented, then, according to Ankersmit, resemblance will be ‘the representational analogue of the criteria of truth that epistemologists always discuss.’11 He argues thus: The equestrian statue of Aurelius at the Piazza del Campidoglio and the person of flesh and blood named Aurelius both exist (or once existed in the case of the latter), and this implies, according to Ankersmit, that they have the same ontological status.12 It follows that: the relationship between the represented and its representation – a world-to-world relationship – could never be modelled on the relationship between world and language (which is the exclusive domain of interest for the epistemologist).13 no need for its representation. See Ankersmit, ‘Political and Historical Representation,’ in Ananta Charana Sukla ed., Art and Representation: Contributions to Contemporary Aesthetics (Westport and London: Praeger, 2011), 69-89, at 70-72. Rousseau assumes this resemblance to be an ‘exact resemblance’, and indeed, his argument holds if this is assumed. However, resemblance always presupposes non-identity and thus excludes ‘exact resemblance’: only two different things can be said to resemble one another. Therefore, Rousseau’s argument does not hold. Elsewhere, as we will see, Ankersmit too emphasizes that resemblance presupposes non-identity between representation and the represented. 11 Ankersmit, Historical Representation, 223. 12 Ibid., 224. He emphasizes this several times. See e.g. Historical Representation, 11 and 81, and Meaning, Truth, and Reference, 161. Here we disregard the fact that the statue still exists whereas Aurelius does not. 13 Ankersmit, Historical Representation, 225.

Resembl ance, Substitution, Expression

81

Why this is so is not immediately clear. Obviously, Aurelius and his statue both exist and we can identify and describe them. It is also clear that the question how language (description) relates to the world (Aurelius and his statue respectively) is of interest to the epistemologist. Thus far, we need not disagree, but we should not forget the difference between Aurelius and his statue: a statue is a representation whereas the person Aurelius is not. Calling the relationship between Aurelius and his statue a world-to-world relationship, as Ankersmit does, then, is misleading, for this world-toworld relationship is asymmetric: Aurelius is not related to his statue as his statue is related to Aurelius. The human being with the name Aurelius who lived from 121 to 180 AD is a necessary condition of his statue and that is obviously not the case the other way around. Aurelius and his statue, therefore, do not have the same ontological status. To be sure, the statue is an object and part of our world, but as a representation, the statue has to function as such, and this functioning is an event rather than an object. Rather than a world-to-world-relationship, the relation between Aurelius and his statue is a world-to-representation-in-the-world relationship, and this is the relationship that epistemologists worry about. The relationship between the represented and its representation thus can be modelled on the relationship between world and language, contrary to what Ankersmit thinks. This refutes his third argument to disqualify the resemblance theory in support of the substitution theory. Although I agree with Ankersmit that the resemblance theory tends ‘to reduce problems of representation to those traditionally investigated by epistemology,’ the point is, I think, not to give in to this tendency. Resemblance gives rise to epistemological worries as long as it is understood as a measure of realism, but if we realize that realism of representation is a matter of habit rather than of fit, a matter of standards of representation of some culture at a given time, these worries do not arise, for then we are no longer bothered by the realist assumption that there is a ready-made or antecedent world for our representations to fit or resemble.14 2.2

Representation and Identity

In his fourth argument, Ankersmit reminds us of Arthur Danto’s distinction between representation in the sense of being present again, as for example Dionysus was believed to be present during certain rituals, and representation in the sense of standing in the place of something else, as for example 14 This is a central argument of Goodman in his Languages of Art, 37-39.

82 

The Exemplifying Past

representatives of parliament stand for us. In the first sense, the thing or person presents itself again. In the second sense, representation is the making present of something that is absent, like members of parliaments make the people they represent present.15 In the first sense, the representation and the represented are one and the same thing, person, or divinity, whereas in the second sense, the representation and the represented are two different things or persons. According to Ankersmit, the resemblance theory cannot account for representation in the sense of being present again, for in this sense of representation, the representation and the represented are identical while resemblance presupposes non-identity: only two different things can be said to resemble one another.16 Some thing or person does not resemble itself and of two identical things we do not say that they resemble one another but we say that they are identical to one another. The argument that resemblance presupposes non-identity is faultless.17 However, it holds for the substitution theory as well. Something can only function as a substitute for something else. Substituting one thing with itself is no substitution. It makes no sense to claim that Dionysus functions as a substitute for himself when he presents himself in art or cult.18 Both the resemblance and the substitution theory cannot account for representation in the first sense of being present again. So we still have no reason to prefer the substitution 15 Danto, The Transfiguration of the Commonplace, 19. The distinction between representation as ‘being present again’ and representation as ‘standing in the place of something else’ has been made before. The first formulation of it that I know of is in Antoine Furetière’s late-seventeenthcentury’s Dictionnaire universel. I came across this reference to Furetière in Chartier, On the Edge of the Cliff, 92. 16 Ankersmit, Historical Representation, 226. 17 The argument is far from self-evident. Goodman holds – and many have followed him – that ‘An object [A] resembles itself to the maximum degree but rarely represents itself: resemblance, unlike representation, is reflexive. Again, unlike representation, resemblance is symmetric: B is as much like A as A is like B, but while a painting may represent the Duke of Wellington, the Duke doesn’t represent the painting.’ Goodman, Languages of Art, 4. I disagree. First, A does not resemble A but is identical with A. Only two different things may resemble one another. Resemblance is therefore not reflexive. Second, we never say that a person resembles his or her portrait. We say that a portrait resembles the person it depicts. Therefore, resemblance is not symmetric. Third, only when the representation is there, A resembles B according to the resemblance theory of representation. We should distinguish between the concept of resemblance apart from the concept of representation and the concept of resemblance in relation to the concept of representation. The resemblance theory is concerned with the latter. It is the function to represent that determines the object’s meaning. 18 Friedrich Nietzsche related the Dionysian impulse in art to ‘being present again’ and the Apollonian impulse to ‘standing in the place of something else.’ See his Die Geburt der Tragödie (Könemann Verlagsgesellschaft: Cologne, 1994), § 10.

Resembl ance, Substitution, Expression

83

view of representation to the resemblance view. I might add that in all other arguments, representation is used in the second sense of standing for something else, i.e. as symbolizing. Ankersmit’s fifth argument against the resemblance theory runs thus: For think of Gombrich’s renowned hobbyhorse. The average table will more closely ‘resemble’ a real horse than a hobbyhorse, and yet, as Gombrich so rightly insisted, the hobbyhorse more successfully represents a real horse than a table (for the simple reason that the hobbyhorse may function in the eyes of the child as a real horse, whereas the table will not).19

I do not see why a table more closely resembles a real horse than a hobby­ horse. Arguably, a table is less successful in representing a horse because galloping with a table is more diff icult than galloping with a stick (or hobbyhorse), but this cannot be decisive, since there is no reason to assume that a table cannot function as a horse and is less successful therein than a stick (or hobbyhorse).20 The art historian Ernst Gombrich did not insist otherwise, as Ankersmit erroneously suggests, for Gombrich points out that any ridable object would do.21 This is what not only Ankersmit himself, but also Danto and Goodman emphasize: anything can be a representation of anything else.22 Whether a table more closely resembles a real horse than a hobbyhorse is of little interest. Resemblance is not a necessary condition of representation. If, however, there is a representation, we may detect the resemblance between the representation and the represented, but this may well be because anything is like anything else in at least some respect. 19 Ankersmit, Historical Representation, 226. It is misleading to compare a table with a hobby­ horse since a hobbyhorse is a representation and a table is not: it would have been better to compare a stick with the table. 20 The reason why a stick functions as a horse whereas a table does not is a matter of habit. One can imagine a society in which its members do not eat at tables nor sit down at tables to converse with one another. They only converse while standing, holding a stick in their hand to signal the fact that they converse and are not to be disturbed by their children. Tables are just there to put the dirty dishes on. In this imaginary society, parents would give their children the tables to play with, not the important sticks. 21 Gombrich writes: ‘It needed two conditions, then, to turn a stick into our hobby horse: first, that its form made it just possible to ride on it; secondly – and perhaps decisively – that riding mattered.’ Gombrich, Meditations on a Hobby Horse and Other Essays on the Theory of Art (London and New York: Phaidon Press, 1978), 7. 22 Goodman, Languages of Art, 5; Danto, The Transfiguration of the Commonplace, 72-73; and Ankersmit, Historical Representation, 227.

84 

The Exemplifying Past

We may conclude that, if a represents b, then (i) a stands for b, (ii) a is not b, and (iii) a sometimes resembles b, a sometimes takes the place of b, and a sometimes expresses c about b. Ankersmit need not disagree with this conclusion, although he cannot deny that ‘standing for’ is not to be identified with ‘taking the place of’, i.e. with ‘substitution’. All representations stand for what they represent, but only some representations are a substitute for what they represent by taking their place. Which sort of representations do so will be discussed shortly. First there are two more arguments to consider. Ankersmit’s sixth argument to favour the substitution theory of representation to the resemblance theory runs thus: What we should note is that the substitution theory differs from the resemblance theory by avoiding the suggestion of a conceptual hierarchy between the notions of representation and of identity. The resemblance theory inadvertently creates such a hierarchy by presenting resemblance as the link between identity and representation: if A sufficiently resembles B, what A represents is (or is identical with) B. Put differently, when A represents B it can be tailored by the criteria of resemblance in such a way that the result of all the tailoring will be (identical with) the represented B all over again. And this is different with the substitution theory: this theory does not posit such a conceptual hierarchy between identity and representation.23

Two flaws can be detected. One is that the argument suggests that resemblance is a condition of representation (‘if A sufficiently resembles B, what A represents is (or is identical with) B’), which it is not. Resemblance is neither a sufficient nor a necessary condition of representation, as Ankersmit too notes elsewhere.24 Resemblance does not establish the relation between the representation and the represented: a may sufficiently resemble b without representing b. Therefore it is not a sufficient condition. It also is not a necessary condition, as we already said. If a represents b, it does not follow that a resembles b. My signature represents me, but it does not look like me. In many cases we will be able to find some resemblance between the representation and the represented, but this is simply so because anything is like anything else in at least some respect. The other flaw in the argument is that it suggests that resemblance creates the identity between representation and represented: if a represents b, then 23 Ibid., 227. 24 Ibid., 226. This was first argued by Goodman in his Languages of Art, 5.

Resembl ance, Substitution, Expression

85

a resembles b to such a degree that a is identical to b. Therefore, so the argument goes, resemblance is the ‘link between identity and representation’. However, only two different things can be said to resemble one another. Since resemblance precludes identity (in the sense of being identical), the resemblance theory does not create a link between the notions of identity and representation. Therefore it cannot suggest a conceptual hierarchy between these notions by making the notion of identity secondary to that of representation. Hence, we still have no reason to prefer the substitution theory to the resemblance theory. The passage following the one just quoted is relevant because it explains why the substitution theory does not suggest a conceptual hierarchy between representation and identity (a hierarchy which, as I have explained, is absent in the resemblance theory as well). Representation is the representation of identity because identity only comes into being by representation, and vice versa; there is not, first an identity, which is or could, next, be represented – whether in agreement with certain criteria of resemblance or not – no, representation and identity both come into being at one and the same time.25

Since identity and representation both come into being at one and the same time, there cannot be a conceptual hierarchy between these two terms. It should be noted that in this passage the notion of identity is used in the sense of ‘knowing what or who something of someone is’ and not in the sense of ‘one thing being identical to another,’ as in the previous passage quoted.26 Obviously, the former sense of identity is more interesting in the context of representation. It allows us to say such sensible things as: a statue of a person should, like portrait painting, represent the identity of the person who is represented.27 The question is: do representation and the identity of a person or thing come into being at one and the same time, as Ankersmit suggests? Obviously, the person named Aurelius is a necessary condition for his statue, but that is not the issue here. Ankersmit reminds us of the fact that if we want to know the identity of Aurelius, we have no other choice but to represent him. This is an important insight into the nature of representation. To be sure, the identity of Aurelius is fabricated, that is to say, the equestrian statue has a particular purpose: it wants us to adopt a certain attitude 25 Ankersmit, Historical Representation, 227. 26 Ankersmit makes this distinction in ibid., 232. 27 Ibid., 232.

86 

The Exemplifying Past

towards Aurelius as an emperor. It tells us viewers what we should take his imperial identity to be. The Roman authorities no doubt had particular motives to present Aurelius in a certain way and fabricate his image. This leaves untouched the important insight won: identity and representation come into being at one and the same time.

3

La Condition Humaine

Six arguments have been discussed, and even though we gained several important insights into the concept of representation, we have not found a convincing reason to favour the substitution view to the resemblance view of representation. In his seventh and last argument, Ankersmit submits the resemblance and substitution theories to a test. Perhaps this test will settle the issue at last. The test is to explain René Magritte’s well-known 1933 painting La condition humaine. Ankersmit asks himself which of the two theories is able to explain the status of this painting as a representation. Not surprisingly, his conclusion is that the substitution theory provides such explanation, whereas the resemblance theory does not. Magritte’s painting pictures a room with a view. In front of the window in the painting stands another painting on an easel. This painting in the painting represents a part of a landscape that would be visible, were it not that it hides precisely that part of the landscape. The landscape represented by the painting in the painting is identical to the landscape it hides from view. So here we have a representation in a representation. A painting in a painting representing a landscape is part of a painting of a room with a view that is largely hidden by that painting.28 The painting in the painting representing the landscape can according to Ankersmit be dealt with from the point of view of the resemblance theory as well as that of the substitution theory. It is both a substitute for the landscape it represents and it resembles the landscape (this is problematic in that here the representation (the painting in the painting) and the represented (the landscape) are identical, while both resemblance and substitution presupposes non-identity: only two 28 As far as I know, at least four paintings of Magritte bear the title La condition humaine. They were painted in 1933, 1935, 1944 or 1945, and 1948 respectively. Also L’appel des cimes (1933), La Belle Captive (the title of two paintings, one from 1950 and one from 1967), Les Promenades d’Euclide (1955), and Le Pays des Miracles (1960) depict a painting in a painting representing a landscape.

Resembl ance, Substitution, Expression

87

different things may resemble one another, and only of two different things one can function as a substitute for the other. I will come back to this). Ankersmit writes: It is true that the relevant part of Magritte’s painting is identical with itself (as part of the painting), but this part of the painting has two functions to fulf ill: (1) to function as the landscape itself (in the way that paintings, according to the substitution theory are required to function as what is represented by them), and (2) to function as the painting of the landscape (i.e., as a representation of the landscape). […] Now, needless to say, the difference between these two functions is exactly analogous to that between (represented) reality and its representation. But the difference of functions disappears if we see the painting with the eyes of the adherent of the resemblance theory: for there is no difference in the resemblance relationship between (1) the landscape itself and that part of the painting for which the painting is the substitute and (2) the landscape itself and its representation as presented by the painting in the room.29

Allow me to repeat the point made here. The painting in the painting fulfils two functions: (1) it functions as the landscape itself and is as such a substitute for the landscape; and (2) it functions as a representation of the landscape. The distinction between these two functions is absent from the point of view of the resemblance theory, so the resemblance theory fails the test and the substitution theory does not. This argument does not, however, as I will show, convince and make us disqualify the resemblance theory of representation. Recall the emphasis we have placed on the function to represent throughout this chapter. If a stick is to represent a horse, it has to function as one, and this means that riding with it is possible and that riding matters. The stick functions as a horse just as the painting in the painting functions as a landscape and in both these cases representation and represented are distinct entities: the stick is no horse, just as the painting in the painting is no landscape. The stick represents the horse and is therefore a substitute for the horse. This is, if we would follow Ankersmit, also true of the painting in the painting. Because the painting in the painting represents the landscape, it is a substitute for the landscape. Now, saying that the painting in the painting functions as the landscape itself – the 29 Ankersmit, Historical Representation, 229.

88 

The Exemplifying Past

first function (1) – is identical to saying that the painting in the painting represents the landscape. But is this not what the second function (2) says? After all, according to the second function (2), the painting in the painting functions as a representation of the landscape. The point is that Ankersmit’s distinction between the two functions is highly questionable. The painting in the painting is a representation of the landscape and functions as a substitute for the landscape, but these are not two different functions of the painting in the painting, but one and the same function. The fact that the painting in the painting hides the landscape from view is unimportant. The painting need not have been situated in front of the window, but, for instance, underneath it. If that were the case, the painting would still represent the landscape and function as a substitute for it, even though we could see the landscape through the window. The fact that the painting in the painting hides the landscape from view does not imply that it is a substitute for the landscape in the sense of the substitution theory: replacing one thing for another is substitution, but that does not give us representation. Substitution, like resemblance, is neither a sufficient nor a necessary condition of representation. Maybe we have missed the point. The landscape as painted by the painting in the painting is identical with the landscape that the painting in the painting hides from view. We are dealing with an extraordinary representation here. We never say that a picture is identical with what it pictures: we would be confusing the one for the other and that is not what we normally do. We do not, for example, pick an apple from a painted bowl of fruit. The fact that the landscape hidden from view is identical with the painting of the landscape is only possible in special cases like Magritte’s painting, but then the issue is not what theory of representation is the most plausible one; then the issue is whether or how a representation can hide our view. We may even question whether the painting in the painting represents the landscape. Since the landscape hidden from view is identical with the landscape as represented by the painting in the painting, there is no way we can distinguish between the representation and the represented. This is what we always should be able to do, according to both the substitution and the resemblance theory of representation. To properly understand Magritte’s painting, we should take the painting as a whole into account. The picture consists of a room with a view and that view is partly obstructed by the painting in the painting representing a landscape that is identical with the landscape it hides from view. We can identify what it is that we see when we look at a painting, just as we can identify what it is that we see when we look through a window. Is Magritte

Resembl ance, Substitution, Expression

89

not making the point that looking at a painting is identical to looking through a window? Magritte himself tells us the following: the tree represented in the painting hid from view the tree situated behind it, outside the room. It existed for the spectator, as it were, simultaneously in his mind, as both inside the room in the painting, and outside in the real landscape. Which is how we see the world: we see it as being outside ourselves even though it only is a mental representation of it that we experience in ourselves.30

I do not think that we see the world ‘as being outside ourselves,’ we simply see the objects and events the world consists of (a landscape, an easel, a tree, a window, a painting). Perceiving a painting is no different from perceiving the world surrounding it (i.e. the waves of light hitting the retina are no different; there might be a difference of processes in our brain caused by the three-dimensional interpretation of a two-dimensional picture, but that is not a philosophical issue, but an issue to be dealt with by the neurosciences).31 The more interesting question is what the painting expresses. On the one hand, art is one of the things in the world; we see a painting on an easel in front of a window. On the other hand, art is expressive and therefore not merely one of the world’s objects: it expresses something about the world. In this sense, art is both within and without the world. Not all representations express something about the world, but all representations that aim to have an audience take an attitude towards what is represented do. It is clear to me that Magritte wants to say something about figurative painting as such. Is the painting of a room with a painting in front of a window not a metaphor for perception? Is it not true that the painting expresses that seeing a figurative painting is like seeing through a window, as if reality is a mere appearance? Is not that part of the human condition? At least, that is how Magritte sees the world and why he explores this facet of human perception in so many of his paintings. Under this interpretation, the painting can be said to embody its meaning.32 30 Quoted in ibid., 228. 31 The point is to distinguish between the retinal processing of stimuli and what is called ‘attentive looking’ in the philosophy of art and aesthetics. The latter is simply not one of the subjects of this book. It is the subject of Bence Nanay, Aesthetics as Philosophy of Perception (Oxford: Oxford University Press, 2016). Nanay discusses the distinction on pp. 139-143. There is a history in the way art is perceived, but only in the sense of perceptibility as a kind of attentiveness, as the romantic poet Novalis would have it. 32 This last point, which is Hegel’s and Danto’s, is discussed in Chapter 6.

90 

The Exemplifying Past

For a proper understanding of this interpretation we should realize that the way Magritte sees the world has nothing to do with perception. To be sure, Magritte’s painting expresses something about perception, but that is how he conceives of perception. This we may agree or disagree with. Not that his painting is true or false. It is not that if we disagree with Magritte on how to conceive of perception that we take his painting to be false. The painting expresses that the world is a mere appearance to our perception, at least it does so under the interpretation given, and with this thought we can agree or disagree, not with the fact that this is what Magritte’s painting expresses. But what if the painting were to express something differently under a different interpretation? Again, we could agree or disagree with the thought that the painting expresses, not with the fact that, under this alternative interpretation, it is this particular thought that the painting expresses. The new question here is how we are to theorize the correctness of interpretations of representations. The answer was already suggested. If c is a correct interpretation of a, then (i) a expresses c, and (ii) a embodies c. A correct interpretation has the representation embody its meaning.33 Correctness of interpretation depends on what a representation is meant to symbolize. Correctness of representation depends on conventions of representation.34 A representation that does not express something about what it represents too is not true or false itself. Obviously, what we say about a representation is either true or false. A label on a tube of paint represents the paint in the tube, and if a yellow label is by mistake printed on tubes of blue paint, we do not say that the label is false (or that the paint in the tube is false): we say that the label misrepresents the paint in the tube.

4

Substitution Once More

We have found no reason to favour the resemblance or the substitution theory of representation over the other. We also found no reason to rehabilitate the resemblance theory. Perhaps the best argument against the resemblance view is Gombrich’s argument that function precedes 33 See Chapter 6. 34 The distinction tells us something about aesthetic evaluation. Too often aesthetic judgment is limited by conventions of correct representation. Educators, artists, art historians, and philosophers would, I presume, rather want correctness of interpretation to play a decisive role in aesthetic evaluation, and have students and audiences see how the artwork embodies its meaning, and evaluate the artwork’s appropriateness as to what it expresses.

Resembl ance, Substitution, Expression

91

resemblance.35 Something needs to function as a representation before the question of resemblance arises. This argument shows that resemblance is no sufficient condition of representation. It also underlines the point that representations stand for what they represent in that they symbolize what they represent: if a stands for b, then a functions as a representation of b and vice versa. To function as a representation, acquaintance with the symbolic system used is a prerequisite, and this acquaintance we acquire through learning.36 Standing for is, however, not taking the place of. This is, then, I think Ankersmit’s mistake. He inferred from the fact that all representations stand for what they represent that representations are substitutes for what they represent, and therefore, that the substitution theory is the plausible theory of representation. The substitution theory is, however, only plausible in some cases. It strikes me that the substitution theory only makes sense if the representation represents an absent body, typically in the case of an absent sovereign37 or divinity and in the case of public and political representation, although also a stick, as in Gombrich’s famous essay, may take the place of an absent horse in the child’s play.38 Gombrich formulates his view on substitution thus: The clay horse or servant, buried in the tomb of the mighty, takes the place of the living. The idol takes the place of the god. The question whether it represents the ‘external form’ of the particular divinity or, for that matter, of a class of demons is quite inappropriate. The idol serves as the substitute of the God in worship and ritual – it is a man-made god in precisely the sense that the hobby horse is a man-made horse; to question it further means court deception.39

The exemplary representations that Gombrich mentions here have in common that they represent an absent body. Think also of how the substitution 35 Gombrich, Meditations on a Hobby Horse, 3-5. 36 The question: ‘How do we understand a representation?’ is not answered by referring to some theory of representation but by referring to the society we grew up in and how we got acquainted with the symbolic systems that are in use in that society. 37 A key work in historiography supporting this claim is Louis Marin’s Le Portrait du Roi (Paris, Éditions de Minuit, 1981). Other historical examples of representations taking the place of an absent body and the history of this type of representation can be found in Carlo Ginzburg, ‘Representation: The Word, the Idea, the Thing’ in his Wooden Eyes. Nine Reflections on Distance (London and New York: Verso 2002), 63-78. 38 This essay titled ‘Meditations on a Hobby Horse or the Roots of Artistic Form’ is reprinted in Gombrich, Meditations on a Hobby Horse. 39 Ibid., 3.

92 

The Exemplifying Past

theory helped us understand Aurelius’ statue with which we began this chapter. Aurelius’ statue was meant to take the place of the real Aurelius and function as a substitute for him. Also think of how members of parliament take the place of the people they represent. Or think of Hyacinthe Rigaud’s famous portrait of Louis XIV, which was placed on the throne when the king himself was absent. It was a huge insult to turn one’s back to the portrait, just as it was a huge insult to turn one’s back to the king. We cannot generalize the substitution view to all representations. A subway map is no substitute for the subway and a label on a tube of paint is no substitute for the paint in the tube, although the subway map represents subway stations and routes and the label represents the paint in the tube. All representations stand for what they represent, but only those representations that stand for absent bodies function as substitutes for those bodies. It follows, as argued in this chapter, that historical representations are no substitutes for the past they represent. If we want to account for the relation between historical representations (narratives) and the past, we should turn to the theory that representations express something about what they represent. Representations that express something about what they represent are, as Danto has it, metaphors for what they represent.40 Historical representations too express something about what they represent, and therefore, they can be said to be metaphors for the past. This is something that Danto never realized. Ankersmit, on the other hand, did. It is a central claim of his narrativist philosophy of history. Like metaphors, historical representations, i.e. narratives, propose a point of view, a ‘seeing-as’ suggesting how the past ought to be viewed. 41 Like statues and paintings, narratives are works, and like narratives, statues and paintings have us see the world in a certain light. To put it in the idiom of the conclusion of our previous chapter: the statue of Aurelius shows us Aurelius as a stoic harbinger of peace.

5 Metaphor Aurelius’ equestrian statue represents him as a victorious emperor who shows clemency to a barbarian king with the gesture he makes. It expresses 40 Danto, The Transfiguration of the Commonplace, 194. 41 See his Narrative Logic. A Semantic Analysis of the Historian’s Language (The Hague: Martinus Nijhoff Publishers, 1983). See also in relation to Danto’s view on representation and metaphor, Ankersmit, Meaning, Truth, and Reference, 74.

Resembl ance, Substitution, Expression

93

that Aurelius is a dignified stoic harbinger of peace. This is how the statue shows us Aurelius and how he is to be seen. Our attitude towards him should follow from this. In order for the sculptor to succeed, Aurelius has to become to us a stoic harbinger of peace. This is where our analysis has brought us. It is supported by how Goodman, and in his footsteps Danto, relates the concept of expression to metaphor.42 Representations that express something about what they represent – and not all representations do so – are metaphors for what they represent. The latter is what Ankersmit argues in the context of historical representation. The statue of Aurelius is a metaphor for him being a stoic harbinger of peace. On the metaphoric structure of representations Danto writes: When Napoleon is represented as a Roman emperor, the sculptor is not just representing Napoleon in an antiquated get-up, the costumes believed to have been worn by the Roman emperors. Rather the sculptor is anxious to get the viewer to take toward the subject – Napoleon – the attitudes appropriate to the more exalted Roman emperors – Caesar or Augustus […]. That figure, so garbed, is a metaphor of dignity, authority, grandeur, power, and political utterness. Indeed, the description or depiction of a as b always has this metaphoric structure. 43

I know of one sculpture of Napoleon being represented as a Roman Emperor, the 1808 sculpture Napoleon in Triumph by François-Frédéric Lemot. (There are also busts of Napoleon wearing a laurel leaf crown; and an interesting neo-classicist statue made by the renowned sculptor Antonio Canova in 1806, representing a heroic nude of Napoleon as Mars the peacemaker, which Napoleon found trop athlétique and ordered not to be shown in public.) Better known are the paintings representing Napoleon as a Roman emperor, for instance the one by Jean-Auguste-Dominique Ingres, who represents Napoleon as Charlemagne, a reference that also features in the sculpture by Lemot.44 Napoleon as Charlemagne expresses the grandeur and dignity of the emperor: that is what the painting is a metaphor for. It is a way of 42 Goodman, Languages of Art, 85-95. Danto, Transfiguration of the Commonplace, 189-194. Here I follow Danto rather than Goodman. The reason for this is technical and has to do with how Goodman relates the notion of expression to the notion of exemplification. See below, n.48. 43 Danto, The Transfiguration of the Commonplace, 167. 44 The metaphor ‘Napoleon as Roman emperor’ can also be related to the Colonne Vendôme at the Place Vendôme in Paris, the Arc du Triomphe and the opera ‘Le Triomphe de Trajan’, first performed in 1807, in which Trajan unmistakably represents Napoleon. ‘Napoleon as Trajan’ is a metaphor for the mercifulness of Napoleon.

94 

The Exemplifying Past

establishing his authority, carefully avoiding any possible connotations of that authority with the Bourbon dynasty. It should be noted that the representation of a as b, of seeing a as b, has nothing to do with perception. Seeing or representing-as is not seeing or representing-that.45 All seeing-that may be expressed in so-called observation sentences: sentences that we would agree upon outright on witnessing the occasion.46 We see that Napoleon sits on a throne, that he wears a golden laurel wreath and a chain around his shoulders, that there is an eagle on the carpet, and that he has a staff in both his hands. Perhaps we know that the sceptre in his right hand is Charlemagne’s – and this sceptre is something we see too. We do not see that it is Charlemagne’s, unless we know it is his, just like we not see Napoleon, unless we know that the portrait is his. If a person lacks the appropriate concepts to describe what he sees – for example because he has never seen or heard of Charlemagne, and therefore, he cannot make the assertion that Napoleon has Charlemagne’s staff in his right hand – others are able to explain to him that the sceptre is Charlemagne’s, so that he learns what it is that he sees. Acquaintance with the symbolic system used and the ability to refine one’s sense of the ins and outs of the symbolic system is a condition of taking one thing to be a representation of some other thing. The specific way Ingres presents Napoleon to the viewer – as Charlemagne – prompts us to conceive of Napoleon as an heir of the Carolingians, and in virtue of that, of the Roman emperors, and take the appropriate attitude 45 The notion ‘seeing-as’ has hardly been theorized. The exception is the analysis of Ludwig Wittgenstein in his Philosophische Untersuchungen. His analysis starts with the observation that we either see a duck or a rabbit in Joseph Jastrow’s famous drawing. Seeing a duck or seeing a rabbit are simply two different perceptual experiences, and I can describe what I see (a rabbit) and what it is that I see (a rabbit-drawing). I cannot, however, see Aurelius as Aurelius, fork and knives as fork and knives, or an F as an F. ‘Das “Sehen als …” gehört nicht zur Wahrnehmung’ [‘Seeing-as …. is not part of perception’], concludes Wittgenstein, Philosophische Untersuchungen (Suhrkamp 1967), 231. Wittgenstein further observes (Untersuchungen, 241) that ‘‘‘Ich sehe es jetzt als ein …“ geht zusammen mit “Ich versuche, es als ein .. zu sehen”, oder “Ich kann es noch nicht als ein … sehen”. Ich kann aber nicht versuchen, das konventionelle Bild eines Löwen als Löwen zu sehen, sowenig wie ein F als diesen Buchstaben. (Wohl aber z.B. als einen Galgen.)’ (‘‘‘Now I see it as a …” is compatible with “I try to see it as a …”, or “I cannot yet see it as a ….”. However I cannot try to see the conventional image of a lion as a lion, neither can I see an F as this letter. (Although I can see it as a gallows for example.))’ [my translation]. ‘seeing-as’ has nothing to do with perception. It should be noted that in this analysis, the metaphorical locution ‘a is as b’, which prompts us to have some thought or take some attitude towards what is represented, is not accounted for. 46 On observation sentences, see W. Quine, Pursuit of Truth. Revised Edition (Cambridge and London 1992), 3-5. See also the comments made about this notion in Chapter 2 of this book.

Resembl ance, Substitution, Expression

95

towards him. Representations cause the interpreter to have an attitude by presenting someone or something in a certain light. That is what makes representation an instrument of rhetoric. 47 To be sure, we may refuse to take this attitude, as did the viewers in 1806 when Ingres’ painting was first exhibited, but such refusal is only possible after the paining is interpreted as expressing something about what it represents.

6 Conclusion All representations stand for what they represent, and sometimes this is all they do. Think of maps, labels, swatches, names, road signs, and logos. With these representations, users may plan routes, determine what suit to order, decide who to call, partake in traffic, and buy their favourite coffee. Some representations take the place of what they represent and function as their substitute. Here one should think of members of parliament, the effigy of kings, hobbyhorses, idols, and statues of sovereigns. These representations have in common that they represent an absent body. Some representations express something about what they represent. This is the class of rhetorical representation to which both artistic and historical representations belong. The term missing in this chapter is ‘exemplification,’ which may surprise since the overall claim of this book is that to see the past in a certain light is to have the past exemplify a historical thesis. How then, are the notions of expression and exemplification related? The expression theory of representation suggests the following relation. If a represents b, then, according to the expression theory of representation, (i) a stands for b, (ii) a is not b, (iii) a expresses c about b, and (iv) b exemplifies c. The statue of Aurelius (a) represents Aurelius, the man of flesh and blood (b): it (i) stands for him, (ii) we do not mistake the statue for the human being, and (iii) the statue expresses that Aurelius (b) is a stoic harbinger of peace (c). If the sculptor succeeds, i.e. if the statue is interpreted correctly, Aurelius becomes to us a stoic harbinger of peace. Aurelius (b) – the man of flesh and blood – then (iv) exemplif ies being a stoic harbinger of peace (c). It should be noted that exemplification depends on the notion of expression here. Only if a expresses c about b, b exemplifies c. This does not rule out that a and b may exemplify all sorts of other things apart from what b exemplifies relative

47 Danto, The Transfiguration of the Commonplace, 165.

96 

The Exemplifying Past

to what a expresses. 48 We may, for example, state that Aurelius’ statue exemplifies being an equestrian statue and that Aurelius exemplifies being an emperor. The point is to distinguish between these two different forms of exemplification. When I use the term exemplification in this book, I have its relation with expression in mind. Historical representations express a thesis about what they represent. If some narrative represents the past, then this narrative (i) stands for it, (ii) is not mistaken for the past reality it represents, (iii) expresses a thesis about the past, and (iv), the past as represented exemplifies the thesis. Take for instance Burckhardt’s famous Die Kultur der Renaissance in Italien. This narrative represents fourteenth and fifteenth-century Italy and expresses the thesis that at that time man becomes a self-conscious individual and acknowledges himself as such (‘der Mensch wird geistiges Individuum und erkennt sich als solches’). 49 This is what makes Europeans modern. The thesis tells us how we ought to conceive of the fourteenth and fifteenth centuries and their place in the history of Europe. It does not assert what that period is. The calculated self-interest of the Italians, their thirst for glory and sense of moral responsibility are but a few aspects of Italian culture that exemplify man becoming a self-conscious individual.50 By means of 48 In his Languages of Art, Goodman distinguishes between mere exemplif ication, as for example a tailor’s swatch exemplifies the colour of some clothing, and what he calls metaphorical exemplification. On the latter he claims that ‘What is expressed [c] is metaphorically exemplified [a exemplifies c]. What expresses sadness [a expresses c] is metaphorically sad [a is sad].’ Goodman, Languages of Art, 85. On this claim, Danto wrote: ‘I have transformed insidiously the suggestion of metaphorical exemplification into the thought that what a work expresses is what it is a metaphor for.’ Danto, Transfiguration of the Common Place, 194. We agreed with Danto’s reformulation, but it should be noted that the term exemplification disappears in this reformulation. I relate expression and exemplification thus: If a expresses c about b, then a is a metaphor for b, and b exemplifies c. 49 The full thesis runs thus: ‘Im Mittelalter lagen die beiden Seiten des Bewußtseins – nach der Welt hin und nach dem Innern des Menschen selbst – wie unter einem gemeinsamen Schleier träumend oder halbwach. Der Schleier war gewoben aus Glauben, Kindesbefangenheit und Wahn; durch ihn hindurchgesehen erschienen Welt und Geschichte wundersam gefärbt, der Mensch aber erkannte sich nur als Rasse, Volk, Partei, Korporation, Familie oder sonst in irgendeiner Forme des Allgemeinen. In Italien zuerst verweht dieser Schleier in die Lüfte; es erwacht eine objektive Betrachtung und Behandlung des Staates und der sämtlichen Dinge dieser Welt überhaupt; daneben aber erhebt sich mit voller Macht das Subjektive; der Mensch wird geistiges Individuum und erkennt sich als solches.’ Burckhardt, Die Kultur der Renaissance in Italien (Vienna [1860]), 76. 50 To make things more complicated. The artworks discussed in Burckhardt themselves express something about what they represent, while Burckhardt’s narrative expresses something about the artwork it represents. Often, a work of history expresses some thesis about some work of art. If so, the artwork a1 expresses c1 about b1, and b1 exemplifies c1; and the work of history a2

Resembl ance, Substitution, Expression

97

Burckhardt’s work, they have become examples of man becoming a selfconscious individual, in the same way that by means of his statue, Aurelius became an example of a dignified, stoic bringer of peace. Burckhardt’s work is a metaphor for the fourteenth and fifteenth centuries and it represents them as a Renaissance. This is how he sheds light on the past. We may add that such a view of the past is not, once presented, accepted once and for all: it tends to shed light on the past as long as no other view is able to absorb the light of its rival.51 It is time we have a closer look at this exemplification theory of history.

expresses c2 about the artwork a1=b2 , and the artwork a1=b2 exemplif ies c2 . The painting of Antoine Watteau discussed in the next chapter is an example of this. 51 Cf. how Burckhardt’s view contrasts with Johan Huizinga’s. See Harry Jansen, ‘Rethinking Burckhardt and Huizinga. A Transformation of Temporal Images,’ Storia della Storiografia, 70(2) (2016), 95-112.

5 Exemplification 1 Introduction In her excellent study The Triumph of Pleasure. Louis XIV & the Politics of Spectacle, Georgia J. Cowart sets herself the task ‘to reveal the forces that effectively transformed the celebration of the monarch into the utopian celebration of public entertainment as a new societal model.’1 This new, libertine societal model, emerging in the late seventeenth century, was based on the ideals of love, social harmony, equality, and freedom. If it were to come into being, pleasure would triumph. On the painter Antoine Watteau she writes: his Pilgrimage to Cythera, based directly on the imagery and ideology of two ballets produced at the Opéra and related works at the théâtre de la foire, may be seen as the most complete expression of an operatic, proto-Enlightenment vision of an alternative, utopian society.2

This meaning of Watteau’s work, painted in 1717-1718, goes beyond the artist’s intentions for obvious reasons. Watteau, as any other mortal, could not have foreseen the future, even if the Enlightenment was about to unfold right in front of him. He may have intended to express a vision of an alternative society; he may even have intended to anticipate what was about to come: that would make him in retrospect a visionary; but he could not have intended to anticipate what was later to be known as the Enlightenment. For only after the events we have learned to associate with the Enlightenment took place did it make sense to speak of a proto-Enlightenment. Perhaps we should criticize Coward for ascribing something to Watteau which he never could have said himself, and ask her to change her language and discuss the past in its own terms only. Should the historian not confine herself to making statements that witnesses of the events could have made? All historians need to do – and this is a task already difficult enough – is to establish a record of what happened in the past, including the experiences and observations of contemporaries, as complete as the past remains and the appropriate methods of studying those remains allow it to be. Is not 1 G. Cowart, The Triumph of Pleasure. Louis XIV & the Politics of Spectacle (Chicago: University of Chicago Press, 2008), xv. 2 Ibid., xviii.

100 

The Exemplifying Past

such a complete record of all that happened in the past what historians in the end should strive for? Arthur Danto contemplated such a complete record in his Analytical Philosophy of History and called it the Ideal Chronicle. It is a list of all events in the past in their chronological order, stating absolutely everything about it. The list is recorded by the Ideal Chronicler: He knows whatever happens the moment it happens, even in other minds. He is also to have the gift of instantaneous transcription: everything that happens across the whole forward rim of the Past is set down by him, as it happens, the way it happens.3

Such a complete record would thus fulf il the dream of the Rankeans and show the past as it actually was, wie es eigentlich gewesen. 4 The Ideal Chronicle, however, has one important and decisive shortcoming, as Danto points out: it does not contain any information on later events with which the recorded events are connected. After all, the Ideal Chronicle contains only descriptions of events that could have been made at the time of their occurrence. It follows that the description of an event at t-1 will always be incomplete since that event can always be re-described in terms of events occurring later in time. Such re-descriptions, which Danto calls narrative sentences, cannot be part of the historical record made by the Ideal Chronicler.5 With his narrative sentence, Danto draws our attention to the re-description of an earlier event in terms of a later event, both of which are in the past of the describer and thus both have to have occurred. The description ‘Jones was planting a rose,’ therefore, is not a narrative sentence: its truth only requires the first event (the planting) and not the second and later event (the seedling having grown into a rose) to have taken place. The sentence ‘Jones was planting a prize-winning rose,’ on the other hand, is a narrative sentence, since here the later event re-describes the earlier event and both the second event (winning the prize) and the first event (planting the rose) must have occurred for the sentence to be true.6 So Adrian Haddock 3 A.C. Danto, Narration and Knowledge (New York: Columbia University Press, 1985), 149. This book includes the integral text of the revised edition of his 1965 Analytical Philosophy of History. 4 Here I misuse Leopold von Ranke’s much quoted dictum, for the histories that Ranke wrote and his ideas about it are nothing like the Ideal Chronicle. 5 Danto, Narration and Knowledge, 151-152. 6 Ibid., 164-165.

Exemplification

101

misinterpreted Danto when he wrote that descriptions of actions in a past tense are narrative sentences.7 The description ‘Jones was planting a rose’ is not a narrative sentence. Since the list of event descriptions compiled by the Ideal Chronicler only contains descriptions of events that are made at the time of their occurrence, it lacks descriptions that connect events with events later in time. The Ideal Chronicler, in other words, has no knowledge of the future. To be sure, the historian lacks such knowledge too, but she has knowledge of the future of the past lives she studies, while the Ideal Chronicler has not. There is a second, related difference between the Ideal Chronicler and the historian. The Ideal Chronicle knows everything that happens the moment it happens the way it happens, including what happens in the minds of people. The historian is not only interested in what happened in the past and the minds of its inhabitants, and what later events some event is connected with; she also wants to know what event is relevant for what reason. If the Ideal Chronicler were to be in the position of the historian, he would have to know ‘which future events are relevant, and this requires predicting the interests of future historians.’8 Danto introduced the narrative sentence to emphasize the retrospective nature of historical knowledge, and more specifically, of historical narratives which make evident why some event is historically significant. Now, narratives are not simply lists of narrative sentences, but a narrative sentence in a narrative is an indication that some event is for some reason significant.9 Danto writes: To ask for the significance of an event, in the historical sense of the term, is to ask a question which can be answered only in the context of a story. The identical event will have a different significance in accordance with the story in which it is located or, in other words, in accordance with what different sets of later events it may be connected.10 7 A. Haddock, ‘Danto’s Dialectic,’ Philosophia (2008), 483-493, at 489. 8 Danto, Narration and Knowledge, 169. 9 Ibid., 167. Here Danto distinguishes clearly between narrative sentences and narratives, as he does elsewhere in his book, although most effort goes to analyzing the f irst. Some scholars make it appear that Danto ‘only’ discusses narrative sentences and takes narratives to be nothing more than sets of such sentences. See for example Lydia Goehr, ‘Afterwords: An Introduction to Arthur Danto’s Philosophies of History and Art,’ History and Theory, 46 (2007), 1-28, at 22. 10 Danto, Narration and Knowledge, 11. On how to interpret the ‘identical event,’ see the discussion in Chapter 3, section 5.

102 

The Exemplifying Past

This is in a nutshell the work of a historian: to narrate earlier events in terms of the later events with which they are connected, thereby exposing their historical significance. Or, as Danto concludes his Analytical Philosophy of History: finding out the historical significance of events is ‘the work of historians: history is made by them.’11 Asking Cowart to change her language and discuss the past in its own terms only would therefore be like asking her to stop being a historian. This we should not do. Rather we should reflect on the distinction between descriptions of the past as they can be found in the Ideal Chronicle’s list and retrospective understandings of the past as they can be found in the historian’s narrative.12 The exemplification theory of history that is proposed in this book is a theory about this distinction. This theory holds that the past as represented in some narrative exemplifies the historical thesis of that narrative. As such the theory explains how the past receives its meaning. The emergent proto-Enlightenment in the late seventeenth century, emphasizing, among other things, social harmony, equality, and the wish to participate in the pleasures and manners of what was then known as noble gallantry, is such a historical thesis. The past actions and what resulted from them mentioned in Cowart’s narrative – Watteau’s Pilgrimage to Cythera, but also for instance the ballets of André Campra and the utopian novels she discusses – all exemplify this thesis.13 Watteau’s painting, the ballets of Campra, and the utopian novels discussed, derive their historical meaning from the historical thesis that her narrative expresses. Of course, past events were meaningful for those individuals living through those events, as did Watteau’s painting, but the historical meaning of past events and artifacts such as paintings is only for historians to see. It is up to them to determine what later events some event is connected with and how such connection brings their significance to the fore. Such is the work of the historian. The painting of Watteau is a case in point. It expresses a 11 Ibid., 284. 12 This is also what Frank Ankersmit recommends. See his ‘Danto, History, and the Tragedy of Human Existence,’ History and Theory, 42 (2003), 291-304, at 298. 13 Jouni-Matti Kuukkanen recently argued that exemplification is an epistemic value that enables us to evaluate historical narratives. See his, Postnarrativist Philosophy of Historiography (Basingstoke: Palgrave Macmillan, 2015), 123-126. This I have no quarrel with, although it is not how I use the term. Curiously, when Kuukkanen uses the term to analyze the work of historians, he uses it in the manner in which I am in habit of using it, and not as an epistemic value. See his Postnarrativist Philosophy, 83, 85 and 94. I introduced the term to analyze the work of historians, among others, in my ‘The Exemplification Theory of History. Narrativist Philosophy and the Autonomy of History,’ Journal of the Philosophy of History, 6(2) (2012), 236-257. This chapter is a thoroughly reworked and expanded version of that essay.

Exemplification

103

utopian vision of society. This is what the painting means to him and to his contemporaries. Its being a proto-Enlightenment painting and vision depends, however, on its relation to later events and its significance in the eyes of the historian. To be sure, our interest may confine itself to Watteau’s painting and what it expresses, but the painting cannot mean to Watteau and his contemporaries what it means to us for the simple reason that we live in another time than they did and know what happened in their future. The possible relations this painting has to other paintings and events is different for them. I will find support for the exemplification theory of history in the work of Danto, Louis Mink, and Frank Ankersmit.14 These authors all use notions that resemble the notion of exemplification and they hint at the distinction that will turn out to be crucial: the distinction between exemplifying social change and providing evidence for social change. This distinction will bring us to the heart of historical understanding. Danto and Ankersmit both emphasize the discrepancy between what events mean to witnesses and contemporaries of those events and what those events mean to historians. This discrepancy is what is commonly referred to as the autonomy of history with regard to the past, and a key claim of their narrativist philosophies of history and the exemplification theory of history. History is, in short, made by historians. In this chapter my aim is to argue for the exemplification theory of history and the claims and arguments I discuss of the authors mentioned support this theory, which is why I discuss them. It is not my aim to do justice to their respective philosophies of history and the possible differences therein. I will, however, characterize their narrativist philosophies of history as ‘Platonic’ and contrast them with their rival, the narrativist philosophies 14 The concept ‘exemplification’ is well known from Nelson Goodman’s philosophy of symbols. A tailor’s swatch functions as a sample and as such it exemplifies certain properties. See Goodman, Languages of Art. An Approach to a Theory of Symbols (Indianapolis and Cambridge: Hackett Publishing, 1976), 53. A swatch only exemplifies some of the properties it has. For example, it will not exemplify a certain shape. This notion of exemplification is different from what I have in mind in the following respect. I will argue that aspects of the past exemplify certain properties after being historically understood. Watteau does not exemplify the proto-Enlightenment because he is a proto-Enlightenment painter (and thus can be attributed the property of being such a person). Only after being historically understood (i.e. in terms of later events with which he is connected) does Watteau exemplify the proto-Enlightenment. Watteau thus only retroactively acquires the property of being a proto-Enlightenment painter. Another difference between my use of the notion of exemplification and Goodman’s is discussed in Chapter 4 of this book, n.48. I ended that Chapter with explaining how the notion of exemplification is related to the notion of expression.

104 

The Exemplifying Past

of such authors as David Carr and Paul Ricoeur, which I will characterize as ‘Aristotelian’. The chapter concludes with some strong reasons why we should prefer the Platonic view of narrative to its Aristotelian counterpart, and why these labels are appropriate.

2

Social Individuals

Near the end of his Analytical Philosophy of History, Danto distinguishes between sentences that refer to individual human beings and their behaviour and sentences that refer to what he calls ‘social individuals’ such as social groups, organizations, large-scale events such as the Thirty Years War, and large-scale social movements such as the Reformation.15 It is hard to deny that both types of sentences are indispensable to the historian. However, the distinction does give rise to some important problems. One such problem is whether there really are such things as social individuals; one might argue that classes, groups, and large-scale events and movements are merely the sum of their constituent parts – individual human beings and their behaviour – and only those individual human beings and their behaviour are real. If one denies that social individuals exist, one can still admit that sentences that refer to these social individuals are valuable or even indispensable to historians, for it might be argued that such sentences simply state something about the individual human beings in the past who comprise the social individual that cannot be stated by referring to them severally. It is not a contradiction to be an ontological individualist and at the same time a methodological collectivist. Talk of social individuals might simply be shorthand for talk about the many individual humans and their many attitudes, desires, and beliefs which constitute them. How else would we be able to talk about the Thirty Years War, the Reformation, the Enlightenment, libertines, or France? Apart from the problem of whether social individuals exist or not, there is a second, more pressing problem we should attend to. It is best addressed by taking into consideration Danto’s treatment of Cicely 15 Danto, Narration and Knowledge, 258. This distinction reminds us of the double notion of individuality central to nineteenth-century historicists such as Wilhelm von Humboldt, Leopold von Ranke, and Wilhelm Windelband. On this double notion of individuality in historicism, see J. Bos, ‘Individuality and interpretation in nineteenth-century German historicism,’ U. Feest ed., Historical Perspectives on Erklären and Verstehen (Dordrecht: Springer Science + Business Media B.V., 2010), 207-220.

Exemplification

105

Wedgwood’s narrative of the Thirty Years War.16 Discussing this second problem will also provide us with an answer to the first. The central issue discussed by Wedgwood that Danto draws our attention to is the change of the Thirty Years War from a primarily religious conflict into a political one. Danto wonders how in a description of this change reference is made to the behaviour of individuals (for the moment we shall leave the question unanswered whether the Thirty Years War was changing itself or whether the change happened in or during the Thirty Years War). First he points out that when Wedgwood refers to individuals, she does so mainly to ‘illustrate this change, or to provide evidence that a change has in fact taken place.’ There is, I think, an important distinction to be made between illustrating social change and providing evidence for social change, one which Danto neglects. I will attend to this distinction below. For now I merely follow Danto’s own analysis. After this initial observation, Danto goes on to argue that the individual human beings mentioned by Wedgwood are selected for a special reason. They are not selected ‘because of any intrinsic interest,’ but ‘because of their historical significance: they make clear to us that a great change in attitude and behaviour of individuals in roughly the same social positions has taken place.’ This is what interests me here: past states of affairs mentioned in some historical narrative are mentioned because of their illustrative use, not because of their intrinsic interest. Danto gives the following example of what he has in mind, based on Wedgwood’s narrative of the Thirty Years War: The battle-cry shouted by soldiers at White Hill was ‘Sancta Maria!’ The battle-cry shouted, later, at Nordlingen, was ‘Viva Espaňa’. Those who might have witnessed these two battles would almost certainly not have seen the significance of these shouts. For the significance lies in the contrast between them, a contrast which is significant to an historian who sees in them signs that ‘insensibly and rapidly, the Cross gave way to the flag.’

Danto notes that very few of the changes associated with the overall change of the conflict from a primarily religious to a political conflict were intended by anyone. Whatever motives, attitudes, and desires individuals had, they could not have known the significance of what they were doing or living through.17 16 The work discussed is C.V. Wedgwood, The Thirty Years War (Pelican Books, 1957). 17 Danto, Narration and Knowledge, 262-263.

106 

The Exemplifying Past

Up to this point, the analysis is clear. It confirms what we said in the introduction to this chapter. Retrospection, Danto emphasized, is central to historical narratives and only in the context of a narrative is it clear why some event is historically significant.18 After remarking that historians select behaviour to highlight their historical significance, Danto claims that the change from a primarily religious to a primarily political conflict ‘may not have been reproduced within the biography of any single individual who lived through the change,’ for ‘the change took place, not in individuals, but in society.’19 This is, I think, an astonishing claim, for on the one hand Danto claims that the historian mentions individual human beings because of the significance of their attitudes and behaviour, which makes clear that a great change has taken place, while on the other he claims that the change did not take place in them nor may it be found in the biography of any of them. Apparently, there is nothing in their attitudes and behaviour that would count as evidence for that social change. Danto does not, however, draw this conclusion: he stops his analysis after making his astonishing claim.20 We may agree with Danto that witnesses and contemporaries cannot view their actions from the perspective of the historian. The soldiers obviously witnessed the battle cries but they could not see their historical significance. Wedgwood retroactively attributed this significance to their shouts. We may also agree with Danto that the behaviour and attitude of individuals are the proper objects of observation for the historian and that social changes cannot be observed as such in their behaviour and attitudes, nor in the remains they left behind. Wedgwood observed the battle cries of the soldiers, but the social change those shouts illustrate cannot be observed. She sees in those cries the change from a primarily religious conflict into a primarily political conflict: ‘insensibly and rapidly, the Cross gave way to the flag.’ This change is not empirically found in their behaviour, even though their behaviour illustrates that change. Historians single out behaviour and 18 David Carr complains that Danto’s emphasis on the retrospective view of the historian downplays the role of the views of eyewitnesses and historical agents. D. Carr, ‘Place and Time: On the Interplay of Historical Points of View,’ History and Theory, 40 (2001), 153-167, at 166. I do not think it does. The viewpoint of the agent and that of the historian differ radically, contrary to what Carr thinks. The historical significance of what agents do and witness cannot be part of the viewpoint of the historical agent. 19 Danto, Narration and Knowledge, 263-264. 20 He does draw this conclusion earlier in his book, see his Narration and Knowledge, 61, but there he does not discuss the issues he discusses in response to Wedgwood’s narrative and the question what social individuals are.

Exemplification

107

attitudes because of the significance they see in them with regard to the social change they are interested in. This we can conclude. Past attitudes and behaviour receive a historical meaning by illustrating a social change that, as such, cannot be found in the past itself, nor can it be inferred from its remains because it only comes into view in retrospect. Think of our examples. Cowart uses individuals and their libertine behaviour and beliefs to illustrate what she calls the proto-Enlightenment vision of a utopian society. Cowart sees in Watteau’s painting a proto-Enlightenment painting, which it was not and could not have been for Watteau. Watteau’s Pilgrimage to Cythera depicts the embarkment onto the island of Cythera, where Venus, the goddess of love, was born. The ship on the left side of the painting symbolizes the state, setting sail into a new direction. The depicted scene as a whole expresses the desire for equality, social harmony, libertine joys, and gallant codes of conduct. This meaning of the painting is, presumably, what Watteau intended with it, and therefore, within his grasp. This meaning is also within the grasp of his contemporaries and within our grasp. However, as a vision anticipating the Enlightenment, it is beyond their grasp, but not beyond ours. This proto-Enlightenment vision is not found in the past itself, i.e. it could not have been observed as such at that time, nor can it presently be inferred from its remains. It can only retroactively be seen in the past. This is how we should understand the claim that social change does not take place in individuals but in societies. Late seventeenth and early-eighteenth-century individuals illustrate a change in society of which not one of them was aware. Two issues need further consideration. The problem we began with is still on the table: can social individuals like the Thirty Years War, France, the Enlightenment, and libertines properly be said to have existed? Or, what do we mean when we say that they existed in the past? The second issue is to distinguish between illustrating social change and providing evidence for social change. This will also make clear whether the Thirty Years War changed itself or whether the change happened in the Thirty Years War.

3

Evidence and Illustration

The difficulty is to make sense of the claim that social change does not take place in individuals but in societies, while reference is made to the behaviour of individuals and their attitudes to show that this social change

108 

The Exemplifying Past

in fact took place. In what follows I will elaborate on this claim. It will be a key argument in support of the exemplification theory of history. Social changes are the primary concern of historians, and a social change is illustrated by the behaviour of individual human beings. Historians are interested in how societies develop over time and they study those societies by studying their remains. These remains are only understood as remains inasmuch as they can be related to individuals, their behaviour, attitudes, and desires that brought them about. What we should make clear is the distinction between referring to individuals to illustrate or exemplify social change and referring to individuals in order to provide evidence for social change. I propose to draw this distinction thus. If reference to the attitudes and behaviour of individuals is made to exemplify social change, then a change in society is at stake that only became apparent in retrospect, while if reference to the attitudes and behaviour of individuals is made to provide evidence of social change, then the identif ication of an aspect of social reality is at stake, proving that society was changing in the eyes of witnesses of that change. More needs to be said on this difference, but for now the following distinction suff ices. If a state of affairs E – an action, event, attitude, desire, belief, book, artwork and so on – illustrates or exemplifies a social change, then E is understood retrospectively in terms of the social change it illustrates. If some state of affairs E (or the report of it) is evidence for a social change, then E (or its report) empirically justifies the claim that the social change in fact took place. Not all social changes are either illustrated or empirically justified. The painting of Watteau is a proto-Enlightenment painting because it illustrates the proto-Enlightenment. We may wonder: Can Watteau’s painting also be used as evidence for the existence of the proto-Enlightenment? Does the painting justify the assertion that there was a proto-Enlightenment? The answer is that it does not, for here we are dealing with something that could not have been observed by Watteau and his contemporaries. No one could observe that Watteau’s libertinism was anticipating eighteenth-century Enlightenment. What could it possibly have meant to say in 1717 that Watteau was anticipating the eighteenth-century Enlightenment? The distinction between illustrating social change and providing evidence for it is not always as evident as this example suggests. It is a subtle distinction in the following case. Watteau’s painting is evidence of his libertine outlook, but his painting is no evidence of the history of libertinism, for that history is only illustrated by such things as paintings. Because the painting illustrates the proto-Enlightenment, the statement

Exemplification

109

that Watteau’s painting is a proto-Enlightenment painting is true. We should not, however, forget that the painting acquired this property as a result of Cowart’s retrospective understanding; it is not a property of the painting that can be empirically found by the historian in the past itself (or more precisely, it cannot be inferred from its remains). It follows that one of the truth-conditions of the statement ‘Watteau’s Pilgrimage to Cythera is a proto-Enlightenment painting’ is the reference to the work of Cowart.21 The question whether the change from a religious conflict to a political one was a change of or a change in the Thirty Years can now be answered. It does not turn out to be a very exciting question. In one sense the Thirty Years War changed itself: it changed from a religious conf lict into a political conflict. In another sense, however, the change happened in the Thirty Years War: individual human beings living during that time interval illustrate the change. In both these senses, the Thirty Years War is a social individual that is known in terms of the history that can be told about it. Social individuals, we may conclude, exist as products of historical understanding. Apart from this, we may still refer to social individuals in the sense given to it above: as statements about the attitudes and behaviour of individual human beings of which the social individual consists that cannot be stated by referring to those individual human beings severally. Only the behaviour of individuals can be observed. If individuals participate in social change or witness that change, and this is reported, we have evidence that society was changing in the eyes of contemporaries of that change. None of them, however, is able to see how the course of events will unfold, how later events are connected with the events they are witnessing, and what the historical significance is of what they witness. That is only for the historian to see, to whom their behaviour and attitudes exemplify a social change they could not be aware of themselves.

4

Mink’s Ingredient Conclusion

To further support the distinction between illustrating social change and providing evidence that a social change in fact took place, I will turn to a central argument of Mink. The argument is concerned with what he calls 21 This I also argued in Chapter 2, section 7, and Chapter 3, section 5. See also my ‘Uitspraak en/of representatie,’ Algemeen Nederlands Tijdschrift voor Wijsbegeerte, 103(4) (2011), 263-267.

110 

The Exemplifying Past

historical conclusions and what we call historical theses. Recall what we said about these historical theses. If E exemplifies a social change, then E is understood retrospectively in terms of the social change it exemplifies. Such exemplification requires a narrative. The battle cries studied by Wedgwood exemplify the change of the war from a religious into a political conflict in her narrative. The painting of Watteau studied by Cowart exemplifies the emergent proto-Enlightenment in her narrative. These exemplified social changes are, in other words, the historical theses expressed by these narratives. Now let us turn to Mink. In his seminal essay ‘The Autonomy of Historical Understanding,’22 Mink argues that the conclusions (historical theses) of historical narratives are ingredient conclusions that cannot be detached from the events as they are represented in those narratives. This is what demarcates the writing of history from other scientif ic disciplines, where conclusions are detachable from the represented empirical content. 23 We already encountered this non-detachable quality of historical conclusions before. The distinction between some state of affairs E illustrating social change and referring to some state of affairs E to provide evidence that a social change took place is another way of formulating Mink’s distinction between the writing of history and other scientif ic disciplines. Think of our example. Cowart uses Watteau’s painting to illustrate the protoEnlightenment, not to provide evidence for the proto-Enlightenment, since, as we argued, what Watteau’s painting illustrates could not have been observed as such. The painting only retroactively, through Cowart’s doings, acquired the property of being a proto-Enlightenment painting, 22 Louis Mink, ‘The Autonomy of Historical Understanding,’ History and Theory, 5 (1966), 24-47. 23 This is what the ‘autonomy’ in the title of Mink’s essay refers to. Historical understanding is autonomous from the social sciences because history has ingredient conclusions whereas the social sciences do not. The need to distinguish history from the social sciences was strongly felt in the 1960s. The autonomy of history with regard to the (social) sciences is to be distinguished from the autonomy of history with regard to the past. These two senses of autonomy are not unrelated, and one moves easily from the first to the second. This so happens in Danto’s introduction, as he writes: ‘There are countless descriptions of any event, only under some of which is the event available for scientif ic explanation, which then connects description and explanation very closely. It is almost certain that the descriptions relevant to science will differ from those of importance to history, hence it is unlikely that an event can be covered by a law under the same description in which it is covered by a historical narrative. A certain autonomy then attaches to history, indeed to narrative history, which cannot become more “scientific” without losing its defining human importance since it is human interests, after all, which determine which events are important and under what sort of descriptions.’ Danto, Narration and Knowledge, xii.

Exemplification

111

and it only illustrates or exemplif ies the proto-Enlightenment relative to Cowart’s narrative. Therefore, as a proto-Enlightenment painting, Watteau’s painting cannot be detached from the proto-Enlightenment it exemplifies. Mink writes: The significant conclusions, one might say, are ingredient in the argument itself, not merely in the sense that they are scattered through the text but in the sense that they are represented by the narrative order itself. As ingredient conclusions they are exhibited rather than demonstrated. Articulated as separate statements in a grand finale, they are not conclusions but reminders to the reader (and to the historian himself) of the topography of events to which the entire narrative has given order. In this one respect at least, history is akin to poetry in its reliance on ingredient rather than detachable conclusions.24

In this passage we f ind the meaning of ‘illustration’ that we have been looking for all along. Historians exhibit rather than demonstrate their conclusions. Past attitudes, behaviour, and desires do not empirically demonstrate particular historical theses (conclusions). The attitudes, behaviour, and desires mentioned in the narrative exemplify – illustrate or exhibit – the historical conclusions of that narrative: that is why conclusions or theses of historians cannot be detached from the events and their order as mentioned in the narratives. Think again of our example. The conclusion that in late-seventeenth-century France a libertine outlook anticipated the eighteenth-century Enlightenment cannot be separated from the events exemplifying that outlook. What exemplifies cannot be detached from what is exemplified in the way that empirical content can be detached from the conclusion it supports. If Watteau’s painting ceases to exemplify the proto-Enlightenment, the painting will no longer be a proto-Enlightenment painting, whereas if the painting demonstrates the conclusion that a recurring theme in Watteau’s paintings is the depiction of people from different social classes, the painting and the conclusion are detachable. The painting does not stop depicting people from different social backgrounds if the conclusion turns out to be false. In contrast, 24 Mink, ‘The Autonomy of Historical Understanding,’ 39. The relation between history and poetry that the last sentence of this passage brings to our attention is extensively discussed in Chapter 3. Mink first proposed the term ‘ingredient conclusion’ in his unpublished 1962 paper ‘Historical Knowledge.’ See Richard T. Vann, ‘Louis Mink’s Linguistic Turn,’ History and Theory, 26(1) (1987), 1-14, at 2-3.

112 

The Exemplifying Past

the conclusion that the late seventeenth century was a period of protoEnlightenment is neither true nor false: it is exhibited in a narrative or not.25 With the distinction between exhibiting and demonstrating, or, as I have put it, between illustrating and providing evidence, we have entered the heart of historical understanding. The historian starts with studying behaviour and attitudes relative to some larger context such as a particular war, a form of social critique, political ideology, a revolution, the exercise of power, the management of a city, poor relief, and so on, but she will end up with an autonomous historical narrative in which the behaviour and attitudes studied exemplify a thesis on, for example, war, social relations, or poverty. This autonomy of the historical narrative is due to the fact that the historian can see in the behaviour and attitudes of individuals something that the individuals themselves and their contemporaries could not possibly see. What the historian sees in past behaviour and attitudes appears to be what once was referred to as the spirit of its age: the state of mind a society is in at a particular historical juncture. (As I promised in the introduction to this book: I make use of a contemporary idiom.) The term sounds somewhat antiquated, but inasmuch as past behaviour and attitudes exemplify a historical thesis, something manifests itself in that behaviour and those attitudes which was not apparent at the moment the behaviour took place and the attitudes were held. We have gained a better understanding of the exemplification theory of history. Theses on the past are exhibited rather than demonstrated, as Mink has it. Or, as I prefer to put it, historical theses are exemplified rather than justified by the available evidence, illustrated rather than proven.

5

Ankersmit’s Narrative Substance

The exemplification theory of history is further supported by the narrativist philosophy of Ankersmit. In his major and influential work Narrative Logic, Ankersmit argues that historical theses are panoramic interpretations of large parts of the past. Instead of historical thesis, Ankersmit also uses the more technical term ‘narrative substance’.26 Cowart’s thesis that a 25 To be sure, historical theses can be taken and claimed to be true. We have to distinguish between our semantic and our pragmatist vocabularies. This I argued in Chapter 3. 26 Ankersmit, Narrative Logic. A Semantic Analysis of the Historian’s Language (The Hague: Martinus Nijhoff Publishers, 1983), 16.

Exemplification

113

proto-Enlightenment vision of society emerged at the end of the seventeenth century is such narrative substance. As a thesis on that part of the past, it offers a panoramic interpretation of it. According to Ankersmit, historical theses (narrative substances) do not refer to identifiable ‘things’ or their aspects in historical reality. They have a purely ‘expository’ function; they are linguistic devices, auxiliary constructions with whose aid historians try to convey a maximally clear and consistent representation of the past.27

Here Ankersmit’s narrative substance resembles Mink’s ingredient conclusion.28 Narrative substances have an expository function, i.e. they exhibit rather than demonstrate a historical thesis, as Mink would have it. The claim that historical theses do not refer to past persons, their behaviour and attitudes, and the events they brought about and witnessed, does not mean that those theses have no relation with the past whatsoever. It is rather obvious that narrative substances are always linked to rather specific periods in the past. Wedgwood’s thesis on the Thirty Years War is concerned with the first half of the seventeenth century, and Cowart’s thesis on the proto-Enlightenment is concerned with the late seventeenth and early eighteenth century. The point is that ‘reference’ belongs to a semantics of statements about the past rather than to a semantics of narratives about the past.29 Ankersmit’s claim that historical theses do not refer to past reality supports the claim that historical theses are exemplif ied rather than justified by the available evidence. He almost states this claim himself when he observes that the states of affairs mentioned in some narrative have an ‘illustrative’ use – a term Danto too used in passing, as we saw. As illustrations of the historical thesis proposed in that narrative, these 27 Ibid., 104. Here Ankersmit implicitly suggests, as he does explicitly elsewhere (Narrative Logic, 81), that the past is amorphous and it is up to the historian to bring form and coherence to it in her narrative. It suggests a distinction between content and form which we should reject. This is one of the main arguments in Chapter 3 of this book. 28 Ankersmit agrees with Mink that historical theses (or conclusions) cannot be separated from the events and their order as represented in the narrative, but he misses the opportunity to identify Mink’s ingredient conclusion with his narrative substance. See Ankersmit, Narrative Logic, 46. Ankersmit is right to disagree with Mink’s conception of the narrative as a network of overlapping descriptions. See Narrative Logic, 48-49. This conception of the narrative is absent in the article by Mink I quoted from in the previous section. 29 See Chapter 3. This is also a central claim of Ankersmit’s Narrative Logic.

114 

The Exemplifying Past

states of affairs contribute ‘to a particular view of the past.’30 Obviously, such theses or views are not empirically found in the past itself (i.e. in its remains from which we infer their past existence); but since these theses are proposals as to how we should see the past, as Ankersmit emphasizes throughout his Narrative Logic, they are obviously connected to the past they are a view of. Think once again of Cowart’s The Triumph of Pleasure. Cowart illustrates her thesis with such things as Watteau’s painting, utopian novels, and the ballets of Campra. All these things are part of past reality to which we can refer, but only in the context of her narrative do they illustrate her thesis. Her narrative is, in other words, a proposal to see these things in a certain light. Ankersmit gives several examples of narrative substances, including intellectual movement, Renaissance, social group, Industrial Revolution, states, and revolutions.31 These entities we encountered before. They are what Danto called social individuals. The examples that Ankersmit offers of narrative substances, together with the claim we quoted that they do not refer to ‘identifiable ‘things’ or their aspects in historical reality,’ easily lead to misunderstandings. So John Zammito, among others, argues: for most historians, the notion that the ‘historical idea’ [narrative substance] is entirely fictive, ontologically restricted to the representation and without any claim to actuality in the past, goes too far. Poland, however unstable its borders, however interrupted by partition, is not just our metaphor: it actually existed and we know that. Bourgeoisie may be harder, and Renaissance harder still, but the practicing historian’s intuition needs to be taken extremely seriously.32

‘Poland’ is indeed a proper name that is commonly used to refer to a specific country. It can, for example, be found on geographical maps, or on a list of the member states of the European Union. But this way of using the name ‘Poland’ must be distinguished from using it to refer to the history of Poland. The same is true of the terms ‘bourgeoisie’ and ‘Renaissance.’ The issue is not whether terms such as ‘Poland,’ ‘bourgeoisie,’ and ‘Renaissance’ refer to something in (past) reality: we can use those terms for doing that; rather the issue is to distinguish between what the term refers to and the 30 Ibid., 128. 31 Ibid., 82, 101. 32 J. Zammito, ‘Ankersmit and Historical Representation,’ History and Theory, 44 (2005), 155-181, at 164.

Exemplification

115

history of what the term refers to. This is a distinction between referring to something in past reality and referring to a narrative substance (the latter which does not refer itself). If the term is used to refer to something in past reality and its properties, its meaning differs from when it is used to refer to a narrative substance and its properties. If one does not make this distinction, one fails to understand what Ankersmit attempts to argue. It is the central distinction of his narrativist philosophy of history and an important insight into the nature of history. The difficulty is not to confuse terms and what they can be used to refer to with the history of what the terms can be used to refer to. Whether such confusion is likely to arise depends on the term under consideration. Some narrative substances have a proper name of their own, others have not.33 Usually large-scale events and social movements have a name of their own. Think of terms such as the Renaissance, the Thirty Years War, the Proto-Enlightenment, and the Industrial Revolution. These are names of narrative substances, and when we use these names, we usually refer to the narrative substance embodying their history. Here confusion is not likely to arise. To be sure, we can also use these terms to refer to a chronometric unit, but anyone with a rudimentary sense of history knows that periods are not mere chronometric units. If asked to individuate a term like Renaissance, a person may start with mentioning (and thus referring to) specific persons, dates, events, conditions, artifacts, techniques, buildings, cities, paintings, and opinions; but he will soon find himself to have moved from referring to past reality to referring to a narrative substance, i.e. to how this period is to be viewed historically. The confusion between what a term refers to and the history of what the term refers to is more likely to arise in case the narrative substance does not have a proper name of its own. Think of terms such as ‘poverty,’ ‘social critique,’ ‘ideology,’ and ‘state’. There are two basic ways to individuate a term such as ‘poverty’: we may refer to things in reality and their properties and we may refer to a narrative substance and its properties. Similarly, the phrase ‘Poverty in England’ may be individuated by referring to a variety of identifiable things such as specific poor laws, specific standards of living, specific attitudes, specific working conditions, and so on, but it also may be individuated by referring to a narrative substance embodying the history of poverty, which, for instance, makes clear that toward the end of the eighteenth century, not for the first but for the last time, ‘poverty

33 Ankersmit, Narrative Logic, 92.

116 

The Exemplifying Past

was removed from nature and brought into the forefront of history.’ This meant that All the changes coming to a head at this time – technological, economic, demographic, political, ideological – affected the poor to a greater degree than any other class and made their poverty more conspicuous, more controversial, and in a sense less ‘natural’ than it had ever been before. And this in spite of notable attempts to ‘naturalize’ poverty, to subject it, as Adam Smith did, to the natural laws of political economy, or, as Malthus did, to the biological laws of food and sex.34

Everything this thesis entails is part of the narrative substance on ‘poverty’ that Gertrude Himmelfarb proposes in her book. Individuating the phrase ‘Poverty in England’ by retelling her narrative is rather different from individuating that phrase by mentioning a series of specif ic opinions, specific authors and their theories, specific conditions, and so on. To be sure, Himmelfarb too refers to specif ic individuals and their opinions, and to specific changes in conditions of which contemporaries were well aware, but all these individuals, opinions, and changing conditions as they were experienced, are not merely referred to in order to create some sort of list of what happened when and where. In all the things she studied Himmelfarb sees how ‘poverty was removed from nature and brought into the forefront of history.’ Her thesis is not proven by the evidence, but exemplified by it. The point is that the same term (e.g. ‘poverty’) may be used differently. It follows that one cannot prove that the term ‘poverty’ refers to reality, even though one can use the term ‘poverty’ to do just that. Just like one cannot prove that the ‘Industrial Revolution’ refers to past reality, even though one can use the term to do just that. Therefore, one cannot refute the claim of narrativists such as Ankersmit and Danto that the history of, for example, poverty is autonomous with regard to past reality with an appeal to the argument that the meaning of the term ‘poverty’ is defined or fixed by reality itself. One would only be proving that a term can be used in a certain way (in this case that it can be used to refer to something in reality). What a term refers to is not to be confused with the history of what the term refers to, i.e. with the narrative substance embodying this history.

34 Gertrude Himmelfarb, The Idea of Poverty. England in the Early Industrial Age (New York: Alfred A. Knopf, 1984), 18.

Exemplification

117

Social individuals, to use Danto’s term for what Ankersmit calls narrative substances, are known through the history that can be told about them. In terms of narrative logic, Ankersmit argues thus. A narrative consists of statements on states of affairs in the past. These statements are components of the narrative and as such properties of the narrative substance proposed in that narrative.35 It follows that narrative substances can only be identif ied by a complete enumeration of all the statements it contains. This ‘holism’ sometimes is thought to be typical of narratives in general and Ankersmit’s views on narratives in particular.36 According to Ankersmit, all statements (s1, s2, s3 , sn) contained in a narrative should be read as statements on the narrative substance proposed in that narrative (‘N is s1’, ‘N is s2’, ‘N is s3’ etc.).37 Therefore, all statements in a narrative are analytical. This is the most fundamental theorem of his narrative logic.38 It is also a very precise formulation of what Mink called ingredient conclusions, and of the behaviour and attitudes of individuals (as described by s1 -sn) illustrating social change (N is s1-sn). Since every statement contained in a narrative asserts something about the past and at the same time attributes a property to the historical thesis proposed in that narrative, they have a double function: asserting that s (making a statement on a state of affairs in the past) and asserting that 35 Ankersmit, Narrative Logic, 94. 36 See Eugen Zeleňák, ‘Exploring holism in Frank Ankersmit’s historical representation,’ Rethinking History (2009) 13(3), 357-369; and Zeleňák, ‘Two Versions of A Constructivist View of Historical Work,’ History and Theory, 54(2) (2015), 209-225, at 214-215. See also Kuukkanen, Postnarrativist Philosophy, 8, 44-49, 76-80, and 95-96, who is critical about this holism. His (correct) criticism is that understanding a thesis does not require knowing everything that supports it (e.g. p.79). This leaves untouched the point, and Kuukkanen is well aware of that, that the individuals and their behaviour referred to in some narrative are part of the thesis expressed in that narrative. The sense of holism associated with narratives and Ankersmit’s views about it are not to be confused with semantic holism (Kuukkanen does make this unfortunate association, e.g. Postnarrativist Philosophy, 79). Perhaps it is simply better not to talk about holism in the context of narratives. On semantic holism, Donald Davidson once wrote: ‘Frege said that only in the context of a sentence does a word have meaning; in the same vein he might have added that only in the context of the language does a sentence (and therefore a word) have meaning.’ Davidson, Inquiries into Truth and Interpretation (Clarendon Press: Oxford, 2001 [1984]), 22. He added that to know the meaning of a sentence is to be able to know the ‘place of the sentence in the language as a whole’, i.e. ‘we would know the role of each significant part of the sentence, and we would know about the logical connections between this sentence and others’ (139). Our ability to read narratives depends on this sort of knowledge, but also on the ability to know the logical connections between sentences in a narrative. The latter is what narrativist philosophy of history is concerned with. 37 Ankersmit, Narrative Logic, 107-110. 38 Ibid., 127.

118 

The Exemplifying Past

N is s (attributing the property of including s to the narrative substance proposed in the narrative).39 I propose to rephrase this double function of statements in terms of the exemplification theory of history. 40 I define that theory as follows: If s is a statement on E (some state of affairs in the past) contained by the historical narrative N, then E (as described by s) can be said to exemplify the thesis expressed by N. The following passage of Cowart’s The Triumph of Pleasure may illustrate this: [i] Perhaps because of their apparent frivolity, the ballets of Campra and his collaborators have never been examined as vehicles of social critique or political ideology. [ii] Yet a study of their libretti reveals a carefully encoded dialogue between the opéra-ballet as a modern fête galante embodying the ideals of love, equality, and freedom, and the court ballet of the 1660s as an archaic fête monarchique embodying the ideals of absolutism, patriarchy, and sovereign praise. […] [iii] The public ballet of the Paris Opéra, under the guise of a libertinage de moeurs, a social libertinism that is shared with the court ballet of Louis’s early reign, also espoused a libertinage d’esprit, a political outlook serving as a bridge between seventeenth-century libertinism and eighteenth-century Enlightenment thought. 41

We should focus here on the sentences (ii) and (iii). In (ii) reference is made to the libretti of André Campra and his collaborators. Their libretti exemplify the historical thesis that is expressed in (iii): they are protoEnlightenment ‘things’. Danto would say that the libretti illustrate a change in society. Mink would say that the libretti are an ingredient and non-detachable part of the historical conclusion: they espouse a libertine spirit bridging seventeenth-century libertinism and eighteenth-century Enlightenment thought. Ankersmit would say that the statements attribute a property to the narrative substance that is known as the ‘protoEnlightenment’. Here we have in a nutshell the exemplification theory of history and the way it is anticipated by these three narrativist philosophers of history.

39 Ibid., 95. 40 Cf. my Beweren en Tonen. Waarheid, Taal en het Verleden (Nijmegen: PhD thesis Radboud University, 2009), 38-42, where I give additional reasons for doing so. 41 Cowart, The Triumph of Pleasure, 167-168.

Exemplification

119

6 Conclusion The past states of affairs that the historian mentions in her narrative exemplify the thesis expressed in that narrative. The thesis of Cowart that in the late seventeenth century a proto-Enlightenment emerged, emphasizing such things as social harmony, equality, and the desire to partake in gallantry, is such a thesis. Past states of affairs exemplifying a particular thesis are of course not stumbled upon. The historian starts by studying individual human beings and their behaviour, attitudes, and desires. Historians are interested in social change and the significance of behaviour and the associated events in relation to such change. In order to bring such significance to the fore, they connect earlier events with later events and aim to see in past behaviour something that individuals and their contemporaries could not see: how the change in their society manifested itself in their behaviour. This is where the historian moves away from her evidence, from which she inferred the past behaviour and attitudes, into the direction of a historical thesis, and why history can be said to be autonomous with regard to the past. The search for the relation between behaviour, desires, attitudes, beliefs, and artifacts on the one hand, and social change on the other, is guided by the presupposition that the latter becomes manifest in the former. The larger context to which historians aim to relate the behaviour they study – a war, social critique, ideology, a state – is only known in terms of its history. When Cowart studied Watteau’s painting, she did not discover that his painting was a proto-Enlightenment painting. On the one hand she saw in the painting something there was: a utopian view on society. On the other hand she saw something in the painting there was not: a utopian view on society anticipating the more fully developed and subsequent eighteenth-century Enlightenment’s view on society. While relating the painting to the larger context of social critique, she started to see how this painting contributed to the history of social critique by taking into account later events. Similarly, when Wedgwood studied the battle cries, she did not discover that the war changed from a religious conflict into a political conflict, rather she saw how in those cries this change manifested itself. That is what the battle cries retroactively exemplif ied. The Slovakian philosopher Eugen Zeleňák worries that the notion of exemplification as I use it results in a ‘passive bottom-up understanding of history by evoking the possibility that the past itself is the source of the theses it exemplifies.’ This, according to Zeleňák, threatens the autonomy

120 

The Exemplifying Past

of history. 42 Indeed, I take the past itself (i.e. the past as it is inferred from its remains or the remains themselves) to exemplify a historical thesis, but the past does so only retroactively, in relation to the historian’s narrative, for only afterwards is it clear to us how society changed and how the behaviour and attitudes of individuals exemplify the change they themselves could neither intend nor foresee. Therefore, Zeleňák has no reason to worry. Historical theses are not inferred from the behaviour and attitudes of individuals in the past, nor are they inferred from the past remains (to be sure, from past remains we do infer past behaviour and attitudes). Historical theses are exemplified rather than being justified by the available evidence. We now begin to understand why the narrativist philosophy of history of Danto, Mink, and Ankersmit, may be called Platonic. A historical thesis is not ascertained through a series of empirical tests that determine its validity, as if the truth of such thesis were to be established by means of verification.43 A historical thesis is ascertained by asking what is immanent to it, i.e. by asking how the thesis manifests itself in the course of events and what is intrinsic to this course. The thesis is to be seen in the behaviour and attitudes studied by the historian in order to have the behaviour and attitudes exemplify that thesis. It is here that what Mink referred to as an ingredient historical conclusion, what Danto referred to as a social individual, and what Ankersmit referred to as a narrative substance, is in agreement with the thesis of the eidos and where it is to be distinguished from the detachable conclusions of scientific disciplines. 44 The historical eidos manifests itself in relation to the historian’s narrative. It retroactively becomes concrete in the past studied by the historian. The Platonic view of narrative is not self-evident. The extensive argument given in favour of this view is proof of that. It contrasts with the narrativist philosophy of history of authors such as Paul Ricoeur and David 42 See Zeleňák, ‘Using Goodman to Explore Historical Representation,’ Journal of the Philosophy of History, 7(3) (2013), 371-395, at 393-394. Zeleňák critically discusses my ‘The Exemplification Theory of History’ in the context of the use of Goodman’s theory of symbols by philosophers of history. 43 Regardless of whether we think that verif ication is an inferential or a representational practice. On this distinction and the reasons why verif ication is an inferential practice, see Chapter 3. 44 On the distinction between the hypothesis in the sciences and the hypothesis of the eidos in Plato, see Hans-Georg Gadamer, Dialogue and Dialectic: Eight Hermeneutical Studies on Plato, translation and introduction by P. Christopher Smith (London: New Haven, 1980), 32-36.

Exemplification

121

Carr. Their narrativist philosophy may be characterized as Aristotelian. 45 Ricoeur and Carr deny the autonomy of historical narratives. Instead they focus on the concept of action which provides them with their model of narrative. According to this Aristotelian view, narratives are a mimesis of action. Ricoeur and Carr insist that to understand human action, we have to recognize its temporal structure. Actions have a pre-narrative structure, a temporal structure ‘that call[s] for narration,’ according to Ricoeur. 46 Carr sees a similarity between the motives-means-end structure of action and the beginning-middle-end structure of narratives. 47 Narrative is thus modelled on action and our understanding of and familiarity with action function as ‘translation rules’ that ‘translate’ past reality into the historian’s narrative. 48 Such translation rules are rejected by Danto and Ankersmit since they are at odds with the autonomy they claim for historical narratives. 49 Carr holds that according to the autonomy thesis, or ‘discontinuity thesis’ as he calls it, narratives distort reality.50 The autonomy thesis, however, does not imply this, for it ‘only’ underlines the discrepancy between what witnesses and contemporaries of events could know and what historians can know. Narratives do not distort reality: they help us understand it. Carr also believes that, according to the autonomy thesis, reality is a meaningless sequence,51 but the autonomy thesis as defended here does not imply this conception of reality. Past events obviously did have a meaning for the people living through those events: but that could not have been a historical meaning. Even a perfect witness to an event, such 45 Carr relates his narrativist philosophy explicitly to Aristotle in his ‘Narrative and the Real World: An Argument for Continuity,’ History and Theory, 25(2) (1986), 117-131, at 126. Ricoeur places Aristotle’s theory of the muthos as the mimesis of action as developed in his Poetics at the centre of his narrativist philosophy in his Time and Narrative, Vol. 1 (Chicago: The University of Chicago Press, 1984), 31 ff. Mark Bevir recently also presented a typically Aristotelian view on narrative. See his ‘Historical explanation, folk psychology, and narrative,’ Philosophical Explorations, 3(2) (2000), 152-168. 46 Ricoeur, Time and Narrative, see especially 59, 64, and 74. There is much richness in Ricoeur’s analysis. Such is made clear by Harry Jansen in his ‘Time, Narrative, and Fiction: The Uneasy Relationship between Ricoeur and a Heterogeneous Temporality,’ History and Theory, 54(1) (2015), 1-24. 47 Carr, ‘Narrative and the Real World,’ 122. This Carr also argues at several places in his Experience and History. Phenomenological Perspectives on the Historical World (Oxford: Oxford University Press, 2014), for instance at 68-69 and 110-111. 48 Ankersmit, Narrative Logic, 79. 49 Mink too rejects such translation rules, as we saw in Chapter 3. 50 Carr, ‘Narrative and the Real World,’ 117. See also Carr, ‘Time and Place,’ 165. 51 Carr, ‘Narrative and the Real World,’ 121.

122 

The Exemplifying Past

as the Ideal Chronicler, having access to the minds of all participants and knowing precisely what happens as it happens the moment it happens, could have no knowledge of the historical significance of that event. The historical significance of an event is never apparent the moment the event happens. (It is sometimes said of a current event that it is historic. Such a statement only makes sense if it is understood as a prediction about future retrospective views.) The autonomy thesis not so much contradicts phenomenological hermeneutics, as Carr believes, as it argues for the fundamental limits of hermeneutics. I have no quarrel with Carr’s claim that actions (or experiences) share (several or even many) features of narratives. Nor do I have a quarrel with the claim that to make sense of an action, we often tell a story. However, hermeneutics with its focus on the agent’s point of view cannot account for the history of the events that the agent has lived through. The autonomy thesis does not deny that, as Carr puts it, ‘events are charged with the significance they derive from our retentions and pretensions’: neither does it imply, as Carr believes it does, that life is a ‘mere sequence’.52 We may agree with Carr on both these matters without contradicting anything we have said about the autonomy thesis, as it is about the retrospective understanding of the events that agents have lived through; it is not a thesis about their experience of life. Therefore, we cannot agree with the following. Carr argues that ‘It is not only novelists and historians who view events in terms of their relation to later events, to use Danto’s formulation of the narrative point of view; we all do it all the time, in everyday life.’53 So much is certainly true, but it is irrelevant in the present context. The point missed by Carr is that if I witness and describe E1 at t-1 and re-describe E1 at t-2 in terms of E2, I make clear what E1 means to me at t-2, whereas if a historian re-describes E1 in terms of E2, he does so in terms of what E1 means to history at t-1 itself. The description of events that witnesses and their contemporaries can make, at the moment the events happened and in retrospect, is to be distinguished from the description of events that historians can make. The description of historians is informed by their thesis on the past. What past agents did and went through, their beliefs, desires, and attitudes, including their retrospective views, exemplify the thesis expressed in the historian’s narrative.

52 Ibid., 122. 53 Ibid., 125.

Exemplification

123

Coda As a coda to this chapter I want to argue that the exemplification theory of history is able to explain what Ankersmit calls ‘historical experience’. I will limit myself to his analysis of Francesco Guardi’s (1712-1793) painting Arcade with a Lantern with which he explains what a historical experience is. If someone were to object that the analysis I offer is not what a historical experience according to Ankersmit is, and therefore, the exemplification theory cannot explain what a historical experience is, my response would be that my interest here is only in the notion of historical experience insofar as it is consistent with the exemplification theory of history that this book argues for.54 At the centre of Guardi’s painting, six Pulcinellas are depicted. One of them is urinating, to the abhorrence of the bystanders. Three appear to make themselves a meal, and two Pulcinellas just sit on the ground, presumably from heavy drinking. The painting and, especially, the Pulcinellas, symbolize boredom, according to Ankersmit. That is what the painting is about. Ankersmit writes that Guardi does not invite us to see the ‘real’ world as if it were a theater (as in the theatrum mundi metaphor); no, he wants to inculcate in us the truth that life is both – real life sometimes is what it seems to be, but at other times it is a harsh comedy in the quite literal sense of the word.55

This is how Ankersmit interprets the painting. To support his interpretation, Ankersmit refers to Richard Sennett’s The Fall of Public Man,56 in which it is argued that in the eighteenth century, acting ‘was not restricted to the theatre, but it was part of behaviour in public, on the street.’ The need to integrate the roles played in theatre and public life into one substratum was not felt, and as a consequence, the idea that public life was a theatre led to the sentiment of boredom. Guardi’s painting, representing the theatre 54 A more adequate exposition of this notion, in the sense of being situated in the larger context of Ankersmit’s work on historical experience and the themes of history and experience, can be found in Martin Jay, Songs of Experience. Modern American and European Variations on a Universal Theme (University of California Press: Berkeley, Los Angeles, London, 2005), 216-260. Reading the section on Ankersmit (255-260) made me think that Ankersmit himself would disagree with the explanation I give of ‘his’ notion of historical experience. 55 Ankersmit, Sublime Historical Experience (Stanford: Stanford University Press, 2005), 271-274. 56 Sennett, The Fall of Public Man (Cambridge: Cambridge University Press, 1974).

124 

The Exemplifying Past

and public life in one scene, expresses this boredom. It brings us ‘right into the heart of eighteenth-century public life,’ according to Ankersmit.57 This is what he calls a historical experience. It is an experience he had when contemplating Guardi’s painting. Inasmuch as Ankersmit interprets the painting in accordance with what Guardi meant by it, the interpretation is something that obviously not only Guardi himself, but also his contemporaries would agree with. Ankersmit, however, also sees something in the painting that Guardi and his contemporaries were unable to see. In terms of the exemplification theory, we should say that Guardi’s painting exemplif ies or illustrates eighteenth-century public life, relative to Sennett’s narrative (it should be noted that Ankersmit does not seem to be aware of the fact that he is referring to a narrative substance, proposed by Sennett, as he explains his experience!).58 ‘Eighteenth-century life’ is a social individual (narrative substance) which can only be known through the history that can be told about it. This history cannot be found in the past itself; it can only be illustrated with such things as paintings. Ankersmit may have experienced the boredom expressed by the painting, but he could only identify this as an eighteenth-century sentiment resulting from public life being a theatre after understanding the painting retrospectively, as the Platonic view on narrative dictates. According to this view, Sennett’s thesis in his The Fall of Public Man has become concrete in Guardi’s painting. In this sense, Guardi’s painting is part of what Sennett’s book is all about, even though Sennett did not write about this particular painting. Ankersmit’s description of a historical experience as a ‘direct and immediate encounter with reality’59 can now be interpreted as the encounter we have the moment an object such as a painting draws us out of our present into an understanding of the past which it exemplifies. Immersing oneself in this understanding of the past is having a historical experience.60 57 Ankersmit, Sublime Historical Experience, 274-275. 58 In his inaugural lecture as Professor, Ankersmit is even more outspoken, and argues that Romanticism did integrate the roles played in public into one substratum, an authentic self that eventually destroyed the ancien régime. See De historische ervaring (Groningen: Historische Uitgeverij, 1993), 23. This is an even more exciting historical thesis in need of illustration. 59 Ankersmit, Sublime Historical Experience, 285. 60 This is not to be confused with the idea of immersing oneself in the past, as for example panorama paintings, open air museums, ‘experience rooms’ in museums, video games, and historical novels and films can be immersive. This is the sort of experience that gives the illusion of experiencing the past as then contemporaries experienced it. Cf. Chapter 2, n.33.

Exemplification

125

In our next and final chapter, we will discuss another example of a thesis on the past that became concrete in the past itself. Therein I will argue that we require a philosophical view of history to account for these theses. It will further support the central claim of this book that historical theses are exemplified rather than being justified by the available evidence. Not only historians present historical theses. Anyone who studies past events to find out their historical significance will at some point need a historical thesis to see this significance in those events.

Maarten Baas, Made in China

© Studio Maarten Baas / Pearl Lam Galleries / 2008 Photographer: courtesy of Pearl Lam Galleries, Shanghai

6

Danto’s End of Art That the chair should in recent days have entered art as a medium or a form, rather than as subject; that the chair should have become art […] strikes me as a sign that a certain barrier has been made visible by being broken. − Arthur Danto, Philosophizing Art1 [M]y philosophy of art would hardly have been the same under skies other than those of Manhattan in the latter decades of the twentieth century. − Danto, The Body/Body Problem2

1 Introduction Maarten Baas’s Made in China (2008), also known as Plastic Chair in Wood, is an exact wooden copy of the plastic lawn chair we all know so well. The object looks like a lawn chair: there can be no misunderstanding about that, but is it ‘merely’ a lawn chair in wood? If we were to come across it in a garden centre on a Saturday afternoon, would we take Made in China to be an object on which to sit? Perhaps we would think that it is a rather peculiar looking lawn chair because of its wood pattern, and even when closer inspection reveals that the chair is actually made of wood, we would have no reason to doubt our belief that it is a lawn chair that we have before us. On the other hand, perhaps we would think that it is a work of art. But why is a lawn chair in wood a work of art while its plastic equivalent is not? It appears that this requires ‘something the eye cannot de[s]cry – an atmosphere of artistic theory, a knowledge of the history of art: an artworld,’ as Arthur Danto had it in 1964.3 The question why of two perceptually indiscernible objects one is a work of art and the other is not, is central to Danto’s philosophy of art, which is mostly concerned with sculpture and painting. Pondering this question eventually resulted in the end-of-art thesis which he proposed in the early 1980s. Art had ended, in the Stable Gallery on East 74th Street in New York in 1 Arthur Danto, ‘The Seat of the Soul,’ Philosophizing Art. Selected Essays (Berkeley, Los Angeles, and London: University of California Press, 1999), 144-163, at 162. 2 Danto, ‘Introduction,’ The Body/Body Problem. Selected Essays (Berkeley, Los Angeles, and London: University of California Press, 1999), 1-18, at 4. 3 Danto, ‘The Artworld’, The Journal of Philosophy, 61 (1964), 571-584, at 580. This Danto also states in his The Transfiguration of the Commonplace. A Philosophy of Art (Harvard: Harvard University Press, 1981), 135.

128 

The Exemplifying Past

1964 to be precise, when Andy Warhol exhibited his now famous Brillo Box – a carefully crafted replica of the packaging of soap pads. Brillo Box is Danto’s favourite example of both the problem of two perceptually indiscernible objects, one of which is a work of art, and his end-of-art thesis. 4 This chapter argues that Danto’s end-of-art thesis presupposes a substantive philosophy of history, i.e. a philosophy of the history of art itself. Such philosophy explains the direction that art has taken, why that direction could not have been different, and why art had ran its course in 1964. Danto makes several somewhat scattered remarks about this philosophy of history, most of which have to do with Hegel’s view on art and history, but he never scrutinizes this philosophy to assess its credibility. This is what I aim to do in this chapter (without saying much about Hegel) by making use of the central arguments made earlier in this book. I must ask the reader for patience for the detour I make before arriving at this philosophy of history and the issues the reader has become familiar with. We cannot too hastily accept that the actual course of events has a direction which could not have been different and which has come to an end.5 This chapter starts with some Dantonian considerations to provide a context for the problem of perceptually indiscernible objects, one of which is a work of art. This will bring the historical dimension of art to the fore: art is made and interpreted in specific times and places, by specific artists for specific audiences, with specific technologies of art production at their disposal. We often acknowledge the historical dimension of what human 4 One could argue that, on close inspection, the eye could reveal a distinction between them, and therefore, the artwork is not perceptually indiscernible from its ordinary counterpart. However, even if someone is able to make a distinction between the two on the basis of stimuli hitting the retina alone, such distinction would be insufficient to establish that one is a work of art and the other is not. 5 The term substantive philosophy of history may surprise some readers. Not only because it refers to an outdated form of philosophy of history that is typically associated with Hegel, but also because Danto developed his own analytical philosophy of history in the 1960s in opposition to substantive philosophies of histories. Throughout this chapter I will be clear about what I mean by these terms and their usefulness. For now it suffices to say that Danto’s end of art thesis presumes a substantive philosophy of history which purports to explain why art developed as it did, why the direction into which it developed could not have been different, and why art came to an end. It should further be noted that this philosophy of history does not suppose that there is some purpose in this development. Finally, the distinction I will make between the philosophical and the common view of history can be interpreted as a replacement of the less adequate and somewhat outdated distinction between substantive and analytical philosophy of history. The distinction between substantive (or speculative) and analytical (or critical) philosophy of history was made in 1951 by W.H. Walsh in the first chapter of his Philosophy of History: An Introduction. Revised Edition (New York and Evanston: Harper & Row, 1968 [1951]).

Danto’s End of Art

129

beings do and bring forth, and without this dimension there would be nothing for historians to do, but such acknowledgment seldom implies a substantive philosophy of history and commits us to hold that the course of events has a direction which could not have been different. The historical dimension of art too does not imply a substantive philosophy of history. The end-of-art thesis on the other hand does. It is important to determine why this is so. Also because the historical dimension of art appears to be at odds with Danto’s more recent strategy to answer the question why of two perceptually indiscernible objects one is a work of art and the other is not by formulating the necessary conditions for something to be a work of art. These conditions imply that the criteria for something’s being a work of art are everywhere and always the same. We will find that this (apparent) contradiction between the historical dimension of art and the conditions for something to be art turns out to be no contradiction at all given Danto’s end-of-art thesis. Danto is, as he himself proclaims, both a historicist and an essentialist about art.6 The substantive philosophy of history, which Danto only tacitly accepts, helps us understand why this is so. Baas’s Made in China was made after the end of art. If Danto’s end-of-art thesis is correct, it should be able to explain its existence ‘on the other side of history’.

2

The Historical Dimension of the Artworld

To answer the question why a lawn chair in wood is a work of art while its plastic equivalent is not, we may start by noting that the wooden chair is not simply a wooden version of the plastic lawn chair, for if that were the case, it would not be a work of art. The chair would then only be a non-artistic imitation of the lawn chair. However, a work of art is not a version of what it imitates, and we are not supposed to take and treat Baas’s chair as an ordinary chair on which to sit.7 This is why most of us would presumably prefer to come across Baas’s chair in a museum or gallery rather than in a garden centre, for then the institutional setting and means of display would provide the right sort of ‘atmosphere’ to identify the object as a work of art (I wonder: has a museum visitor ever taken a warden’s chair for a work of art?). 6 Danto, After the End of Art. Contemporary Art and the Pale of History (Princeton: Princeton University Press, 1997), 95, 193. 7 The artist plays with the fact that a work of art is not supposed to be a version of what it imitates when he creates a trompe l’oeil effect. Curiously, too often a work that is meant to provide a trompe l’oeil effect is exhibited in such a way that the effect is not created, and a label next to the work is needed to tell the visitor of the intended effect.

130 

The Exemplifying Past

This is also how Baas would prefer it, I presume, for he made the chair for the artworld rather than the gardenworld. But even so, the question remains why a lawn chair in wood is a work of art while its plastic equivalent is not.8 Perhaps we should say that the chair is an artistic imitation of the plastic lawn chair, the sort of imitation we only find in the artworld, rather than a non-artistic imitation. But a moment’s thought makes us realize why this would be of no help. I do not doubt that Baas’s chair can be taken for an ordinary chair. I would however not dare to inform the person who believes that it is an ordinary chair that the chair is actually a work of art with an appeal to the theory that works of art are a mimesis, because if I would, I would have to indicate on the basis of what I believe that the wooden chair is an artistic imitation rather than a non-artistic one, and I do not know wherein lies the difference between an artistic and a non-artistic imitation of a plastic lawn chair.9 In any case, the difference is not a perceptual one, for Baas’s Made in China looks exactly like a plastic lawn chair, albeit one made of wood. Still, there has to be a difference, for Made in China is a work of art and not a mere lawn chair on which to sit. The theory that works of art are a mimesis does not help us with the problem why of two perceptually indiscernible objects one is a work of art and the other is not. Rather than a wooden version or an imitation, Made in China is a work of art that is appropriated. Baas did not design the chair himself: the designer of the plastic lawn chair did. Baas also did not carry out the manual work to reproduce the plastic chair in wood; Chinese craftsmen did that for him. Baas however did make the work, whereas the designer of the plastic chair and the Chinese craftsmen did not. The craftsmen he hired only made a chair in wood, faithfully copying the plastic model; and the designer of the plastic chair only created an ordinary chair on which to sit. To be sure, it has always been the case that the artist did not have to make the art object himself to be its author (think for instance of architecture, much of the performance arts, and large mural paintings), and copying designs has been a common 8 Because this question remains, the institutional setting in which the artwork is exhibited does not explain why of two perceptually indiscernible objects, one is a work of art and the other is not. Danto, Transfiguration of the Commonplace, 5, 99. It should be noted that his ‘The Artworld’ essay is not to be taken as a defence of the so-called Institutional Theory of Art, as George Dickie thought. 9 In book X of his Republic, Plato notoriously dismisses mimetic art. Interestingly, although he does not explicitly distinguish between artistic and non-artistic imitation, his main problem with art seems to be that the ordinary person is not able to distinguish between these two types of imitation and takes what is imitated to be what reality is or should be like, rather than as something which contrasts with reality, as is the case with artistic imitations. It reads as a warning on the corrupting influence of popular media on the youth and morale.

Danto’s End of Art

131

practice of art making since ancient times. New in the twentieth century, however, is that ordinary objects are appropriated to be works of art, and therefore only in twentieth-century philosophy of art does the question emerge as to why of two perceptually indiscernible objects one is a work of art while the other is not. Danto’s problem is, in other words, historically indexed, for it stems from particular developments in art. If the problem were raised before the early twentieth century, before, let us say, Marcel Duchamp’s In Advance of the Broken Arm (1915), it would not have made any sense, for there simply were no indiscernible objects one of which was art and the other was not. One may at times have mistaken an ordinary object for a work of art and vice versa, but there was no philosophical problem before Duchamp ‘made’ his appropriated sculpture, indiscernible from a snow shovel, which, apart from being a work of art, it also happens to be. The appropriation of consumer goods is, apparently, an artistically relevant predicate, for it allows us, at least in these instances, to distinguish In Advance of the Broken Arm from a snow shovel and Made in China from a lawn chair, even though they are a snow shovel and a lawn chair, respectively, and there are presumably more important artistically relevant predicates that allow us to identify these objects as works of art. If at some moment in time it is decided that p is artistically relevant for some work, p becomes relevant for all works – some or most of which will lack p. Such retroactive enrichment of the entities in the artworld makes it possible to discuss different artists together, even if some artist could have no knowledge of the future artists and the predicates they would introduce with which he would be retroactively related. This is Danto’s theory. He writes: ‘The greater the variety of artistically relevant predicates, the more complex the individual members of the artworld become; and the more one knows of the entire population of the artworld, the richer one’s experience with any of its members.’10 His theory of retroactive enrichment of the artworld with artistically relevant predicates emphasizes the historical dimension of the artworld, which becomes richer over time and allows re-descriptions of earlier works based on new ones. Incidentally, here one sees how Danto’s analytical philosophy of history, presented in his Analytical Philosophy of History and in which sentences that describe an earlier event in terms of a later event are central, importantly influences his views on art in his 1964 ‘The Artworld’ essay.11 10 Danto, ‘The Artworld,’ 583-584. 11 These sentences he calls ‘narrative sentences’. See Danto, ‘Narrative Sentences,’ History and Theory, 2(2) (1962), 146-179, and Analytical Philosophy of History (Cambridge: Cambridge University Press, 1965). I discussed these sentences in Chapter 5.

132 

The Exemplifying Past

Danto’s theory of the retroactive enrichment of the artworld also provides an answer to the question why of two perceptually indiscernible objects one is a work of art and the other is not, in that at a particular moment in time some artwork introduces an artistically relevant predicate into the artworld hitherto unknown to it which allows distinguishing between this artwork and its ordinary counterpart. Given our example, this predicate is ‘being perceptually indistinguishable from real things,’ which thus not only enables us to count in In Advance of the Broken Arm as one of the artworld’s members, but at the same time introduced the philosophical problem central to Danto’s philosophy of art. To be sure, not all philosophy is historically indexed in this way, for not all philosophical problems arise from actual developments.

3

Art and Interpretation

Danto’s theory of the retroactive enrichment of the artworld with artistically relevant predicates is not fully satisfactory in that it leaves the question open whether there are necessary and suff icient conditions for something to be a work of art. Such conditions would enable us to distinguish once and for all between two perceptually indiscernible objects one of which is a work of art, independent of our knowledge of the history of art and the artistically relevant predicates that are introduced at specific times. To these conditions Danto turned in his The Transfiguration of the Commonplace, and later he realized that they were already formulated by Hegel in his 1820s lectures on fine art.12 The conditions for something to be a work of art are: (i) works of art are about something, they have a certain content or meaning; and (ii) works of art embody the meaning they express, i.e. their means of presentation is appropriate for the meaning they express.13 These two conditions mark the difference between ordinary 12 For this realization, see Danto, ‘The End of Art: A Philosophical Defense,’ in David Carrier ed., History and Theory. Theme Issue 37. Danto and his Critics: Art History, Historiography and After the End of Art (1998), 127-143, at 130. 13 Danto, After the End of Art, 194-195. A central claim of Danto’s Transfiguration of the Commonplace is that art is expression. On the conditions, see also Danto, ‘Art and Meaning,’ in Noël Carroll ed., Theories of Art Today (Madison and London: University of Wisconsin Press, 2000), 130-140, at 132. In this essay, Danto addresses several objections to the two conditions. Hegel writes in his Aesthetics. Lectures on Fine Art, trans. T.M. Knox, Volume I (Oxford: Clarendon Press, 1975), 11: ‘What is now aroused in us by works of art is not just immediate enjoyment but our judgment also, since we subject to our intellectual consideration (i) the content of art, and (ii) the work of art’s means of presentation, and the appropriateness of both to one another.’

Danto’s End of Art

133

objects and artworks.14 Let us take Baas’s Made in China as an example. The plastic chair in wood is a work of art because, on the one hand, it is about ‘mass production and the making of products as cheap as possible.’ The title of the work, Made in China, already points in this direction. ‘On the other hand, China is known for its ref ined handwork in china and painting.’15 These extremes of production in China are what Baas’s work is all about. His work is a comment on consumer society in a globalized world. The wooden lawn chair suits this purpose very well: the work embodies its meaning. It is crucial to note that Baas’s chair asks for an interpretation. Without such interpretation, the chair would simply be a chair, an object on which to sit. All artworks require interpretation according to Danto, otherwise they would not express anything about their subject matter. The interpretation thus constitutes the object as a work of art in that only in relationship to an interpretation the material object is a work of art. Not all interpretations are equally satisfying. Danto’s view is that the correct interpretation coincides with the interpretation that the artist of the work had in mind when creating the work.16 He could have been a bit more precise, and state, first, that an interpretation of a work of art is correct if the meaning expressed by the work is embodied by it, i.e. if the means of presentation is appropriate relative to the work’s content and vice versa, and second, that such embodiment is what the artist aims at when making his work. A different, although related, interpretation of a work of art concerns its possible art historical significance. This we turn to below. The distinction between the artistic and the historical interpretation of a work of art is important because the end-of-art thesis concerns the second type of interpretation in that after the end of art, art is still being made and still Hegel further elaborates on his definition in several sections, see in particular 70-73 and 95. I identify what Hegel refers to as the appropriateness of the means of presentation with what Danto refers to as embodying. 14 Noël Carroll objects that the two conditions may also apply to ordinary objects and argues that a sports car is about ‘speed’ and embodies that, and a sword may be about ‘being feared’ and embody that. See Carroll, ‘Danto’s New Definition of Art and the Problem of Art Theories,’ British Journal of Aesthetics, 37(4) (1997), 386-392, at 387. However, if the sports car is about ‘speed’ and embodies that, then it is an artwork and not an ordinary object. If the conditions apply, we are dealing with a work of art. In other words, the conditions are necessary and jointly sufficient. 15 These are Baas’s own words in an interview by Homme Siebenga in Museum Tijdschrift, 2 (2010), 24. 16 Danto, ‘The Appreciation and Interpretation of Works of Art,’ in The Philosophical Disenfranchisement of Art (New York: Columbia University Press, 2005), 23-46, at 44-45.

134 

The Exemplifying Past

needs to be interpreted, but, as we will see, art no longer carries any historical significance, and therefore the (historical) interpretation of the historical significance of art is ruled out. Both Duchamp and Baas could not have made their works in any other time than they did. Not only because of the practical reason that there had to be mass-produced snow shovels and plastic lawn chairs for sale before they could be appropriated to be works of art, but also because those works are to be interpreted in such a way that they embody the meaning they express. If someone had made Made in China two hundred years ago, it would not have been a work of art but a rather poor looking chair, and we might even agree with that judgment, for there are better looking chairs in the world than this plastic chair in wood. We found that interpretation constitutes works as works of art in that without an interpretation objects cannot be works of art. Now we find that interpretation is historically indexed. Interpretation depends, in part, on the available or newly introduced artistically relevant predicates. It also depends on the world in relation to which the artwork is interpreted – a world with garden lawn chairs, consumer society, china, bulk production in China, and so on. Only in relation to this world can Made in China mean what it does. This is the historical dimension of art. The two conditions for something to be a work of art do not contradict the view that interpretation is historically indexed. If the conditions are satisfied, then the art object is interpreted as a work, and correctly inasmuch as the interpretation coincides with the interpretation that the artist had in mind. If the conditions are not satisfied because we do not know how to interpret the object or because we do not know that the object should be interpreted in the first place, then the object is not a work, and in the case of Baas’s Made in China, the object would be a somewhat peculiar looking lawn chair. If the conditions are satisfied but the artwork is not interpreted in accordance with how the artist would have interpreted it – and this is Danto’s criterion of correct interpretation – then the object has turned into a new work, meaning something different than envisioned by the artist. Baas’s 2008 Made in China was made after art ended in 1964. One would expect this to enter its interpretation. The historical dimension of art in terms of the techniques of producing art and interpreting art is, as far as I know, non-controversial. Art is made and interpreted in specific times and places using specific technologies and symbols.17 Danto’s thesis that art has developed in a certain direction 17 Art is, in other words, a cultural practice. Noël Carroll, ‘Art, Practice, and Narrative,’ The Monist, 71(2) (1988), 140-156.

Danto’s End of Art

135

and has ran its course, on the other hand, is controversial and presumes a substantive philosophy of history which the historical dimension of art in said terms does not. This I turn to now.

4

The End of Art

Danto claims that art ‘ends with the advent of its philosophy,’ i.e. the ‘historical stage of art is done with when it is known what art is and means.’ This happened in modernist art – the period stretching from impressionism to pop art. Modernist artists started to question what art is with their works: as art, painting and sculpture became objects for themselves, and that is how art found out what art is and means.18 Duchamp’s In Advance of the Broken Arm is a case in point. The conditions for something to be a work of art help us understand why. The work questions what art is by questioning the art historical concept of art of his time. It is about how ordinary objects can be works of art, in contrast to what art history, up till then, was teaching. Duchamp’s appropriation therefore is part of the meaning of the work. Art history, the work tells us, was taking itself far too seriously. If you want to comment on art history and its dominant concept of art, then the introduction of a snow shovel in a museum is a good way to do so. The snow shovel then embodies its meaning. In Advance of the Broken Arm is a work of art for the conditions for something to be a work of art are satisfied in the artistic interpretation given. Another clear example of a work of art questioning what art is, is Malevich’s Black Square, also from 1915. This work tells us that the religious experience is not dependent on Christian iconography but on art itself – the painting was to be hanged at the same place where iconographic paintings in houses were hanged. This is what the painting is about and how it embodies its meaning. Duchamp’s sculpture is an example of new realism: nothing is as real as reality itself, and it would be impossible to find a more realistic visualization of a snow shovel than his work. Malevich’s painting is, by contrast, an example of abstraction. What connects these works is that they both question what art is and the idea of art that the history of art tells is the correct one. The whole concept of mimesis is put aside by them and deemed irrelevant. In the analysis of Karsten Harries, who in retrospect anticipated 18 Danto, ‘The End of Art’ (1984), reprinted in his Philosophical Disenfranchisement, 81-115, quoted phrases at 107 and 111 respectively. Already in his Transfiguration of the Commonplace, vii and 56, Danto posited the thesis that art had ended.

136 

The Exemplifying Past

Danto’s end-of-art thesis, there had always been a certain balance between mimesis and abstraction in the visual arts until 1915: the moment Duchamp made his work, which is the opposite of abstraction, and Malevich made his work, which is the opposite of mimesis. The balance between mimesis and abstraction was disturbed for good and there was no coming back of it. Abstraction and mimesis would go their own and separate ways.19 Duchamp and Malevich shook up the artistically relevant predicates of the artworld, and the viewers of the works not only had to ponder what their works meant, but also were forced to ponder art and its history in general, for that was part of the meaning of those works. Danto’s astonishing thesis is that by questioning what art is, the conditions for something to be a work of art were brought to consciousness in the domain of art itself, and this happened in modernist art. Art ended the moment it was known what art is, and when that moment arrived, art, as a historical phenomenon, was over. Danto is very precise about this end given his favourite example of this end. Art ended with Warhol’s exhibition of his Brillo Box at the Stable Gallery on East 74th Street in Manhattan in April in 1964.20 Like Duchamp’s In Advance of the Broken Arm and Malevich’s Black Square, Warhol’s Brillo Box is about art, but his work says that anything can be a work of art: that is what his work means. Brillo Box, perceptually indiscernible from its ordinary counterpart, although the former is painted and made of wood whereas the latter is made of printed paperboard, suits the meaning it expresses very well: it embodies its meaning, and it can only do that for the f irst time at that particular moment in time. In 1964, so Danto tells us, it became clear that there was no longer any way that art was supposed to be, and no artwork was truer than any other. Now that anything could be a work of art, future developments were ruled out. There is art after the end of the art, but, Danto claims, ‘its existence carries no historical signif icance’. There is, in other words, no historical interpretation of art after the end of art possible. In 1964 a point was reached ‘where there can be change without development, where the engines of artistic production can only combine and recombine known forms.’21

19 Karsten Harries, ‘Hegel on the Future of Art,’ Review of Metaphysics, 27 (1974), 677-696, at 690-691. 20 Danto, Transfiguration of the Commonplace, vii and 208; and Danto, After the End of Art, 35-36. 21 Danto, ‘The End of Art,’ 84 and 85 respectively.

Danto’s End of Art

137

Again Duchamp’s In Advance of the Broken Arm is a case in point. The work could not have been made after the end of art, since after the end of art, the question what art is has been answered. To be sure, the art object could still be made (and bought), but it could never mean what it did in 1915 as a work. One difference between the sculptures of Duchamp and Baas which we did not mention earlier is instructive here. Duchamp appropriated an existing object without physically altering it, whereas Baas appropriated an existing design by having it manually reproduced. Duchamp made his work before the end of art, whereas Baas made his work after the end of art – on the other side of history so to speak. After the end of art it no longer makes sense to appropriate an object from a warehouse and turn it into a work of art in the manner in which Duchamp did, for the question what art is has been answered. One can, however, appropriate a design, as Baas did, but for the very different reason of expressing something about our present day consumer society. It can be argued that Danto’s thesis that the conditions for something to be art were brought to consciousness in the domain of art entails a substantive philosophy of history in that it explains why art has developed as it did and why it has ran its course. However, it can also be argued that Danto’s thesis is a historical thesis in the sense in which all historical narratives present a thesis on the past: as a means to provide narrative coherence, Zusammenhang, to the manifold of events. His thesis gives coherence to what we refer to as ‘modernist art’ and allows us to discuss a wide range of artists and their works in relation to one another. The end of art is in other words an artistically relevant predicate which retroactively enriches the entities in the artworld. If, however, the end of art thesis presumes a substantive philosophy of history, as I think it does, then the end of art does much more than provide coherence to the manifold of events that we associate with modernist art, for in that case the thesis also explains how and why modernist art was destined to come into existence and end. I will now discuss the question how and in what sense the end-of-art thesis presumes a substantive philosophy of history. This enables me to determine how we are to assess the credibility of the end-of-art thesis, and with that, the credibility of the philosophy of history that supports it. But first there is one final hurdle to overcome.

5

Analytical and Substantive Philosophy of History

How can it be, one wonders, that Danto, a well-known philosopher of history, does not scrutinize the substantive philosophy of history that he

138 

The Exemplifying Past

presupposes with his end-of-art thesis?22 One reason may be this. Substantive philosophy of history is not proper philosophy since it is concerned with the past itself. It is, in other words, a form of history writing rather than a form of philosophy. Because Danto meant his end-of-art thesis to be a contribution to the philosophy of art rather than as a contribution to art history, he had no reason to scrutinize the substantive philosophy of history presupposed by his end-of-art thesis. He knew, of course, that his end-of-art thesis was a thesis about art and its development in the past, but if we want to assess the philosophical credibility of his thesis as a thesis on the history of art, we should know what criteria there are for the assessment of such theses, and this implies that we turn to analytical philosophy of history rather than to substantive philosophy of history, for analytical philosophy of history is, in contrast to substantive philosophy of history, philosophy proper and concerned with the nature of historical knowledge. Any historical assessment of the end-of-art thesis, as to it its historical correctness, accuracy, plausibility, consistency, and so on, is up to art historians. These are, to be sure, mere speculations about the reasons Danto might have had. What should however be clear is that in this view, the end-of-art thesis presents us with knowledge about art, its development and coming to an end in the past, and the substantive philosophy of history it presupposes is left unanalyzed. What may have made Danto reluctant to scrutinize the substantive philosophy of history that he appears to accept is that he criticized the substantive philosophies of history in his Analytical Philosophy of History. This is what Danto has to say about the issue: Art history must have an internal structure and even a kind of necessity. This was the conviction that motivated my essay ‘The End of Art,’ and the other writings which undertake to articulate a philosophy of the history of art in exactly the grand manner I had learned from Hegel, and which it astonished me that I was accepting, since my first book, the Analytical Philosophy of History of 1965, had pretty much taken a stand against its possibility in principle. Right or wrong, my view now was that the artworld was not demanding only a philosophy of art. It was demanding a philosophy of its own history.23 22 Danto is well known as a philosopher of history for his Analytical Philosophy of History. The book was republished twenty years later in 1985 with three additional chapters as Narration and Knowledge (New York: Columbia University Press, 1985), a year after Danto presented his end-of-art thesis in his ‘The End of Art’. The book starts with a chapter criticizing substantive philosophies of histories in the manner detailed in this section. 23 Danto, ‘Preface,’ Philosophical Disenfranchisement, xxviii.

Danto’s End of Art

139

The philosophy of history that motivated Danto’s ‘End of Art’ essay is, as we concluded in the previous section, still in need of clarification. This I aim to do in the next section. Here our focus is on Danto’s stance against the very possibility of a substantive philosophy of history. One criticism of substantive philosophies we already discussed: substantive philosophies of history are histories rather than philosophies. A second criticism is that they are mistaken about the nature of historical knowledge. Substantive philosophies of history, Danto argued, presume to give an account of the whole of history. They are not only concerned with the past, as all histories are, but also with the ‘historical future’, i.e. with events and their historical meaning as they will unfold in the future. Substantive philosophies of history therefore are prophetic. This betrays a deeply mistaken conception of history. One cannot write the history of events before they have happened and determine their historical meaning in advance, for, as Danto argues – and this is the central claim of his Analytical Philosophy of History – the historical meaning of future events cannot be known before those events have come to past and can be understood in relation to later events which give earlier events their historical meaning. Historical knowledge is, essentially, retrospective, as it provides descriptions of earlier events in terms of later events, both of which are known to the historian. We already saw that this central claim concerning historical knowledge importantly influenced his ‘The Artworld’ essay.24 Two conclusions follow from this. First, Danto’s criticism of substantive philosophies of history, which we may accept, is not concerned with the direction of a particular course of events and how that course may come to an end as long as that end lies in the past. His end-of-art thesis and the substantive philosophy of history it presupposes therefore do not contradict his earlier criticism of substantive philosophies of history. It should also be noted that Danto never renounced his criticism of substantive philosophies of history, and that his end-of-art thesis does not presume that the historical process has a purpose. Secondly, because the end-of-art thesis is a thesis about art and its history, it is, as a thesis, a form of knowledge providing narrative coherence to modernist art, and therefore, the thesis might well be 24 In his Analytical Philosophy of History, Danto also attempted to bring narratives under the covering-law model of explanation. This attempt was in 1965 already at the brink of becoming outdated. In retrospect, it was his discussion of re-description of earlier events in terms of later events that helped bring about the so called narrative turn in the philosophy of history that turned out to be historically significant. See Michael Beaney, ‘Historiography, Philosophy of History and the Historical Turn in Analytical Philosophy,’ Journal of the Philosophy of History, 10(2) (2016), 211-234, at 228-229.

140 

The Exemplifying Past

in agreement with Danto’s analytical philosophy of history. Such agreement is not ruled out given that Danto’s criticism of substantive philosophies of histories does not apply to the substantive philosophy of history presupposed by his end-of-art thesis. These two conclusions assume that the distinction between analytical and substantive philosophy of history is helpful with regard to the issue sub judice here. However, the distinction is helpful only up to the point we have just reached: the philosophical domain of the nature of historical knowledge is to be distinguished from the historical domain of the past and there is no such a thing as the historical meaning of future events. The conclusions do not bring us any further than were we left off at the end of the previous section. The end-of-art thesis is, apart from being both a contribution to the philosophy of art and the history of art, also a philosophy of the history of art, and we want to know what this philosophy is. The central issue thus remains: Danto did not scrutinize the substantive philosophy of history that his end-of-art thesis presumes beyond what he says about it in his ‘The End of Art’ essay and some scattered references to the work of Hegel. Explicating this philosophy of history, then, is the task I set myself in the next section. It will further support the central claims of this book.

6

The Common and the Philosophical View of History

My argument is that what I refer to as the standard or common view of history cannot account for Danto’s end-of-art thesis and the philosophy of history which supports it. I take it that the common view of history consists of two levels and the question how they are related. On the one hand there is 1) the level of the past and its remains. This is the level of past reality, the level of the actual course of events and its remains which are still present in our day. On this level we find the artists and their works, their attitudes, desires, and beliefs, the circumstances in which they worked and lived, and all that affected them during their careers. In the common or standard view of history, this is what the past consists of and what it is that historians try to understand, explain, and represent. On the other hand there is 2) the level of our knowledge of the past and its remains, which includes the methods, techniques and approaches used to attain this knowledge and the way this knowledge is structured and presented. The common view of history is well equipped to account for the historical dimension of art as we discussed it. It locates the practices and technologies of art production, its appreciation, the associated attitudes, beliefs and desires, and the context

Danto’s End of Art

141

in which this all happened, in the past; and we acquire knowledge of the past by means of the appropriate methods of studying its remains.25 When we say that something is historically indexed, it is the common view of history that we have in mind. This two-level view of history is, as far as I know, uncontroversial and widely used0. Also part of the common view of history is the question how the past is related to our knowledge of the past, as for example in the questions how to arrive at a truthful and sincere account of the past,26 and if and how the coherence of the narrative reflects the coherence of the past. These are just two examples of the many questions that the common view of history gives rise to. Danto’s Analytical Philosophy of History provides other questions following from the common view of history. There is much to be said about how 1) the level of the actual course of past events and 2) our knowledge of it are related, and the sometimes heated debates about that relation, but that is not my concern here. I merely want to establish that there is such a thing as the ‘common view of history’, with the purpose of bringing to the fore that this view does not make room for the end-of-art thesis and its supporting philosophy of history, and that is why we should not asses the credibility of Danto’s end-of-art thesis in terms of the common view of history. Perhaps the reader is surprised by this strategy and objects that the common view of history does make room for Danto’s end-of-art thesis and its associated philosophy of history, for surely, so the objection goes, the philosophy of history that Danto tacitly adheres to should be located on 1) the level of the past and the end-of-art thesis as something he proposed and thought of should be located on 2) the level of our knowledge of the past. After all, Danto’s thesis is that the conditions for something to be work of art were brought to consciousness in the domain of art itself, and clearly, this happened in the past itself. This Danto, nor anyone else, can deny. To name just one obvious example. How else could he be so precise about the moment when art fully knew what it was? As we saw, according to Danto, art ended in 1964 in the Stable Gallery in New York with Warhol’s Brillo Box. No reasonable person can deny that Brillo Box was exhibited at that time and place in the past and that Danto was one of its witnesses. The way art developed, the direction it took, and the moment it ended, are all to be 25 Carroll’s ‘historical or narrative approach to the cultural practice of art’ as presented in his ‘Art, Practice, and Narrative,’ is a good example of thinking about art and history using the common view of history. 26 This is one of the questions Bernard Williams raises in his Truth and Truthfulness (Princeton: Princeton University Press, 2002).

142 

The Exemplifying Past

located in the past, so the objector argues. Therefore, the common view of history provides all the room that the philosophy of history supporting the end-of-art thesis needs. The thesis as expressed in Danto’s ‘End of Art’ and other essays is a form of knowledge and therefore it should be located at the second level of our knowledge of the past. It follows, so the objector continues, that any assessment of the credibility of Danto’s thesis depends on the common view of history and the way we (ought to) think about the relation between 1) the past and its remains and 2) our knowledge of it and the way this knowledge is attained and presented. There are criteria associated with the common view of history to evaluate the credibility of our knowledge of the past, and these criteria are to be used if we want to assess the credibility of Danto’s end-of-art thesis. This sensible objector is, however, mistaken. First we should observe that assessing the credibility of the end-of-art thesis to the mind of the objector depends either on i) a relation of confirmation and disconfirmation between the levels 1) and 2): our 2) knowledge of the past is confirmed or disconfirmed by the 1) evidence we find on the level of the past and its remains; 27 or the credibility of the thesis depends on ii) criteria with which we evaluate our knowledge of the past and the form it has, such as consistency, scope, fruitfulness, accuracy, adequacy, and coherence, or any other such criteria with which we evaluate theses on the past.28 But both do not apply to Danto’s thesis, or at least not in the way that our objector assumes they do, and that is why he is mistaken. Against i) I will hold that there is no evidence for Danto’s thesis whatsoever. Therefore his end-of-art thesis cannot be confirmed or disconfirmed, and that is why the common view of history cannot account for its credibility. His thesis is, however, supported by a philosophy of history. With regard to ii), I admit that we may evaluate Danto’s thesis in terms of the criteria mentioned,29 but that is not all that there is to it, for in Danto’s view, art did become an object for itself in modernist art and that is why art 27 In Chapter 3 I argued that confirmation is an inferential rather than a representational practice. For the argument here it is immaterial whether one agrees with that particular argument or not. It suffices to acknowledge that the idea of confirmation and disconfirmation is concerned with the relation between the levels 1) and 2). How that relation is to be understood is a different matter. 28 On such criteria, see e.g. Jouni-Matti Kuukkanen, Postnarrativist Philosophy of Historiography (Basingstoke: Palgrave Macmillan, 2015), 155-158. 29 Inasmuch as those criteria suggest a relation between the thesis and the past which it is about, we are back at objection i). Criteria such as adequacy and accuracy may suggest a relation of confirmation and disconfirmation between the levels of 1) the past and its remains and 2) our knowledge of it. These criteria may however also stand for norms of scholarly behaviour and therefore only apply to the level of 2) our knowledge of the past.

Danto’s End of Art

143

reflected on itself, gained self-knowledge, and ended as a result of that, and this did happen in the past, but not in the sense allowed by the common view of history. Against ii) I hold that we only are able to evaluate the end-of-art thesis if we take the philosophy of history which it supports seriously. My task now is to substantiate the claim that Danto’s thesis cannot be confirmed or disconfirmed and to take its supporting philosophy of history seriously. I will do so as part of presenting what I refer to as the non-standard or philosophical view of history, which does not so much as contradict the common view of history, but adds two more levels to it. The argument is that although the end-of-art thesis cannot be confirmed or disconfirmed – there is no evidence for or against it –, the thesis is supported by a philosophy of history. To bring this to the fore we need in addition to the levels of 1) the past and its remains and 2) our knowledge of it and the way we present that knowledge, two more levels. Our first extra level is the level of necessary conditions of what actually exists. This is level 0) and here we should locate the conditions for something to be a work of art. These conditions manifest themselves on the level of 1) the past and its remains. Level 1) is thus according to the philosophical view of history a manifestation or realization of level 0). Put differently, level 0) becomes concrete on level 1), in the actual course of past events. For a proper understanding of the philosophical view of history, we have to realize that we only become aware of 0) the necessary conditions that come manifest 1) in the actual course of past events if we take another and fourth level into account. This fourth level is level 3), the level of exposition, and this is the appropriate level of the end-of-art thesis. This four-level model and their relations constitute what I call the philosophical view of history. It allows us to conceive of Danto’s end-of-art thesis as follows. His 3) thesis that the conditions for something to be art were brought to consciousness in the domain of art itself makes us aware of how 0) the necessary conditions of art became manifest in 1) the actual works of art that we associate with 2) modernist art. Put differently, the 3) end-of-art thesis is only properly understood on the basis of 1) artworks illustrating or exemplifying that 3) thesis. Duchamp’s In Advance of the Broken Arm and Malevich’s Black Square are not evidence of modernist art and the conditions for something to be art to come to consciousness in the domain of art itself, rather these works illustrate that thesis. The relation between the 1) past and its remains and the 3) end-of-art thesis is therefore not a relation of confirmation and disconfirmation.30 30 The philosophical view of history accepts that the relation between levels 1) and 2) is a relation of confirmation and disconfirmation, as the common view of history holds.

144 

The Exemplifying Past

There is no evidence to be found on the level 1) of the past and its remains that either confirms or disconfirms the 3) thesis. The philosophical view of history holds that the credibility of 3) the thesis depends on how well it is able to make us aware of the 0) necessary conditions of what exists and how they become manifest in 1) the actual course of events and all we 2) know about it. The philosophical view of history does not claim that it is able to predict the 1) actual course of events. It does claim that the 1) actual course of past events is only determinate in retrospect inasmuch as the 3) thesis of the past exposes the 0) necessary conditions of 1) what actually have been found to have existed.31 Now we may understand why Brillo Box ended art. From the point of view of the common view of history, it is rather curious that art ends with this specific work at the specific time and place of its exhibition in the Stable Gallery in 1964. Not because it seems implausible that it was Warhol’s intention to end art, for the common view of history can account for all of Warhol’s intentions, even the eccentric ones, and determine whether there is evidence for the intention or not. But why of all the artworks made is this specific work at this specific moment the work to end all art? Why was not, for instance, Richard Hamilton’s collage Just What Is It That Makes Today’s Houses So Different, So Appealing? at the 1956 This is Tomorrow exhibition the work that ended all art, as the German art historian Hans Belting argued? According to him, the This is Tomorrow exhibition was a ‘farewell to art history,’ for ‘It was no longer the intention to guide art history into a yet unknown future but to give up the notion that art history still was to go on altogether.’32 We may doubt whether this was indeed the intention of the exhibition, but here the point is that the common view of history would have to say that art either ended twice (in 1956 or 1964) and Danto and Belting are both right, or art ended once and one of them is mistaken. (Another option in the common view of history is that art did not end and therefore both Danto and Belting are wrong. Here the end of art is interpreted as meaning that there is no art after the end of art, but that is not what Danto claims. Needless to say, there is an abundance of evidence of art after the end of art. Yet another option is that Danto’s and Belting’s theses are different and only share the same name, but this option would leave the issue at hand intact: why does Brillo Box end all art?) The philosophical view of history has a better way to deal with this. Since both Brillo Box and Just What Is It 31 I discussed the indeterminate past being determinate in retrospect in Chapter 3. 32 Hans Belting, Art History after Modernism (Chicago and London: University of Chicago Press, 2003), 75. This is a revised version and translation of his 1983 book.

Danto’s End of Art

145

That Makes Today’s Houses So Different, So Appealing? are no evidence of the end of art theses, it is not that art ended twice or that Danto or Belting is mistaken. Brillo Box exemplifies the end of art, and it does so inasmuch as it allows us to see in it how art ends. The same holds for Hamilton’s collage, and it may well be that only after reading Danto’s ‘End of Art’ essay one sees how this collage ends art. In that case, Danto’s end-of-art thesis would have become concrete in that work. When Danto visited the Stable Gallery in 1964, the artwork prompted him to reconsider all he knew about art. Rather than ‘merely’ describing and re-describing it as adequately as possible in the historical context of its existence, providing an adequate and accurate account of the past in agreement with the common view of history, Brillo Box prompted Danto to ponder how art and its development could have culminated in Brillo Box, and this eventually led to his end-of-art thesis. He wondered, we might say, what state of mind art was in, and how that state of mind came into existence, given the actual course of events. This is the question that requires the philosophical view of history if it is to be answered. In the introduction to this book I wrote that its aim was to make sense of the claim that the spirit of an age retroactively becomes concrete in the artifacts under consideration. This claim, we now know, makes sense if we have a philosophical view of history. Brillo Box exemplifies the spirit of its age, and it does so inasmuch as Danto has us see in it how art came to an end.

7

After the End of Art

Modernist art was a thing of art history. Not because modernist art referred to earlier works of art, that would have been nothing new, for such reference can be found in works of art throughout time, but because art from impressionism to pop art comments on the history of art and the concept of art presented by that history. Modernist art questioned what art according to its own standards and history was and should be, and it is precisely by virtue of commenting that modernist art reflected on itself and became an object for itself and thus gained self-knowledge. Art ended the moment it was realized that as far as appearances go, anything can be a work of art. A good example of such self-knowledge is Rob Scholte’s 1986 painting Utopia. The painting’s composition is an appropriation of a postcard which Scholte found in London, probably at the Cabaret Mechanical Theatre, depicting the mechanical sculpture Manet’s Olympia by automata-maker Paul Spooner, who used a drawing doll model in his automata of Edouard

146 

The Exemplifying Past

Manet’s famous painting. Scholte painted the drawing doll model lying on a sofa, with a black cat beside it and a black servant like in Manet’s painting, but this time in wood bringing her something to drink. His painting refers to Manet’s Olympia, which is often thought to be the first modernist painting, via the postcard of Spooner’s mechanical sculpture, and in virtue of that it refers to the Venus of Urbino of Titian to which Olympia refers, and these references are part of the meaning of Scholte’s painting, but not in the sense in which in modernist art the history of art and its concept of art was commented upon. The reference to Manet in Utopia is not a rehearsal of motif or theme, which is a common artistic practice which can be found throughout time, nor does the painting question the concept of art of its time, which is typical of modernist art, for Scholte knows all too well that anything can be a work of art. The latter is precisely what Utopia is about and the painting tells us that if anything can be a work of art, then so can art. This is, I think, what Scholte’s painting means, and why the work is such a fine example of art’s self-knowledge. Earlier I made a distinction between the artistic and historical interpretation of art. The artistic interpretation I gave of Scholte’s Utopia is concerned with what the painting is about and what meaning it embodies. These are the conditions for something to be a work of art. Interpretation, Danto argued, constitutes a work as a work of art in that all art requires interpretation. Any artistic interpretation is historically indexed in that it depends on the available artistically relevant predicates and the world in relation to which the artwork is interpreted. The art historical interpretation of a work of art, on the other hand, concerns its historical significance, i.e. its contribution to some historical development, as for instance in the contribution that Manet’s Olympia made to modernism in art. In the words of the art historian Timothy Clark, the ‘peculiar freedom with the usual forms of representation was later held to be the essence of Olympia […] and made it the founding monument of modern art.’33 Put differently, mimesis was destined to become an artistically irrelevant predicate, which, retroactively, would become irrelevant for all art works. Not only the artistic interpretation is historically indexed in the sense just described, the historical interpretation is too and depends not only on artistically relevant predicates but also on what we might call ‘historically relevant 33 T.J. Clark, The Painting of Modern Life. Paris in the Art of Manet and his Followers (Princeton: Princeton University Press, 1984), 79. Clark is not only interested in the art historical interpretation of the work, but also in its social historical interpretation and how the work relates to the nineteenth-century social question.

Danto’s End of Art

147

predicates’ and the world in relation to which the artwork is interpreted, such as ‘Renaissance,’ ‘pre-Raphaelite,’ ‘impressionist,’ and ‘modernist.’ There is, however, no historical interpretation of Scholte’s Utopia possible, for the work was made after the end of art. Art ended in 1964, and even though there was and is a future for art in that art will keep on being made and appreciated, the historical future of art to come was closed the moment art ended. There were no longer any contributions to the history of art to be made. The common view of history can perfectly account for the fact that the artistic and historical interpretations of art are historically indexed. If, on the other hand, we want to grasp why art developed in the direction in which it developed, and why art after the end of art is no longer historically significant, we need the philosophical view of history.

8 Conclusion In his ‘The End of Art’ essay Danto quotes Hegel’s famous remark from his lectures on the fine arts that ‘Art invites us to intellectual consideration, and that not for the purpose of creating art again but for knowing philosophically what art is.’34 This is what Danto, prompted by Warhol’s Brillo Box, aimed to do, first in his ‘Artworld’ essay, then in his The Transfiguration of the Commonplace, and finally in his ‘End of Art’ essay. He not only defined the necessary and sufficient conditions for something to be a work of art, which, as we saw, were already formulated by Hegel; he made clear that these conditions were brought to consciousness in the domain of art itself.35 Hegel did not anticipate that and how art would become an object for itself, and he only would have been confused by and unable to interpret the works we discussed in this chapter. Hegel also did not claim to know the actual course of art and history beforehand, nor did Danto. The actual course of events and its direction are only determinate in retrospect. Danto’s criticism 34 Hegel, Aesthetics, 11. Quoted in Danto, ‘The End of Art,’ 114. 35 Art being an object for itself is central to a Hegelian understanding of the development of art. In this context, Robert Pippin arrives at the same conclusion as Danto in his ‘What Was Abstract Art? (From the Point of View of Hegel),’ Critical Inquiry (2002), 1-24, as he (p.17) writes: ‘Hegel’s narrative of an expanding critical self-consciousness thus fits the modernist refusal to take for granted what a painting or art was, what writing or being an artist was. Such notions were now treated as norms, neither fixed by nature nor [by] human nature, but actively (and in Hegel historically) legislated and subject to criticism. With such questions raised this way, it would come as no surprise that art making and novel writing would themselves become the subjects of art; Proust and James, de Kooning and Pollack are only the most obvious examples.’

148 

The Exemplifying Past

of substantive philosophies of history, which, in his view, purport to write the history of events before they have happened, thus has no bearing on his end-of-art thesis. Nevertheless, his end-of-art thesis presumes a philosophy of the history of art. Although Hegel and Danto did not claim to know the actual course of events, they did claim to have identified the general state of mind that art was in their day and how that state of mind had come into existence.36 Such identification, we saw, presumes a philosophy of history which can only be accounted for by the philosophical view of history.37 I find support for this view in William Dray, who, by sheer coincidence, in 1964, the year in which Warhol exhibited his Brillo Box, contended that ‘the Hegelian philosopher cannot say anything about history’s actual course unless he knows first what stage of spirituality has been reached.’ And this does not ‘entail the predictability of history. What it entails is the retrodictability of the necessary conditions of what has actually been found to exist.’38 In this sense, Danto is a Hegelian philosopher. Warhol’s Brillo Box prompted Danto to determine the ‘stage of spirituality’ that art had reached in his days, and that, eventually, resulted in his end-of-art thesis. Hegel would add that here philosophy is as it should be: its own time expressed in thought.39

36 Danto’s thesis is very different from Hegel’s. Hegel claims that ‘art, considered in its highest vocation, is and remains for us a thing of the past [ein Vergangenes],’ for art no longer satisfies the spiritual needs of man. ‘We have got beyond venerating works of art as divine and worshipping them. […] Thought and reflection have spread their wings above fine art.’ Hegel, Aesthetics, 10-11. Later Hegel writes: ‘[N]o matter how we see […] Mary so estimably and perfectly portrayed: it is no help; we bow the knee no longer’ (103). Cf. the many different interpretations of Hegel’s thesis, among which Danto’s, by Martin Donougho in his ‘Art and History: Hegel on the End, the Beginning, and the Future of Art,’ in Stephan Houlgate ed., Hegel and the Arts (Evanston: Northwestern University Press, 2007), 179-215. 37 Not allowing for the philosophical view of history as distinct from the common view of history explains a certain shortcoming in the interesting essays that deal with Danto’s end of art thesis in volumes such as the one edited by Mark Rollins, Danto and his Critics. Second Edition (Malden, Oxford, and Chichester: Blackwell Publishing, 2012), and the aforementioned theme issue of History and Theory edited by David Carrier. 38 Dray, Philosophy of History (Englewood Cliffs: Prentice-Hall, 1964), 76. 39 Hegel, Grundlinien der Philosophie des Rechts oder Naturrecht und Staatswissenschaft im Grundrisse: mit Hegels eigenhändigen Notizen und den mündlichen Zusätzen (Frankfurt am Main: Suhrkamp, 1989), 26.

Bibliography Akker, Chiel van den. 2009. Beweren en Tonen. Waarheid, Taal en het Verleden (Nijmegen) — 2011. ‘Uitspraak en/of representatie,’ Algemeen Nederlands Tijdschrift voor Wijsbegeerte, 103(4): 263-267 — 2012. ‘The Exemplification Theory of History. Narrativist Philosophy and the Autonomy of History,’ Journal of the Philosophy of History, 6(2): 236-257 Ankersmit, Frank. 1983. Narrative Logic. A Semantic Analysis of the Historian’s Language (The Hague: Martinus Nijhoff Publishers) — 1990. De navel van de geschiedenis. Over interpretatie, representatie en historische realiteit (Groningen: Historische Uitgeverij) — 1993. De historische ervaring (Groningen: Historische Uitgeverij) — 1998. ‘Danto on Representation, Identity, and Indiscernibles,’ History and Theory, 37(1): 44-70 — 2000. ‘Representation as the Representation of Experience,’ Metaphilosophy, 31(1/2): 148-168 — 2001. Historical Representation (Stanford: Stanford University Press) — 2003. ‘Danto, History, and the Tragedy of Human Existence,’ History and Theory, 42: 291-304 — 2005. Sublime Historical Experience (Stanford: Stanford University Press) — 2009. ‘White’s “New Neo-Kantianism”: Aesthetics, Ethics, and Politics,’ in F. Ankersmit, E. Domańska, and Hans Kellner eds., Re-Figuring Hayden White (Stanford: Stanford University Press), 34-53 Ankersmit, F., E. Domańska, and Hans Kellner eds. 2009. Re-Figuring Hayden White (Stanford: Stanford University Press) — 2011. ‘Political and Historical Representation,’ in Ananta Charana Sukla ed., Art and Representation: Contributions to Contemporary Aesthetics (Westport and London: Praeger), 69-89 — 2012. Meaning, Truth, and Reference in Historical Representation (Ithaca and New York: Cornell University Press) Anscombe, Elizabeth. 1979. ‘Under a Description,’ Noûs, 13(2): 219-233 Aristotle. 1995. Poetics, trans. N. van der Ben and J.M. Bremer (Amsterdam: Athenaeum, Polak & Van Gennep) Beiser, Frederick. 2014. After Hegel. German Philosophy 1840-1900 (Princeton and Oxford: Princeton University Press) Bevir, Mark. 2000. ‘Historical explanation, folk psychology, and narrative,’ Philosophical Explorations, 3(2): 152-168 Boghossian, Paul. 2006. Fear of Knowledge. Against Relativism and Constructivism (Oxford: Clarendon Press) Bos, Jacques. 2010. ‘Individuality and interpretation in nineteenth-century German historicism,’ in U. Feest ed., Historical Perspectives on Erklären and Verstehen (Dordrecht: Springer Science + Business Media B.V.), 207-220 Brandom, Robert. 1998 [1994]. Making It Explicit. Reasoning, Representing, and Discursive Commitment (Cambridge MA and London: Harvard University Press) — 2011. Perspectives on Pragmatism. Classical, Recent, & Contemporary (Cambridge MA and London: Harvard University Press) Belting, Hans. 2003. Art History after Modernism (Chicago and London: University of Chicago Press) Bucciantini, Massimo, Michele Camerota, and Franco Giudice. 2015. Galileo’s Telescope. A European Story (Harvard University Press: Cambridge MA and London), trans. Catherine Bolton

150 

The Exemplifying Past

Burckhardt, Jacob. 1878 [1860]. The Civilization of the Renaissance in Italy [Die Kultur der Renaissance in Italien], trans. S.G.C. Middlemore (Vienna) Canary, R.H. and H. Kozicki eds. 1978. The Writing of History. Literary Form and Historical Understanding (Madison: University of Wisconsin Press) Carr, David. 1986. ‘Narrative and the Real World: An Argument for Continuity,’ History and Theory, 25(2): 117-131 — 2001. ‘Place and Time: On the Interplay of Historical Points of View,’ History and Theory, 40: 153-167 — 2014. Experience and History. Phenomenological Perspectives on the Historical World (Oxford: Oxford University Press) Carrier, David ed. 1998. History and Theory. Theme Issue 37. Danto and his Critics: Art History, Historiography and After the End of Art Carroll, Noël. 1988. ‘Art, Practice, and Narrative,’ The Monist, 71(2): 140-156 — 1997. ‘Danto’s New Def inition of Art and the Problem of Art Theories,’ British Journal of Aesthetics, 37(4): 386-392 — ed. 2000. Theories of Art Today (Madison and London: University of Wisconsin Press), 130-140 Clark, T.J. 1984. The Painting of Modern Life. Paris in the Art of Manet and his Followers (Princeton: Princeton University Press) Chartier, Roger. 1997. On the Edge of the Cliff. History, Language, and Practices (Baltimore and London: Johns Hopkins University Press), trans. Lydia G. Cochrane Collingwood, Robin. 1978 [1946]. The Idea of History (Oxford: Oxford University Press) Cowart, Georgia. 2008. The Triumph of Pleasure. Louis XIV and & the Politics of Spectacle (Chicago: University of Chicago Press) Danto, Arthur. 1962. ‘Narrative Sentences,’ History and Theory, 2(2): 146-179 — 1964. ‘The Artworld,’ The Journal of Philosophy, 61: 571-584 — 1965. Analytical Philosophy of History (Cambridge: Cambridge University Press) — 1981. The Transfiguration of the Commonplace. A Philosophy of Art (Harvard: Harvard University Press) — 1985. Narration and Knowledge (New York: Columbia University Press) — 1997. After the End of Art. Contemporary Art and the Pale of History (Princeton: Princeton University Press) — 1998. ‘The End of Art: A Philosophical Defense,’ in David Carrier ed., History and Theory. Theme Issue 37. Danto and his Critics: Art History, Historiography and After the End of Art, 127-143 — 1999. Philosophizing Art. Selected Essays (Berkeley, Los Angeles, and London: University of California Press) — 1999. The Body/Body Problem. Selected Essays (Berkeley, Los Angeles, and London: University of California Press) — 2000. ‘Art and Meaning,’ in Noël Carroll ed., Theories of Art Today (Madison and London: University of Wisconsin Press), 130-140 — 2005. The Philosophical Disenfranchisement of Art (New York: Columbia University Press) Davidson, Donald. 2001 [1984]. Inquiries into Truth and Interpretation (Oxford: Clarendon Press) — 1990. ‘The Structure and Content of Truth,’ Journal of Philosophy, 87(6): 279-328 — 1996. ‘The Folly of Trying to Define Truth,’ The Journal of Philosophy, 93: 263-278 — 2005. Truth, Language and History (Oxford: Clarendon Press) Donougho, Martin. 2007. ‘Art and History: Hegel on the End, the Beginning, and the Future of Art,’ in Stephan Houlgate ed., Hegel and the Arts (Evanston: Northwestern University Press), 179-215 D’Oro, Guiseppina. 2004. ‘Re-enactment and Radical Translation,’ History and Theory, 198-208

Bibliogr aphy

151

D’Oro, Guiseppina and James Connelly. 2017. ‘Collingwood, Scientism and Historicism,’ Journal of the Philosophy of History, 11(3): 275-288 Dray, William. 1964. Philosophy of History (Englewood Cliffs: Prentice-Hal) — 1995. History as re-enactment. R.G. Collingwood’s Idea of History (Oxford: Clarendon Press) Dummett, Michael. 2004. Truth and the Past (New York: Columbia University Press) — 2006. Thought and Reality (Oxford: Oxford University Press) Dussen, Jan van der. 1980. History as a Science: Collingwood’s Philosophy of History (Krips Repro: Meppel) Feest, U. ed. 2010. Historical Perspectives on Erklären and Verstehen (Dordrecht: Springer Science + Business Media B.V.) Frege, Gottlob. 1967. ‘The Thought: A Logical Inquiry,’ in P. Strawson ed., Philosophical Logic. Oxford Readings in Philosophy (Oxford: Oxford University Press), 17-38 — 1993 [1892]. ‘On Sense and Reference’ [1892] in: A.W. Moor ed. Meaning and Reference (Oxford: Oxford University Press), 23-42 Gadamer, Hans-Georg. 1980. Dialogue and Dialectic: Eight Hermeneutical Studies on Plato, trans. and intro. by P. Christopher Smith (New Haven: Yale University Press) Gilbert, Felix. 1990. History: Politics or Culture? Reflections on Ranke and Burckhardt (Princeton and Oxford: Princeton University Press) Ginzburg, Carlo. 2002. Wooden Eyes. Nine Reflections on Distance (London and New York: Verso) Goehr, Lydia. 2007. ‘Afterwords: An Introduction to Arthur Danto’s Philosophies of History and Art,’ History and Theory, 46: 1-28 Goodman, Nelson. 1976. Languages of Art. An Approach to A Theory of Symbols (Indianapolis and Cambridge: Hackett Publishing) Gombrich, Ernst. 1978. Meditations on a Hobby Horse and Other Essays on the Theory of Art (London and New York: Phaidon Press) Graf, Rüdiger. 2003. ‘Interpretation, Truth, and Past Reality. Donald Davidson meets history,’ Rethinking History, 7(3): 387-402 Grigoriev, Serge. 2008. ‘Beyond Radical Interpretation: Individuality as the Basis of Historical Understanding,’ European Journal of Philosophy, 17(4): 489-503 Hacking, Ian. 1995. Rewriting the Soul. Multiple Personality and the Sciences of Memory (Princeton: Princeton University Press) Haddock, Adrian. 2008. ‘Danto’s Dialectic,’ Philosophia, 483-493 Harries, Karsten. 1974. ‘Hegel on the Future of Art,’ Review of Metaphysics, 27: 677-696 Hegel, Georg. 1975. Aesthetics. Lectures on Fine Art, trans. T.M. Knox, Volume I (Oxford: Clarendon Press) — 1989. Grundlinien der Philosophie des Rechts oder Naturrecht und Staatswissenschaft im Grundrisse: mit Hegels eigenhändigen Notizen und den mündlichen Zusätzen (Frankfurt am Main: Suhrkamp) Hempel, Carl. 1942. ‘The Function of General Laws in History,’ The Journal of Philosophy, 39(2) (1942), 35-48 Himmelfarb, Gertrude. 1984. The Idea of Poverty. England in the Early Industrial Age (New York: Alfred A. Knopf) Houlgate, Stephan. 2007. ed., Hegel and the Arts (Evanston: Northwestern University Press) Jansen, Harry. 2015. ‘Time, Narrative, and Fiction: The Uneasy Relationship between Ricoeur and a Heterogeneous Temporality,’ History and Theory, 54(1): 1-24 — 2016. ‘Rethinking Burckhardt and Huizinga. A Transformation of Temporal Images,’ Storia della Storiografia, 70(2): 95-112

152 

The Exemplifying Past

Jay, Martin. 2005. Songs of Experience. Modern American and European Variations on a Universal Theme (Berkeley, Los Angeles, and London: University of California Press) Kuhn, Thomas. 1996. The Structure of Scientific Revolution (Chicago and London: The University of Chicago Press, 3th edition) Kuukkanen, Jouni-Matti. 2015. Postnarrativist Philosophy of Historiography (Basingstoke: Palgrave Macmillan) Lorenz, Chris. 1998. ‘Can Histories by True? Narrativism, Positivism, and the “Metaphorical Turn,”’ History and Theory, 37(3): 309-329 Lucas, D.W. 1980 [1968]. Aristotle Poetics. Introduction, Commentary and Appendix (Oxford: Clarendon Press) Malpas, Jeff. 2008. ‘On Not Giving Up the World – Davidson and the Grounds of Belief,’ International Journal of Philosophical Studies, 16(2): 201-215 — 2011. ed., Dialogues with Davidson. Acting, Interpreting, Understanding (Cambridge MA and London: The MIT Press) Marchetti, Giancarlo. 2011. ‘Davidson and the Demise of Representationalism,’ in Jeff Malpas ed., Dialogues with Davidson. Acting, Interpreting, Understanding (Cambridge MA and London: The MIT Press), 113-128 Margaret, Jacob. 2006. The Origins of Freemasonry. Facts & Fictions (Philadelphia: University of Pennsylvania Press) Marin, Louis. 1981. Le Portrait du Roi (Paris: Éditions de Minuit) Mink, Louis. 1966. ‘The Autonomy of Historical Understanding,’ History and Theory, 5: 24-47 — 1978. ‘Narrative form as a cognitive instrument,’ in R.H. Canary and H. Kozicki eds., The Writing of History. Literary Form and Historical Understanding (Madison: University of Wisconsin Press) — 1987. Historical Understanding (Ithaca and London: Cornell University Press) Moor, A.W. 1993. ed. Meaning and Reference (Oxford: Oxford University Press) Nanay, Bence. 2016. Aesthetics as Philosophy of Perception (Oxford: Oxford University Press) Nietzsche, Friedrich. 1994. Die Geburt der Tragödie (Könemann Verlagsgesellschaft: Cologne) Paul, Herman. 2011. Hayden White. The Historical Imagination (Cambridge: Polity Press) Peters, Rik. 2013. History as Thought and Action. The philosophies of Croce, Gentile, de Ruggiero and Collingwood (Exeter: Imprint Academics) Pippin, Robert. 2002. ‘What Was Abstract Art? (From the Point of View of Hegel),’ Critical Inquiry, 1-24 Quine, Willard Van Orman. 2003 [1953], ‘Two dogmas of empiricism,’ From a Logical Point of View (Cambridge and London: Harvard University Press), 20-46 — 1992. Pursuit of Truth. Revised Edition (Cambridge and London) Rorty, Richard. 1998. Truth and Progress. Philosophical Papers, Volume 3 (Cambridge: Cambridge University Press) Roth, Paul. 1978. ‘Paradox and Indeterminacy,’ The Journal of Philosophy, 75(7): 347-367 — 1989. ‘How Narratives Explain,’ Social Research, 56(2): 449-478 — 2002. ‘Ways of Pastmaking,’ History of the Human Sciences, 15(4): 125-143 — 2012. ‘The Pasts,’ History and Theory, 51: 313-339 Ricoeur, Paul. 1984. Time and Narrative, Vol. 1 (Chicago: The University of Chicago Press) Rollins, Mark. 2012. Danto and his Critics. Second Edition (Malden, Oxford, and Chichester: Blackwell Publishing) Sennett, Richard. 1974. The Fall of Public Man (Cambridge: Cambridge University Press) Strawson, Peter. 1967. ed., Philosophical Logic. Oxford Readings in Philosophy (Oxford: Oxford University Press)

Bibliogr aphy

153

— 1967 [1961]. ‘Singular Terms and Predication,’ in Strawson ed., Philosophical Logic (Oxford: Oxford University Press), 69-88 Sukla, Ananta Charana. 2011. ed., Art and Representation: Contributions to Contemporary Aesthetics (Westport and London: Praeger) Vann, Richard T. (1987). ‘Louis Mink’s Linguistic Turn,’ History and Theory, 26(1): 1-14 Walsh, W.H. 1968. Philosophy of History: An Introduction. Revised Edition (New York and Evanston: Harper & Row) Weberman, David. 1997. ‘The Nonfixity of the Historical Past,’ The Review of Metaphysics, 50(4): 749-768 Wedgwood, Cicely. 1957. The Thirty Years War (Pelican Books) White, Hayden. 1973. Metahistory. The Historical Imagination in Nineteenth-Century Europe (Baltimore and London: Johns Hopkins University Press) — 1978. Tropics of Discourse. Essays in Cultural Criticism (Baltimore and London: Johns Hopkins University Press) — 1982. ‘The Politics of Historical Interpretation: Discipline and De-Sublimation,’ Critical Inquiry, 9(1): 113-137 — 1987. The Content of the Form. Narrative Discourse and Historical Representation (Baltimore and London: Johns Hopkins University Press) — 2010. The Fiction of Narrative. Essays on History, Literature, and Theory, 1957-2007, ed. and intro. R. Doran (Baltimore: Johns Hopkins University Press) Williams, Bernard. 2002. Truth and Truthfulness (Princeton: Princeton University Press) — 2006. The Sense of the Past. Essays in the History of Philosophy, ed. and intro. by Myles Burnyeat (Princeton: Princeton University Press) Wittgenstein, Ludwig. 1967. Philosophische Untersuchungen (Suhrkamp 1967) Wright, Georg Hendrik von. 2004 [1971]. Explanation and Understanding (Ithaca and London: Cornell University Press) Zammito, John. 2005. ‘Ankersmit and Historical Representation,’ History and Theory, 44: 155-181 Zeleňák, Eugen. 2009. ‘Exploring holism in Frank Ankersmit’s historical representation,’ Rethinking History, 13(3): 357-369 — 2013. ‘Using Goodman to Explore Historical Representation,’ Journal of the Philosophy of History, 7(3): 371-395 — 2015. ‘Two Versions of A Constructivist View of Historical Work,’ History and Theory, 54(2): 209-225

Index aesthetic evaluation 90n34 anachronism 14-15 analytical philosophy of history 16, 131 and substantive philosophy of history 16n8, 128n5, 137-140 analytic-synthetic distinction 71 Ankersmit, Frank 66, 102n12, 103, 120 and anti-representationalism 19n14 and historical experience 123-125 on narrative substance 61n62, 112-118, 120 on rationality of history writing 18 on representation 19, 69n38, 77-88, 91-93, 114 on translation rules 63, 121 anti-realism 28n13 anti-representationalism 19n14, 69, 78n7; see also Davidson on conceptual schemes Aristotle 27n12, 51-52, 57, 61, 121n45 Aurelius, Marcus 75-77, 80-81, 85-86, 92-97 autonomy of history 20, 103, 112, 116, 119, 121-122 two senses of 110n23 Baas, Maarten 126-127, 129-130, 133-134, 137 Belting, Hans 144-145 Brandom, Robbert 72n43, 72n45 Bucciantini, Massimo 35, 49 Burckhardt, Jacob 13, 17, 96-97 Camerota, Michele 35, 49 Campra, André 102, 114, 118 Canova, Antonio 93 Charlemagne 93-94 Carr, David 104, 106n18, 121-122 Carroll, Noël 133n14, 135n18, 141n26 chronicle 51, 53-55, 57, 59-60, 62, 65-68; see also Ideal Chronicle Cigoli, Ludovico 21-22, 28-29, 33-34, 38, 46 Collingwood, Robin 22-23, 32-38, 46 Clark, Timothy 146 common view of history 128n5, 140-145, 147, 148n38 conceptual relativism 40, 43-45 conceptual schemes see Davidson confirmation 17, 56-57, 59, 71-73, 142-144; see also evidence concreteness of historical thesis 13, 17-18, 120, 124-125, 143, 145 Constantine 75 constructivist literature 39n36 Cowart, Georgia 99, 102, 107, 109-114, 118-119 Danto, Arthur and autonomy of history 20, 103, 110n23, 116, 121 on conditions for something to be a work of art 132-133, 136-137

and end of art 19-20, 127-148 and exemplification 96n48, 103, 113, 118 on historical significance 16, 59, 66, 101-102, 106, 136 and Ideal Chronicler 100-102, 122 and indiscernible objects 127-128, 129, 131-132 and interpretation of art 133-134, 146 and narrative sentences 56n13, 67, 100-101, 122 on narrative as temporal whole 46-47 on openness of the future 16 and re-description of action 14n2, 15-16, 47-48, 131 on representation 81-83, 92-93 and social individuals 104-106, 114, 117, 120 on substantive philosophy of history 137-140 Davidson, Donald and anti-representationalism 19n14 on conceptual schemes 39-45, 69 and interpretation 22-25, 30-34, 35n30, 38 and meaning 24n4, 25 and realism 28n13 on semantics holism 117n36 on truth 24n4, 25, 41-42 Donne, John 21, 49 D’Oro, Guiseppina 22n2, 32, 36n31 Dray, William 37n34, 148 Duchamp, Marcel 131, 134-137, 143 Dummett, Michael 26n10, 28n13, 55n11 eidos 120 Einfühlen 35 evaluation of statements 56, 72 of historical thesis 18, 102n13, 142-143 of art see aesthetic evaluation evidence and historical thesis 17-18, 48, 103, 105-110, 112-113, 116, 119-120, 125, 142-145 and justification 18, 27, 55, 65, 108, 112-113, 120, 125 and meaning 24n4, 28, 55n11 and knowledge of the past 19n14, 22, 28, 36-37, 48, 63, 142, 144 see also confirmation exemplification passim and expression 95-96 and Goodman 18, 93n42, 96n48, 103n14 and historical experience 123-124 and instance of type 61 as semantic term 18-19, 71 theory of history 17, 70, 102-103, 108, 112, 118, 123

156  experiencing the past, the illusion of 36n33, 124n60 Franklin, Benjamin 72 Frege, Gottlob 26n8, 26n9, 117n36 fiction 52-53, 56-58, 60-62, 64-65, 67-68, 72 Galilei, Galileo 21-23, 25-28, 33-35, 38, 42n43, 46, 49 Giudice, Franco 35, 49 Gombrich, Ernst 83, 90-91 Goodman, Nelson 18-19, 77n4, 81n14, 82n17, 83, 84n24, 93, 96n48, 103n14, 120n42 Grappa, Andrea Vittorelli of Basano del 21 Grigoriev, Serge 22n2, 32, 38n35 Guardi, Francesco 123-124 Haddock, Adrian 100 Hacking, Ian 14, 16 Hamilton, Richard 144-145 Harries, Karsten 136 Hegel, Georg 16n8, 89n32, 128, 132, 138, 140, 147-148 Herodotus 51 hermeneutics 35, 122 Himmelfarb, Gertrude 116 historical experience see Ankersmit and historical experience historical research as distinguished from historical writing 56, 64-65 historical thesis passim historical understanding, heart of 13, 60, 103, 112 historicists 104n15, 129 holism 117 Homer 15 Horky, Martin 33 Ideal Chronicle(r) see Danto identity see representation and identity imitation see mimesis indeterminacy of interpretation 25 of the past 13-16, 19, 64, 68-69, 144 of translation 25n6 Ingres, Jean-Auguste-Dominique 93-95 interpretation of art 89-90, 133-136, 146-147 correctness of 90, 133-134 as being panoramic 112-113 of other minds 22-38 as part of historical research 65 see also Davidson on interpretation Impressionism 135, 145 irony 29, 58n20 Jacob, Margaret 60 Jansen, Harry 121n46 jokes 28

The Exemplifying Past

justification of beliefs 23, 24n4, 27, 28n13; see also evidence justificationist theory of meaning 24n4, 25n6, 28, 55n11 Kepler, Johannes 27 Kuhn, Thomas 39n36 Kuukkanen, Jouni-Matti 13n1, 18, 102n13, 117n36 Lemot, François-Frédéric 93 literary theory 54, 57, 62 Lorenz, Chris 57n14 Louis XIV 92 Mackenzie, Alexander 14 Malevich, Kazimir 135-136, 143 Magritte, René 86-90 Manet, Édouart 146 meaning 23, 24n4, 25-26, 28, 30, 47-49, 54, 55n11, 67, 70n41, 71-72, 115-116, 117n36 of art 89-90, 132-136, 146 of the future 140 of the past 16-17, 102, 107, 121, 139 metaphor see representation methodological collectivism 104 Mink, Louis 52, 54-61, 63-66, 68-71, 73, 103, 109-113, 117-118, 120, 121n49 mimesis 27n12, 51-52, 121, 130, 136, 146 Napoleon, Bonaparte 93-94 narrative passim Aristotelian view of 104, 121 as artifice 56, 58-61 as cognitive instrument 58-61, 65 having ontological force 64 as guide 18, 73 as intersecting 61-62 Platonic view of 103-104, 120, 124 and theory 17, 59 as temporal whole 46-47, 69 see also plot narrative coherence 19, 55-56, 69, 113n27, 137, 139, 141-142 narrative sentence see Danto narrative substance see Ankersmit narrative truth 18, 52-57, 60, 66, 71-73, 112, 120 narrativist philosophy of history 13n1, 19, 57, 70, 92, 112, 115, 117n36, 120-121 Nietzsche, Friedrich 82n18 Novalis 89n31 ontological individualism 104 Paul, Herman 52n6 Petrarch 17, 47-49 Pippin, Robert 147n36 philosophical view of history 19-20, 125, 143-145, 147-148 Plato 120n44, 130n9

157

Index

Pope Paul V 46 pop art 135, 145 plot 27n12, 51-52, 55-57, 58n15, 58n20, 59-60, 62 pragmatism 54, 57, 71-73 principle of charity 31 principle of humanity 31 Quine, Willard von Orman 19n14, 24n4, 25n6, 71 radical interpretation see Davidson on interpretation radical translation 24n4 rational relation with the world, as distinguished from causal relation with the world 45-46 rationality of beliefs 30-32 of history writing 18 realism 28n13, 81, 135 reality, causal relation to 45 reasons and sense of the past 16 and beliefs 23, 27, 31, 34, 38, 44, 46 re-enactment of past thought 22, 34-38 reference 18, 26n9, 66, 71, 108, 113 anaphoric 55 and history of what the term refers to 114-116 relativism see conceptual relativism representation accuracy of 19n14, 138, 142, 145 correctness of 90, 138 empiricist notion of 69-70, 78n7 and identity 81-86 as instrument of rhetoric 77-78, 95 and metaphor 89, 92-94, 96n48, 97 as proposals 19, 78, 114 as symbolization 19, 83, 90-91, 94 theories of 19, 75-98 and truth see truth see also mimesis representing-as 69n38, 94 representing-that 69n38, 94 retroactive alignment 59-60, 65-67 rhetoric see representation Rigaud, Hyacinthe 92 Ricoeur, Paul 104, 120-121 Rorty, Richard 19n14 Rousseau, Jean-Jacques 79n10 Roth, Paul 14n2, 16, 68n35 Scholte, Rob 145-147 seeing, ambiguity of 44, 94 seeing-as 69n38, 92, 94 seeing-in 13, 16, 20, 105-107, 112, 116, 119-120, 124-125, 145 seeing-that 69n38, 94 Sennett, Richard 123-124

Sforza, Francesco 13 social change 68, 103, 117, 119 as the primary concern of historians 60-61, 108, 119 as taking place not in individuals but in societies 105-110 social individuals as product of historical understanding 108-109 see Danto and social individuals spirit of an age 13, 112, 145 Spooner, Paul 145-146 state of mind 112, 145, 148 stimuli, as reasons for holding beliefs to be true 44, 46 symbolization see representation Tarski, Alfred 42 thought and sensation 36-37 Titian 146 translation rules 62-63, 121 truth 26-28, 31, 40-43, 54-56, 100 historians on 54, 66 and historical thesis see narrative truth and holding true 23-26, 29-30, 33, 46, 48 possessing concept of 22, 39, 49 pragmatist understanding of 71-73 of (artistic) representations 90 as semantic concept 22, 25, 57, 71-73 truth-conditions 23-24, 26, 30, 38, 48-49 in art 27n12 as consisting in part of reference to a narrative 48, 67, 109 and meaning 24n4, 25n6, 26, 28, 30, 32n21, 37, 55n11, 71-72 truthful 141 truth-value 23, 26-27, 54, 72 verification and historical thesis 120 and statements 28, 48 theory of meaning 71 see also evidence Von Humboldt, Wilhelm 104n15 Von Ranke, Leopold 100n4, 104n15 Warhol, Andy 128, 136, 141, 144, 147-148 Watteau, Antoine 99, 102-103, 107-111, 114, 119 Weberman, David 64 Wedgwood, Cicely 105-106, 110, 113, 119 White, Hayden 52, 58-59, 60n23, 61-62, 63n28, 66 Windelband, Wilhelm 104n15 Wittgenstein, Ludwig 94n45 Zammito, John 114 Zeleňák, Eugen 77n4, 119-120 Zusammenhang see narrative coherence