Biblical and Ancient Greek Linguistics [7 — 2018] 9781532673467, 1532673469

Citation preview

Biblical and Ancient Greek Linguistics Volume 7 — 2018 Contents STANLEY E. PORTER Editor’s Foreword: More Important Words on Words and More

5

PAUL L. DANOVE The Conceptualization of Communication in the New Testament: A Feature Description

7

NICHOLAS P. LUNN Categories of Contrast in New Testament Greek

31

JOHN J.H. LEE Cohesive Harmony Analysis for Ancient Greek: SelPap I:112 and PMich VIII:491 as a Test Case

81

RYDER A. WISHART Monosemy: A Theoretical Sketch for Biblical Studies

107

STANLEY E. PORTER Aspect and Imperatives Once More

141

JOSEPH D. FANTIN May the Force Be with You: Volition, Direction, and Force: A Communicative Approach to the Imperative Mood

173

2

Biblical and Ancient Greek Linguistics 7

JAMES D. DVORAK “Evidence that Commands a Verdict”: Determining the Semantics of Imperatives in the New Testament Ancient Sources Index Modern Authors Index

201

225 238

Senior Editors Professor Dr. Stanley E. Porter Dr. Matthew Brook O’Donnell Assistant Editors Dr. Christopher D. Land Dr. Francis G.H. Pang Editorial Board Dr. Martin Culy (Briercrest College and Seminary, Canada) Dr. Paul Danove (Villanova University, USA) Dr. Christopher D. Land (McMaster Divinity College, Canada) Dr. Matthew Brook O’Donnell (University of Pennsylvania, USA | McMaster Divinity College, Canada) Professor Dr. Stanley E. Porter (McMaster Divinity College, Canada) Dr. Catherine Smith (University of Birmingham, UK) Dr. Jonathan Watt (Geneva College, USA) Dr. Cynthia Long Westfall (McMaster Divinity College, Canada) Biblical and Ancient Greek Linguistics (BAGL) is an international journal furthering the application of modern linguistics to the study of Ancient and Biblical Greek, with a particular focus on the analysis of texts, including but not restricted to the Greek New Testament. The journal is hosted by McMaster Divinity College (mcmasterdivinity.ca) and works in conjunction with its Centre for Biblical Linguistics, Translation and Exegesis (cblte.org) and the OpenText.org organization (www.opentext.org) in the sponsoring of conferences and symposia open to scholars and students researching in Greek linguistics who are interested in contributing to advancing the discussion and methods of the field of research. BAGL is a refereed online and print journal dedicated to distributing the

4

Biblical and Ancient Greek Linguistics 7

results of significant research in the area of linguistic theory and application to biblical and ancient Greek, and is open to all scholars. Accepted pieces are in the first instance posted online in page-consistent pdf format, and then (except for reviews) are published in print form each volume year. Submissions to BAGL BAGL accepts submissions in five categories, and manuscripts are to be labeled as such at the time of submission: Article Exploration Note Response Review/Review article

Submissions should follow the BAGL style-guide found at http://bagl.org, and should include an abstract, not longer than 100 words, two to six keywords, and identification of the type of article (which will be noted at the time of posting and publication). Submissions not following the style-guide will be returned for revision before being considered. Submissions should be sent in electronic form (Word or RTF) to Stanley E. Porter at [email protected]. Assessment and response will be made within approximately two months of submission. Accepted submissions should be posted online within two months of acceptance. The online form of BAGL is found at http://bagl.org. Copyright © 2018 Wipf and Stock Publishers. All rights reserved. Except for brief quotations in critical publications or reviews, no part of this book may be reproduced in any manner without prior written permission from the publisher. Write: Permissions, Wipf and Stock, 199 W. 8th Ave., Eugene, OR 97401. ISBN 13: 978-1-5326-7346-7 www.wipfandstock.com Manufactured in the U.S.A.

[BAGL 7 (2018) 5–6]

EDITOR’S FOREWORD: MORE IMPORTANT WORDS ON WORDS AND MORE Stanley E. Porter McMaster Divinity College, Hamilton, ON, Canada

The papers in this volume are organized into two broad categories. The first grouping is a set of four papers reflecting widely differing approaches to Greek description. The first is an article by Paul Danove using Construction Grammar to identify five features of how communicative events are grammaticalized in New Testament Greek verbs. The results provide identification, description, classification, interpretation, and translation of these verbs. Readers familiar with Greek linguistics will be well aware of Danove’s continuing work in the area of Construction Grammar, pioneered by such linguists as Charles Fillmore and Paul Kay. The second article in this group is by Nicholas Lunn on contrast within New Testament Greek. Lunn identifies and describes seven different contrastive formulations, and then exemplifies them within the Greek New Testament. He also offers concluding comments that provide insights into how his previous descriptions can aid translation and exegesis. The third paper, by John Lee, draws upon Ruqaiya Hasan’s Cohesive Harmony Analysis (CHA), developed within a Systemic Functional Linguistics (SFL) framework, to provide a reading of two ancient Greek papyri. In this article, Lee first defines CHA within an ancient Greek framework and then shows how, with suitable modifications to account for the fact that Greek is a fusional language, CHA helps use to describe ancient texts. Lee’s method is rigorous and detailed, but it takes the notion of cohesion to new levels suitable for the study of ancient Greek. The final paper in this first group, by Ryder Wishart, deals with the broad topic of monosemy. Wishart weighs in on

6

Biblical and Ancient Greek Linguistics 7

the side of monosemy with an examination of the major issues involved and then application to three examples within the field of biblical studies. To define monosemy, Wishart utilizes a number of different approaches, including lexical semantics, corpus study, and a sign-based approach, drawing upon the work of Charles Ruhl and the Columbia School of Linguistics. He develops what he calls a lowest common denominator approach. Once he has defined his terms and concepts, Wishart applies the notion of linguistic monosemy—not just lexical monosemy—to the πίστις Χριστοῦ debate, discussion of the Greek article, and treatments of verbal aspect, all topics of continuing and recent debate. The second group of papers is focused upon the Greek imperative and command/prohibition. These three papers were first presented at the Biblical Greek Language and Linguistics Section of the Society of Biblical Literature 2016 annual meeting in San Antonio, TX. The first paper, by Stanley Porter, focuses upon verbal aspect and imperatives and their related commands and prohibitions. Utilizing SFL, he discusses imperatives in relation to the Greek mood system, verbal aspect, and three abiding issues regarding frequency, speech functions, and use in indirect speech acts. By contrast, Joseph Fantin approaches imperatives from what he calls a communicative approach. He outlines his approach by focusing upon a number of environments in which the imperative is used in ways that require more explanation than traditional studies can provide. In the third paper of this group, James Dvorak outlines two basic expositional frameworks for imperatives. Along the way, Dvorak offers a short course in SFL and its context-dependent descriptions and then a critique of the notion of contextless meanings. We recognize that this volume is longer than most previous volumes of the journal, but we believe that the papers merit timely publication, as there are a number of significant ways that current discussions might benefit from the ideas proposed in them.

[BAGL 7 (2018) 7–29]

THE CONCEPTUALIZATION OF COMMUNICATION IN THE NEW TESTAMENT: A FEATURE DESCRIPTION Paul L. Danove Villanova University, Villanova, PA, USA Abstract: This article develops five features that describe the conceptualizations of the event of communication grammaticalized by New Testament verbs and uses these features to formulate a model of the observed New Testament usages of communication. The discussion resolves all NT occurrences of verbs that designate communication into one of twenty-one usages with distinct feature descriptions, offers guidelines for interpreting and translating verbs with each usage, and clarifies elements of the conceptualization of communication in relation to specific examples. (Article) Keywords: Feature, communication, semantic, syntactic, verbal usage.

1. The Event of Communication An event is a cognitive schema of an action in which two, three, or four entities are set in a particular relation to each other.1 The event of communication logically includes three entities: one who communicates, what is communicated, and the interpreter of the communication. These descriptions correspond to the thematic roles, Agent (the entity that actively instigates an action and/or is the ultimate cause of a change in another entity), Content (the content of a sensory, cognitive, or emotional event or activity), and Experiencer (the animate entity that undergoes a sensory, cognitive, or emotional event or activity).2 In this event, 1. 2.

See Goddard, Semantic Analysis, 197–98. These and subsequent semantic functions receive description

8

Biblical and Ancient Greek Linguistics 7

the Agent produces the Content of communication, and the Experiencer interprets that Content. 2. Features of the Conceptualization of the Event of Communication The bare concept of communication is qualified in general and specific ways. General qualifications are described by features that specify the conceptualization of communication associated with a number of verbs. Specific qualifications are associated with the unique denotation of each verb. Verbs that grammaticalize a conceptualization of communication with the same features constitute a verbal usage. The following discussion develops five usage features that specify various constraints on the conceptualization of communication: secondary emphasis, orality, subject affectedness, functionality, and suppression.3 2.1 Feature #1: Secondary Emphasis In the conceptualization of communication, the Agent communicates Content to an Experiencer. Non-passivized verbs of communication grammaticalize this event by placing primary emphasis on the Agent and raising it as the first (subject) complement. The selection of the second complement, however, depends on the relative emphasis that each verb places on the Content and Experiencer. Among the 121 NT verbs of communication, 101 place secondary emphasis only on the Content; fifteen place secondary emphasis only on the Experiencer; and five present distinct usages with secondary emphasis on the Content and on the Experiencer. At least among the last five verbs, usages with secondary emphasis on the Experiencer have the implication that the Content is interpreted successfully.4 Secondary emphasis on the Content highlights the

according to the thematic roles developed in Saeed, Semantics, 139–71, and Danove, Linguistics and Exegesis, 31–45. 3. These usage features constitute a development of those introduced in Danove, “αἰτέω/αἰτέοµαι,” 101–18. 4. Lehrer, “Verbs of Speaking,” 155.

DANOVE Conceptualization of Communication

9

production and transmission of the Content (the Agent causes the message to go to the Experiencer), whereas secondary emphasis on the Experiencer highlights the reception and interpretation of the Content (the Agent causes the Experiencer to receive the message).5 When the verbs are passivized, they consistently raise the entity with secondary emphasis as the verbal subject. The descriptions of features note secondary emphasis by arranging the three entities of communication in the order of decreasing emphasis. In the following examples, λαλέω ‘speak’ places secondary emphasis on the Content (ACE) and διδάσκω ‘teach’ places secondary emphasis on the Experiencer (AEC): ACE

ἄλλην παραβολὴν ἐλάλησεν αὐτοῖς (Matt 13:33) He spoke another parable to them.

AEC

κύριε, δίδαξον ἡµᾶς προσεύχεσθαι, καθὼς καὶ Ἰωάννης ἐδίδαξεν τοὺς µαθητὰς αὐτοῦ (Luke 11:1) Lord, teach us to pray just as John taught his disciples.

In the former example, the secondary emphasis on the Content indicates that the parable (Content) was successfully produced and moved to “them.” In the latter example, the secondary emphasis on Jesus’s disciples (Experiencer) indicates that they will receive Jesus’s teaching on praying. The Feature Model notes secondary emphasis at the beginning of the usage description so that the usages of λαλέω and διδάσκω in the noted occurrences have the following format: λαλέω: ACE διδάσκω: AEC

5. Most frequently, discussions of the Content/Experiencer emphasis employ a movement (cause to go)/possession (cause to have) distinction: cf. Pinker, Learnability, 48, 63; Speas, Phrase Structure, 87–89; Pesetsky, Zero Syntax, 135–38; and Hovav and Levin, “Dative Alternation,” 134. However, the movement (cause to go)/reception (cause to receive) distinction better explains the implication that the emphasized Experiencer successfully interprets the Content/message: cf. Goldberg, “Argument Structure,” 46, 49–52. Although these authors are concerned with “dative alternation” among various classes of English verbs (X tells Y Z or X tells Z to Y), the movement/reception constraints on the English verbs appear to parallel exactly those of Greek verbs of communication.

10

Biblical and Ancient Greek Linguistics 7

2.2 Feature #2: Orality Some verbs are restricted to conceptualizations of oral communication in which the Agent produces speech or sound, and the remaining verbs either are restricted to or tolerate the conceptualization of other modes of communication. For example, the conceptualization of λέγω ‘say’ is restricted to oral communication; that of δείκνυµι ‘show’ is restricted to non-oral (i.e., visual) communication; and that of µιµνῄσκω ‘remind’ accommodates both oral and non-oral modes of communication (reminding by speaking or writing or showing). This discussion distinguishes between verbs that are restricted to oral communication (+or) and all other verbs of communication (–or): +or λέγει αὐτοῖς ὁ Ἰησοῦς, Πόσους ἄρτους ἔχετε; (Matt 15:34) Jesus says to them, “How many loaves do you have?” –or δείκνυσιν αὐτῷ πάσας τὰς βασιλείας τοῦ κόσµου καὶ τὴν δόξαν αὐτῶν (Matt 4:8) He shows to him all the kingdoms of the world and their glory.

The Feature Model introduces the ±or distinction after the notation on secondary emphasis. λέγω: ACE +or δείκνυµι: ACE –or

2.3 Feature #3: Subject Affectedness Greek verbs designate communication with usages that employ active, middle, and passive base forms, and a majority of verbs with active and middle base forms admit to passivization. Greek active, middle, and passive base forms signal differing conceptualizations of the affectedness of the first complement (subject/Agent).6 Active base forms of Greek (and English) verbs typically provide no guidance in determining whether the first complement is affected. Thus, both ὁράω ‘see’ and πάσχω ‘suffer’ use active base forms in Greek (and English), even though the one who sees need not be affected, while the one who suffers

6. Lyons, Linguistics, 373, discusses the nature of this affectedness; cf. Allan, Middle Voice, 19–20. Saeed, Semantics, 162–65, considers various categories of affectedness.

DANOVE Conceptualization of Communication

11

necessarily is affected. Middle base forms of verbs of communication indicate that the subject’s affectedness is not introduced by an entity internal to the event (Content or Experiencer).7 Passive base forms, in contrast, indicate that the subject’s affectedness is introduced directly by one of the entities internal to the event (Content or Experiencer). This discussion defines the subject affectedness signaled by middle base forms as “external affectedness” because an entity external to the event introduces the affectedness. The affectedness signaled by passive base forms is defined as “internal affectedness” because an entity internal to the event introduces the affectedness.8 An example of active/middle usages of communication with the same Greek verb appears with αἰτέω ‘ask’. Since English verbs do not mark for external affectedness, this and following translations of verbs with middle usages introduce “with affect” in brackets after the subject. Act. τίνα δὲ ἐξ ὑµῶν τὸν πατέρα αἰτήσει ὁ υἱὸς ἰχθύν (Luke 11:11) The son will ask what father among you for a fish? Mid. ᾐτήσατο τὸ σῶµα τοῦ Ἰησοῦ (Matt 27:58) He [with affect] asked [Pilate] for the body of Jesus.

Semantically, verbs with both active and middle usages of communication (Cmm.) are three-place because they require completion by three arguments. Syntactically, they are ditransitive (ditr.) because the subject is an Agent and all three arguments must be realized when they do not satisfy the requirements for permissible omission of verbal complements.9

7. Further discussion of subject affectedness appears in Danove, Verbs of Transference, 22–23. 8. Usages characterized by the feature “external affectedness” receive further consideration in Allan, Middle Voice, 112–14; Rijksbaron, Syntax and Semantics, 147–50; Bakker, “Voice, Aspect, and Aktionsart,” 36. 9. Null complements may be either definite, that is, having a definite semantic content retrievable from the context, or indefinite, that is, having no definite semantic content retrievable from the context. Definite null complements receive development in Mittwoch, “Idioms,” 255–59; Matthews, Syntax, 125–26; Allerton, Valency, 34, 68–70. Indefinite null complements receive development in Fraser and Ross, “Idioms,” 264–65; Sag and Hankamer,

12

Biblical and Ancient Greek Linguistics 7

Passive base forms signal that the subject/Agent is internally affected (pass.) because it is co-referential with another entity of the event, either the Content [C = A] or the Experiencer [E = A]. As a consequence, the Agent acts on itself. When verbs adopt passive base forms, they consistently omit the co-referential Content or Experiencer complement. Active/passive usages of communication appear with ἀναµιµνῄσκω ‘remind of/remind oneself of’ [= remember]).10 Since English verbs do not mark for internal affectedness, this and following translations of verbs with passive usages introduce the omitted but co-referential Content or Experiencer complement. Act. [Τιµόθεον] ὃς ὑµᾶς ἀναµνήσει τὰς ὁδούς µου τὰς ἐν Χριστῷ [Ἰησοῦ] (1 Cor 4:17) [Timothy], who will remind you of my ways in Christ [Jesus]. Pass. ἀναµιµνῄσκεσθε δὲ τὰς πρότερον ἡµέρας (Heb 10:32) But remind yourselves of [= remember] the former days.

Semantically, verbs with both active and passive usages of communication require completion by three arguments and are three-place. Syntactically, the English verbs that translate Greek verbs with passive usages of communication are ditransitive (ditr.) and lexically realize the omitted Content or Experiencer by a form of the reflexive pronoun “self.” Since Greek verbs with passive base forms consistently omit the co-referential Content or Experiencer, they are syntactically transitive (trans.). The Feature Model notes active, middle, and passive usages by introducing after the orality notation consecutively and in parentheses the event (Cmm.), the affectedness (act., mid., or pass.), and the syntactic notation (ditr. or trans.). With passive usages, the Feature Model introduces brackets around the

“Anaphoric Processing,” 325–45. 10. In English, remind is better characterized as requiring completion by a stimulus (the event or entity that brings about a change in a mental or psychological state, event, or activity), Content, and Experiencer because the subject need not be animate, as in “Cold rains remind me of Spring.” In the NT, however, µιµνῄσκω compounds consistently require completion by an animate first complement that functions as an Agent.

DANOVE Conceptualization of Communication

13

omitted co-referential entity to note its omission and places after the secondary emphasis description and in brackets the coreferential entities, here [E = A]: ἀιτέω: AEC +or (Cmm. act. ditr.) αἰτέω: AEC +or (Cmm. mid. ditr.) ἀναµιµνῄσκω: AEC –or (Cmm. act. ditr.) ἀναµιµνῄσκω: A[E]C [E = A] –or (Cmm. pass. trans.)

2.4 Feature #4: Functionality Each logical entity of the event of communication is associated with a specific semantic function (Agent, Content, and Experiencer). The conceptualization of communication, however, also accommodates the grammaticalization of information about the Content and not the Content directly. When this occurs, the grammaticalized entity functions as a Topic (the topic or focus of a mental or psychological state, event, or activity), and the Feature Model interprets this as a change in functionality of the Content to a Topic (C→T). Such changes in functionality produce usages of communication that differ only in this feature, as in the following examples of λαλέω (speak): C

ἐλάλει αὐτοῖς τὸν λόγον (Mark 2:2) He was speaking to them the word.

C→T ἀποδεξάµενος αὐτοὺς ἐλάλει αὐτοῖς περὶ τῆς βασιλείας τοῦ θεοῦ (Luke 9:11) Welcoming them, he was speaking to them about the reign of God.

The Feature Model introduces notations about changes in functionality after the secondary emphasis description. If more than one entity in that description receives special notation, these notations appear after the secondary emphasis description and in the same order as the entities within that description: λαλέω: ACE +or (Cmm. act. ditr.) λαλέω: ACE C→T +or (Cmm. act. ditr.)

2.5 Feature #5: Suppression Some verbs grammaticalize conceptualizations of communication by placing a definite semantic referent directly onto the

14

Biblical and Ancient Greek Linguistics 7

Content argument, suppressing the Content complement, and raising only the Agent and Experiencer as complements. This discussion marks suppression by placing the suppressed Content in braces, {C}. The omission of the Content complement by suppression differs from the omission of complements whose definite referent may be retrieved from another entity of the event (pass.). All but one of the verbs that suppress the Content complement have cognate nouns and pronouns with the same referent as the suppressed Content complement: ἀληθεύω ‘speak the truth’ (cf. ἀλήθεια), ἀποκρίνοµαι ‘speak a response’ (cf. ἀπόκρασις), δηµηγορέω ‘speak a public address’ (cf. δηµηγορία), ἐξοµολογέω ‘speak a declaration’ (cf. ὁµολογία), ἐπιτάσσω ‘speak an order’ (cf. ἐπιταγή), ἐπιτιµάω ‘speak a rebuke’ (cf. ἐπιτίµησις, ἐπιτιµόα), ἑτεροδιδασκαλέω ‘teach something different’ (cf. ἕτερος + διδασκαλία), εὐαCελίζω ‘proclaim good news’ (cf. εὐαCέλιον), ὁµολογέω ‘speak a declaration’ (cf. ὁµολογία), προσεύχοµαι ‘say a prayer’ (cf. προσευχή), ὑπερεντυγχάνω ‘speak an intercession’ (cf. ἔντυξις), and φωνέω ‘produce a sound’ (cf. φωνή). The remaining verb, ἀποτάσσοµαι ‘say good-bye’, has a specialized meaning that clarifies the definite referent of the Content. The NT presents no occurrences of suppression with verbs that have transitive passive usages of communication or that otherwise are restricted to grammaticalizing communication with a change in the functionality of the Content to a Topic. Since the Content is not realized, suppression produces for verbs transitive active and middle usages of communication. For some verbs, this is the only usage of communication. For the remaining verbs, suppression produces transitive usages with the same orality, secondary emphasis, and affectedness as their corresponding ditransitive usage(s), so that the ditransitive and transitive usages differ only in this feature. This receives illustration in the following occurrences of εὐαCελίζω ‘proclaim [as good news]/proclaim good news’: C

ἰδοὺ γὰρ εὐαγγελίζοµαι ὑµῖν χαρὰν µεγάλην (Luke 2:10) For, look, I proclaim to you great joy.

{C} [µε] εὐαγγελίσασθαι πτωχοῖς (Luke 4:18) [me] to proclaim good news to the poor.

DANOVE Conceptualization of Communication

15

The Feature model notes suppression by placing the suppressed Content of the event in braces, A{C}E or AE{C}, and specifying in braces after the secondary emphasis description that the referent of the Content is retrievable directly from the verb, {C = V}: εὐαγγελίζω: ACE +or (Cmm. mid. ditr.) εὐαγγελίζω: A{C}E {C = V} +or (Cmm. mid. trans.)

3. Usages ACE +or This discussion considers the seven observed usages of oral communication with secondary emphasis on the Content. 3.1 Usage #1: ACE +or (Cmm. act. ditr.) The forty-nine verbs with Usage #1 grammaticalize oral communication (+or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Content (ACE), and no suppression or change in the functionality of the Content.11 As the first example indicates, the translation of Greek verbs of commanding with usages ACE is difficult because English verbs of commanding consistently place secondary emphasis on the Experiencer (AEC). To address

11. Usage #1: ἀγγέλλω ‘announce’, αἰτέω ‘ask’, ἀναβοάω ‘cry out’, ἀναγγέλλω ‘tell’, ἀναγινώσκω ‘read’, ἀνακράζω ‘cry out’, ἀναφωνέω ‘speak loudly’, ἀπαγγέλλω ‘announce’, ἀπειλέω ‘speak a warning’, βοάω ‘shout’, γογγύζω ‘grumble’, διαγγέλλω ‘proclaim’, διαγογγύζω ‘complain’, διαλαλέω ‘discuss’, διασαφέω ‘explain’, διατάσσω ‘speak detailed orders’, διδάσκω ‘teach’, διερµηνεύω ‘interpret’, ἐκλαλέω ‘tell’, ἐξαγγέλλω ‘proclaim’, ἐπιλύω ‘explain’, ἐπιτάσσω ‘speak orders’, ἐπιτιµάω ‘speak strict orders’, ἐπιφωνέω ‘shout’, εὐαγγελίζω ‘proclaim [as good news]’, καταγγέλλω ‘announce’, κηρύσσω ‘proclaim’, κράζω ‘cry out’, κραυγάζω ‘cry out’, λαλέω ‘speak’, λέγω ‘say’, ὀµνύω ‘swear’, ὁµολογέω ‘declare’, παραγγέλλω ‘speak orders’, προκαταγγέλλω ‘announce previously’, προκηρύσσω ‘proclaim previously’, προλέγω ‘say previously’, προσλαλέω ‘speak’, προστάσσω ‘speak orders’, προσφωνέω ‘call’, προφητεύω ‘prophesy’, συζητέω ‘discuss’, συλλαλέω ‘speak’, συµβουλεύω ‘speak advice’, συντάσσω ‘speak orders’, φάσκω ‘claim’, φηµί ‘say’, φράζω ‘speak’, and φωνέω ‘speak loudly’.

16

Biblical and Ancient Greek Linguistics 7

this, the translation of these Greek verbs uses a paraphrase combining “speak” (ACE) plus “commands” or “orders.” τὸ ἄλαλον καὶ κωφὸν πνεῦµα, ἐγὼ ἐπιτάσσω σοι, ἔξελθε ἐξ αὐτοῦ καὶ µηκέτι εἰσέλθῃς εἰς αὐτόν (Mark 9:25) “Unspeaking and deaf spirit,” I speak orders to you, “Come out of him and no longer enter into him!” εἶπεν αὐτοῖς, Γεννήµατα ἐχιδνῶν (Matt 3:7) He said to them, “Offspring of vipers.”

3.2 Usage #2: ACE +or (Cmm. mid. ditr.) The twenty-eight verbs with Usage #2 grammaticalize oral communication (+or) with an externally affected Agent (mid.), secondary emphasis on the Content (ACE), and no suppression or change in the functionality of the Content.12 The translations signal external affectedness by placing “[with affect]” after the subject. αὐτὸς δὲ οὐδὲν ἀπεκρίνατο αὐτῷ (Luke 23:9) But he [with affect] did not respond anything to him. µαρτύροµαι δὲ πάλιν παντὶ ἀνθρώπῳ περιτεµνοµένῳ ὅτι ὀφειλέτης ἐστὶν ὅλον τὸν νόµον ποιῆσαι (Gal 5:3) Again I [with affect] testify to every man being circumcised that he is obliged to do the entire law.

3.3 Usage #3: AC[E] +or (Cmm. pass. trans.) Only διαλέγοµαι (discuss with one another) with Usage #3 grammaticalizes oral communication (+or) with an internally

12. Usage #2: αἰτέω ‘ask’, ἀπειλέω ‘speak warning’, ἀποκρίνοµαι ‘respond’, ἀπολογέοµαι ‘speak in defense’, ἀποφθέγγοµαι ‘speak’, διαλέγοµαι ‘discuss’, διαλογίζοµαι ‘discuss’, διαµαρτύροµαι ‘declare’, διαστέλλοµαι ‘speak strict orders’, διατάσσω ‘speak detailed orders’, διηγέοµαι ‘describe’, διϊσχυρίζοµαι ‘affirm’, ἐκδιηγέοµαι ‘declare’, ἐκτίθηµι ‘explain’, ἐντέλλοµαι ‘speak commands’, ἐξηγέοµαι ‘explain’, ἐξοµολογέω ‘declare’, ἐπαγγέλλοµαι ‘promise’, εὐαγγελίζω ‘proclaim [as good news]’, εὔχοµαι ‘pray’, µαρτύροµαι ‘testify’, προσεπαγγέλλοµαι ‘promise previously’, προευαγγελίζοµαι ‘proclaim [as good news] previously’, προµαρτύροµαι ‘foretell’, προσεύχοµαι ‘pray’, πυνθάνοµαι ‘ask’, συλλογίζοµαι ‘discuss’, and φθέγγοµαι ‘speak’. This discussion considers only the occurrences of ἀποκρίνοµαι with middle base forms. The occurrences of ἀποκρίνοµαι with passive base forms grammaticalize a different event.

DANOVE Conceptualization of Communication

17

affected Agent (pass.), secondary emphasis on the Content (ACE), and no suppression or change in the functionality of the Content. The translation addresses the internal affectedness by introducing for the omitted co-referential Experiencer a reflexive pronoun.13 In the following example, the πρός prepositional phrase is interpreted as a Locative complement that specifies the locale of the discussion or dispute. πρὸς ἀλλήλους γὰρ διελέχθησαν ἐν τῇ ὁδῷ τίς µείζων (Mark 9:34) For among themselves on the way they discussed with one another who is greater.

3.4 Usage #4: A{C}E {C = V} +or (Cmm. act. trans.) The seven verbs with Usage #4 grammaticalize oral communication (+or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Content (ACE), suppression of the Content {C}, and no change in the functionality of the Content.14 The Content is unrealized, and its semantic referent is supplied directly by the verb {C = V}. ὥστε ἐχθρὸς ὑµῶν γέγονα ἀληθεύων ὑµῖν; (Gal 4:16) So have I become your enemy speaking the truth to you? τακτῇ δὲ ἡµέρᾳ ὁ Ἡρῴδης . . . ἐδηµηγόρει πρὸς αὐτούς (Acts 12:21) On an appointed day Herod . . . was addressing a speech to them.

13. Of the two Greek means of indicating internal affectedness with Usages ACE +or, the omission of a required complement that is co-referential to the Agent and the use of passive base forms appears only in Mark 9:34; and the use of an active (Usage #1) or middle (Usage #2) verb form with a reflexive or reciprocal pronoun that is co-referential to the Agent appears on twenty-two occasions (διαλαλέω, Luke 6:11; διαλογίζοµαι, Mark 8:16; 11:31; Luke 20:14; ἐξοµολογέω, Jas 5:16; εὔχοµαι, Jas 5:16; κηρύσσω, 2 Cor 4:5; λαλέω, Luke 2:15; Acts 26:31; 1 Cor 14:28; λέγω, Mark 4:41; 10:26; 12:7; Luke 8:25; John 4:33; 7:35; 12:19; 16:17; 19:24; ὁµιλέω, Luke 24:14; and συλλογίζοµαι, Luke 4:36; 20:5). This indicates a strong preference for the latter means over the former means of indicating internal affectedness. 14. Usage #4: ἀληθεύω ‘speak the truth’, δηµηγορέω ‘address a speech’, ἐπιτάσσω ‘speak an order’, ἐπιτιµάω ‘speak a rebuke’, ἑτεροδιδασκαλέω ‘teach something different’, ὁµολογέω ‘speak a declaration’, and ὑπερεντυγχάνω ‘speak an intercession’.

18

Biblical and Ancient Greek Linguistics 7

3.5 Usage #5: A{C}E {C = V} +or (Cmm. mid. trans.) The four verbs with Usage #5 grammaticalize oral communication (+or) with an externally affected Agent (mid.), secondary emphasis on the Content (ACE), suppression of the Content {C}, and no change in the functionality of the Content.15 The Content is unrealized and its semantic referent is supplied directly by the verb {C = V}. The translations signal external affectedness by placing “[with affect]” after the subject and the suppression by introducing the Content. ἀποταξάµενος αὐτοῖς ἐξῆλθον εἰς Μακεδονίαν (2 Cor 2:13) [With affect] saying good-bye to them, I went forth into Macedonia. οὗ εἵνεκεν ἔχρισέν µε εὐαγγελίσασθαι πτωχοῖς (Luke 4:18) Because he anointed me to [with affect] proclaim good news to the poor.

3.6 Usage #6: ACE C→T +or (Cmm. act. ditr.) The nine verbs with Usage #6 grammaticalize oral communication (+or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Content (ACE), no suppression of the Content, and a change in the functionality of the Content to a Topic (C→T).16 ἀπήγγειλαν Ἰωάννῃ οἱ µαθηταὶ αὐτοῦ περὶ πάντων τούτων (Luke 7:18) [John’s] disciples reported to John about all these things. ἐλάλει περὶ αὐτοῦ πᾶσιν τοῖς προσδεχοµένοις λύτρωσιν Ἰερουσαλήµ (Luke 2:38) She was speaking about him to all awaiting the redemption of Jerusalem.

3.7 Usage #7: ACE C→T +or (Cmm. mid. ditr.) The six verbs with Usage #7 grammaticalize oral communication (+or) with an externally affected Agent (mid.), secondary emphasis on the Content (ACE), no suppression of the Content, and a change in the functionality of the Content to a Topic

15. Usage #5: ἀποτάσσοµαι ‘say good-bye’, ἐξοµολογέω ‘speak praise to/praise’, εὐαγγελίζω ‘proclaim good news’, and προσεύχοµαι ‘say a prayer /pray’. 16. Usage #6: ἀπαγγέλλω ‘report’, ἐντυγχάνω ‘appeal’, λαλέω ‘speak’, λέγω ‘say’, ὁµιλέω ‘converse’, ὁµολογέω ‘declare’, προκαταγγέλλω ‘announce previously’, προφητεύω ‘prophesy’, and συνοµιλέω ‘speak together’.

DANOVE Conceptualization of Communication

19

(C→T).17 The translations signal external affectedness by placing “[with affect]” after the subject. διηγήσαντο αὐτοῖς οἱ ἰδόντες . . . περὶ τῶν χοίρων (Mark 5:16) The ones seeing [it] [with affect] explained to them . . . about the pigs. προσεύχεσθε περὶ τῶν ἐπηρεαζόντων ὑµᾶς (Luke 6:28) [With affect] pray concerning those mistreating you.

4. Usages AEC +or This discussion considers the five observed usages of oral communication with secondary emphasis on the Experiencer. 4.1 Usage #8: AEC +or (Cmm. act. ditr.) The nine verbs with Usage #8 grammaticalize oral communication (+or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Experiencer (AEC), and no suppression or change in the functionality of the Content.18 ἐκεῖνος ὑµᾶς διδάξει πάντα (John 14:26) That one will teach you all things. ἐκέλευσέν τε τοὺς δυναµένους κολυµβᾶν ἀπορίψαντας πρώτους ἐπὶ τὴν γῆν ἐξιέναι (Acts 27:43) And he ordered those able to swim, jumping first, to go forth to the land.

4.2 Usage #9: Usage AEC +or (Cmm. mid. ditr.) The four verbs with Usage #9 grammaticalize oral communication (+or) with an externally affected Agent (mid.), secondary emphasis on the Experiencer (AEC), and no suppression or change in the functionality of the Content.19 The

17. Usage #7: διαβεβαιόοµαι ‘speak confidently’, διαλέγοµαι ‘discuss’, διηγέοµαι ‘explain’, ἐντέλλοµαι ‘speak commands’, προσαπειλέοµαι ‘speak threats’, and προσεύχοµαι ‘say a prayer/pray’. 18. Usage #8: αἰτέω ‘ask’, διδάσκω ‘teach’, διερωτάω ‘ask’, ἐπερωτάω ‘ask’, ἐρωτάω ‘ask’, εὐαγγελίζω ‘tell [as good news]’, κατηχέω ‘teach’, κελεύω ‘order’, and παρακαλέω ‘beg’. 19. Usage #9: αἰτέω ‘ask’, ἐξαιτέοµαι ‘ask’, εὐαγγελίζω ‘tell [as good news]’, and προαιτιάοµαι ‘accuse beforehand’.

20

Biblical and Ancient Greek Linguistics 7

translations signal external affectedness by placing “[with affect]” after the subject. οὗτος προσελθὼν τῷ Πιλάτῳ ᾐτήσατο τὸ σῶµα τοῦ Ἰησοῦ (Matt 27:58) Coming to Pilate, this one [with affect] asked [him] for the body of Jesus. προῃτιασάµεθα γὰρ Ἰουδαίους τε καὶ Ἕλληνας πάντας ὑφ᾿ ἁµαρτίαν εἶναι (Rom 3:9) For we [with affect] accused beforehand both Jews and Greeks that they all are under sin.

4.3 Usage #10: Usage AE{C} +or (Cmm. act. trans.) Only παρακαλέω ‘speak encouragement / encourage’ with Usage #10 grammaticalizes oral communication (+or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Experiencer (AEC), a suppressed Content {C}, and no change in the functionality of the Content. ὁ παρακαλῶν τοὺς ταπεινοὺς παρεκάλεσεν ἡµᾶς ὁ θεὸς ἐν τῇ παρουσίᾳ Τίτου (2 Cor 7:6) The God encouraging the humble encouraged us by the arrival of Titus.

4.4 Usage #11: Usage AE{C} +or (Cmm. mid. trans.) Only εὐαCελίζω ‘tell good news / evangelize’ with Usage #11 grammaticalizes oral communication (+or) with an externally affected Agent (mid.), secondary emphasis on the Experiencer (AEC), a suppressed Content {C}, and no change in the functionality of the Content. The translation addresses external affectedness by placing “[with affect]” after the subject. . . . συµβιβάζοντες ὅτι προσκέκληται ἡµᾶς ὁ θεὸς εὐαγγελίσασθαι αὐτούς (Acts 16:10) . . . being convinced that God has summoned us to [with affect] evangelize them.

4.5 Usage #12: Usage AEC C→T +or (Cmm. act. ditr.) The six verbs with Usage #12 grammaticalize oral communication (+or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the

DANOVE Conceptualization of Communication

21

Experiencer (AEC), no suppression of the Content, and a change in the functionality of the Content to a Topic (C→T).20 τὸ αὐτοῦ χρῖσµα διδάσκει ὑµᾶς περὶ πάντων (1 John 2:27) His anointing teaches you about all things. ὁ παρακαλῶν ἡµᾶς ἐπὶ πάσῃ τῇ θλίψει ἡµῶν . . . (2 Cor 1:4a) The one encouraging us concerning our every affliction . . .

5. Usages ACE –or This discussion considers the five observed usages of non-oral communication with secondary emphasis on the Content.21 5.1 Usage #13: ACE –or (Cmm. act. ditr.) The twenty-one verbs with Usage #13 grammaticalize non-oral communication (–or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Content (ACE), and no suppression or change in the functionality of the Content.22 Except for the feature –or, Usage #13 is identical to Usage #1. πάντα ἃ ἤκουσα παρὰ τοῦ πατρός µου ἐγνώρισα ὑµῖν (John 15:15) All the things, which I heard from my Father, I made known to you. δείξω σοι τὴν νύµφην τὴν γυναῖκα τοῦ ἀρνίου (Rev 21:9) I will show to you the bride, the wife of the Lamb.

20. Usage #12: ἀποστοµάζω ‘interrogate’, διδάσκω ‘teach’, ἐπερωτάω ‘ask’, ἐρωτάω ‘ask’, νουθετέω ‘warn’, and παρακαλέω ‘beg/encourage’. 21. This discussion of –or usages develops and corrects ideas first introduced in Danove, “Non-Spoken Communication,” 41–58. 22. Usage #13: ἀναδείκνυµι ‘make known’, ἀποδείκνυµι ‘make known’, ἀποκαλύπτω ‘reveal’, γνωρίζω ‘make known’, γράφω ‘write’, δείκνυµι ‘show’, δεικνύω ‘show’, δηλόω ‘make clear’, διανεύω ‘signal’, ἐµφανίζω ‘reveal/make known’, ἐννεύω ‘signal’, ἐπιδείκνυµι ‘show’, ἐπιστέλλω ‘write’, κατανεύω ‘signal’, µηνύω ‘make known’, νεύω ‘signal’, προγράφω ‘write previously’, σηµαίνω ‘indicate’, ὑποδείκνυµι ‘show’, φανερόω ‘make known’, and χρηµατίζω ‘warn’.

22

Biblical and Ancient Greek Linguistics 7

5.2 Usage #14: ACE –or (Cmm. mid. ditr.) Only ἐνδείκνυµαι (make known) with Usage #14 grammaticalizes non-oral communication (–or) with an externally affected Agent (mid.), secondary emphasis on the Content (ACE), and no suppression or change in the functionality of the Content. Except for the feature –or, Usage #14 is identical to Usage #2. The translation addresses external affectedness by placing “[with affect]” after the subject. . . . πᾶσαν ἐνδεικνυµένους πραΰτητα πρὸς πάντας ἀνθρώπους (Tit 3:2) . . . [community members] showing [with affect] every gentleness to all human beings.

5.3 Usage #15: A[C]E [C = A] –or (Cmm. pass. trans.) Three verbs admit to interpretation with Usage #15 as grammaticalizing non-oral communication (–or) with an internally affected Agent (pass.), secondary emphasis on the Content (ACE), and no suppression or change in the functionality of the Content, which is co-referential with the Agent [C = A] and never realized as a complement. All occurrences of the verbs that admit to interpretation with Usage #15 also admit to interpretation with the passivized form of Usage #13, which received description above.23 The polysemy arises because the contexts of these occurrences do not permit a determination of the referent and semantic function of the unrealized complement, which may receive interpretation either as the Content that is co-referential to the Agent (Usage #15) or as an unspecified Agent (Usage #13). Usage #15 has no parallel among verbs of oral communication. ἐν τῷ δευτέρῳ ἀνεγνωρίσθη Ἰωσὴφ τοῖς ἀδελφοῖς αὐτοῦ (Acts 7:13) #15 On the second [visit], Joseph [Agent] revealed himself [Content] to his brothers [Experiencer]. #13 On the second [visit], Joseph [Content] was revealed to his brothers [Experiencer] [by an unspecified Agent].

23. Ambiguous between Usage #15/Passivized Usage #13: ἀναγνωρίζω ‘make oneself known/be made known’; ἀποκαλύπτω ‘reveal oneself/be revealed’; and ἐµφανίζω ‘reveal oneself/be revealed’.

DANOVE Conceptualization of Communication

23

εἰσῆλθον εἰς τὴν ἁγίαν πόλιν καὶ ἐνεφανίσθησαν πολλοῖς (Matt 27:53) #15 They came into the holy city and they [Agent] revealed themselves [Content] to many [Experiencer]. #13 They came into the holy city and they [Content] were revealed to many [Experiencer] [by an unspecified Agent].

The absence of any +or passive usages with the Agent and Content co-referential is best explained in relation to the conceptualization of oral communication grammaticalized by most verbs and the constraint that the Agent and Content be coreferential. Among the seventy-three verbs with Usages #1 and #2, only six (διηγέοµαι, ἐξηγέοµαι, εὐαCελίζω, καταCέHω, κηρύσσω, and ὁµολογέω) appear with an animate Content on a total of twenty-one occasions.24 Except for the six noted verbs and other ἀCέHω and εὐαCελίζω compounds, verbs with Usages #1 and #2 do not appear with animate Content complements because they grammaticalize a conceptualization of oral communication that imposes a proportional relationship on the action of producing the Content and the Content that is produced. That is, the conceptualization is that, one quarter or one half of the way through communicating the Content, one quarter or one half of the Content has been communicated.25 The interpretation of a proportional production of the Content proves very convivial for the introduction of Content complements constituted by verbatim quotations, which may be viewed as a series of similar incremental units (words) built up over time, but excludes the introduction of animate Content complements, because living beings cannot be viewed as a series of similar incremental units. The resulting restriction of animate Content complements to six verbs and the lack of appropriate contexts to introduce the actions of describing (διηγέοµαι), explaining

24. Animate Content complements with Usages #1 and #2: διηγέοµαι (Acts 8:33); ἐξηγέοµαι (John 1:18); εὐαγγελίζω (Acts 5:42; 8:35; 11:20; Gal 1:16); καταγγέλλω (Acts 17:3; Phil 1:17, 18; Col 1:28); κηρύσσω (Acts 8:5; 9:20; 15:21; 19:13; 1 Cor 1:23; 15:12; 2 Cor 1:19; 4:5; 11:4); ὁµολογέω (1 John 4:3; 2 John 7). 25. This indicates that the Content complements of the sixty-seven verbs with Usages #1 and #2 have the characteristic “incremental theme” as developed in Dowty, “Proto-roles,” 574–619.

24

Biblical and Ancient Greek Linguistics 7

(ἐξηγέοµαι), proclaiming [as good news] (εὐαCελίζω), announcing (καταCέHω), proclaiming (κηρύσσω), and declaring (ὁµολογέω) oneself explains the absence of a passive usage of oral communication with the Agent and Content co-referential. 5.4 Usage #16: A{C}E {C = V} –or (Cmm. act. trans.) Only ἐπιστέHω ‘write a letter’ with Usage #16 grammaticalizes non-oral communication (–or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Content (ACE), suppression of the Content {C}, and no change in the functionality of the Content. The Content is unrealized, and its semantic referent is supplied directly by the verb (C = V). Except for the feature –or, Usage #16 is identical to Usage #4. περὶ δὲ τῶν πεπιστευκότων ἐθνῶν ἡµεῖς ἐπεστείλαµεν κρίναντες φυλάσσεσθαι αὐτοὺς τό τε εἰδωλόθυτον καὶ αἷµα καὶ πνικτὸν καὶ πορνείαν (Acts 21:25) Deciding, we wrote [a letter] concerning the Gentiles having come to believe that they keep themselves from what is offered to idols and blood and the strangled animal and sexual immorality.

5.5 Usage #17: ACE C→T –or (Cmm. act. ditr.) The two verbs with Usage #17 grammaticalize non-oral communication (–or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Content (ACE), no suppression of the Content, and a change in the functionality of the Content to a Topic (C→T).26 Except for the feature –or, Usage #17 is identical to Usage #6. περὶ δὲ τῆς φιλαδελφίας οὐ χρείαν ἔχετε γράφειν ὑµῖν (1 Thess 4:9) You do not have a need [for anyone] to write to you about the love of brothers and sisters. . . . περὶ οὗ γενοµένου µου εἰς Ἱεροσόλυµα ἐνεφάνισαν οἱ ἀρχιερεῖς καὶ οἱ πρεσβύτεροι τῶν Ἰουδαίων αἰτούµενοι κατ᾿ αὐτοῦ καταδίκην (Acts 25:15) [Paul] . . . about whom, when I arrived into Jerusalem, the chief priests and elders of the Jews informed [me], [with affect] asking [me] for a sentence against him.

26. Usage #17: γράφω ‘write’ and ἐµφανίζω ‘reveal’.

DANOVE Conceptualization of Communication

25

6. Usages AEC –or This discussion considers the four observed usages of non-oral communication with secondary emphasis on the Experiencer. 6.1 Usage #18: AEC –or (Cmm. act. ditr.) The four verbs with Usage #18 grammaticalize non-oral communication (–or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Experiencer (AEC), and no suppression or change in the functionality of the Content.27 Except for the feature –or, Usage #18 is identical to Usage #8. ὁ δὲ παράκλητος . . . ὑποµνήσει ὑµᾶς πάντα ἃ εἶπον ὑµῖν [ἐγώ] (John 14:26) But the Advocate . . . will remind you of all the things that I said to you. Κορνήλιος ἑκατοντάρχης . . . ἐχρηµατίσθη ὑπὸ ἀγγέλου ἁγίου µεταπέµψασθαί σε εἰς τὸν οἶκον αὐτοῦ καὶ ἀκοῦσαι ῥήµατα παρὰ σοῦ (Acts 10:22) Cornelius, a centurion . . . was warned by a holy angel to summon you to his house and to hear words from you.

6.2 Usage #19: AC[E] [E = A] –or (Cmm. pass. trans.) Only ἀναµιµνῄσκω ‘remind oneself of’ (= remember) with Usage #19 grammaticalizes non-oral communication (–or) with an internally affected Agent (pass.), secondary emphasis on the Experiencer (AEC) that is co-referential with the Agent [E = A] and consistently omitted [E], and no suppression of or change in the functionality of the Content. There is no comparable usage of oral communication. The translation of ἀναµιµνῄσκω is difficult because the English “remind” does not license noun phrase Content complements. As a consequence, ἀναµιµνῄσκω (along with other µιµνῄσκω compounds) does not admit freely to translation by the verb with a reflexive pronoun as do verbs with the previously considered passive usages. Thus this and following translations of Greek verbs of remembering use “remember.”

27. Usage #18: ἀναµιµνῄσκω ‘remind’, ἐπαναµιµνῄσκω ‘remind again’, ὑποµιµνῄσκω ‘remind’, and χρηµατίζω ‘warn’.

26

Biblical and Ancient Greek Linguistics 7 τὰ σπλάγχνα αὐτοῦ περισσοτέρως εἰς ὑµᾶς ἐστιν ἀναµιµνῃσκοµένου τὴν πάντων ὑµῶν ὑπακοήν, ὡς µετὰ φόβου καὶ τρόµου ἐδέξασθε αὐτόν (2 Cor 7:15) His affections are even more for you when he remembers the obedience of all of you, how with fear and trembling you received him.

6.3 Usage #20: AEC C→T –or (Cmm. act. ditr.) The three verbs with Usage #20 grammaticalize non-oral communication (–or) with no guidance concerning the affectedness of the Agent (act.), secondary emphasis on the Experiencer (AEC), no suppression of the Content, and a change in the functionality of the Content to a Topic (C→T).28 Except for the feature –or, Usage #20 is identical to Usage #12. In the first example, καρδία ‘heart’ is able to function as an animate Experiencer because it specifies a human being viewed from the perspective of the ability to reason.29 εἴ τις δοκεῖ θρησκὸς εἶναι µὴ χαλιναγωγῶν γλῶσσαν αὐτοῦ ἀλλὰ ἀπατῶν καρδίαν αὐτοῦ, τούτου µάταιος ἡ θρησκεία (Jas 1:26) If someone thinks that one is religious, not bridling one’s tongue but deceiving one’s heart [about one’s actions], this one’s religion is empty. πίστει χρηµατισθεὶς Νῶε περὶ τῶν µηδέπω βλεποµένων, εὐλαβηθεὶς κατεσκεύασεν κιβωτὸν εἰς σωτηρίαν τοῦ οἴκου αὐτοῦ (Heb 11:7) By faith Noah, being warned about things not yet seen [and] acting reverently, built an ark for the salvation of his household.

6.4 Usage #21: A[E]C [E = A] C→T –or (Cmm. pass. trans.) Only ὑποµιµνῄσκω ‘remind oneself of’ (= remember) with Usage #21 grammaticalizes non-oral communication (–or) with an internally affected Agent (pass.), secondary emphasis on the Experiencer (AEC) that is co-referential to the Agent [E = A] and never realized as a complement [E], no suppression of the Content, and a change in the functionality of the Content to a Topic (C→T). There is no comparable usage of oral communication.

28. Usage #20: ἀπατάω ‘deceive’, ὑποµιµνῄσκω ‘remind’, and χρηµατίζω ‘warn’. 29. Further discussion of the animate use of particular parts of the body to describe the human being from specific perspectives appears in Danove, “δίδωµι,” 27.

DANOVE Conceptualization of Communication

27

ὑπεµνήσθη ὁ Πέτρος τοῦ ῥήµατος τοῦ κυρίου ὡς εἶπεν αὐτῷ ὅτι πρὶν ἀλέκτορα φωνῆσαι σήµερον ἀπαρνήσῃ µε τρίς (Luke 22:61) Peter remembered the word of the Lord, how he said to him, “Before the cock crows today, you will deny me three times.”

7. The Feature Model The Feature Model identifies one usage as the “basic” usage with a particular configuration of features and describes all other usages by their deviations from that basic configuration. This discussion chooses as the basic usage the one grammaticalized most frequently by verbs. Among verbs of communication, the basic usage is Usage #1, which accommodates an absolute majority of NT occurrences of verbs that grammaticalize communication. The Feature Model places the five usage features in a row with the following abbreviations: Oral (orality), Emph (secondary emphasis), Supp (suppression), Funct (functionality), and Aff (affectedness). Beneath each feature and in bold type appears a description of the manner in which it is realized with Usage #1. This becomes the default description in relation to which all other usages of communication are described. Beneath this row appear the descriptions of the features of all other observed usages of communication, with the numerical designation for that usage on the far right after an equal sign. The presence of a dash marks no divergence from the default description, with divergences indicated by an abbreviated statement of the divergence. Oral +or – – – – – – – – – – –

Emph ACE – – – – – – AEC AEC AEC AEC AEC

Supp C – – {C} {C} – – – – {C} {C} –

Funct C=C – – – – C→T C→T – – – – C→T

Aff act mid pass – mid – mid _ mid – mid –

= = = = = = = = = = = = =

Usage #1 #2 #3 #4 #5 #6 #7 #8 #9 #10 #11 #12

28

Biblical and Ancient Greek Linguistics 7 –or –or –or –or –or –or –or –or –or

– – _ – _ AEC AEC AEC AEC

– – {C} – – – – –

– – – – C→T – – C→T C→T

– mid pass – – – pass. – pass

= = = = = = = = =

#13 #14 #15 #16 #17 #18 #19 #20 #21

8. Conclusion This article has described five features of the conceptualization of communication and analyzed twenty-one observed New Testament usages in accordance with these features. The discussions of usages has also considered the interpretation and translation of verbs with each usage and clarified elements of the conceptualization of communication. Bibliography Allan, Rutger J. The Middle Voice in Ancient Greek: A Study in Polysemy. ASCP 11. Amsterdam: Gieben, 2003. Allerton, David J. Valency and the English Verb. New York: Academic Press, 1982. Bakker, Egbert. “Voice, Aspect, and Aktionsart: Middle and Passive in Ancient Greek.” In Voice: Form and Function, edited by Barbara Fox and Paul J. Hopper, 23–47. Typological Studies in Language 27. Amsterdam: Benjamins, 1994. Danove, Paul L. “The αἰτέω/αἰτέοµαι Distinction in the New Testament: A Proposal.” Filología Neotestamentaria 25 (2012) 101–18. ———. A Grammatical and Exegetical Study of New Testament Verbs of Transference: A Case Frame Guide to Interpretation and Translation. LNTS 329; SNTG 13. London: T. & T. Clark, 2009. ———. “The Licensing Properties of New Testament Verbs of Non-Spoken Communication.” Filología Neotestamentaria 24 (2011) 41–58. ———. Linguistics and Exegesis in the Gospel of Mark: Applications of a Case Frame Analysis and Lexicon. JSNTSup 218; SNTG 10. Sheffield: Sheffield Academic, 2002.

DANOVE Conceptualization of Communication

29

———. “The Usages of δίδωµι in the Septuagint: Its Interpretation and Translation.” Bulletin of the International Organization for Septuagint and Cognate Studies 43 (2010) 23–40. Dowty, David R. “Thematic Proto-Roles and Argument Selection.” Language 67 (1991) 574–619. Fraser, Bruce, and John Robert Ross. “Idioms and Unspecified N[oun] P[hrase] Deletion.” Linguistic Inquiry 1 (1970) 264–65. Goddard, Cliff. Semantic Analysis: A Practical Introduction. Oxford Textbooks in Linguistics. Oxford: Oxford University Press, 1998. Goldberg, Adele E. “The Inherent Semantics of Argument Structure: The Case of the English Ditransitive Construction.” Cognitive Linguistics 3 (1992) 37–74. Hovav, Malka Rappaport, and Beth Levin. “The English Dative Alternation: The Case for Verb Sensitivity.” Journal of Linguistics 44 (2008) 129–67. Lehrer, Adrienne. “Checklist for Verbs of Speaking.” Acta Linguistica Hungarica 38 (1988) 143–61. Lyons, John. Introduction to Theoretical Linguistics. London: Cambridge University Press, 1969. Matthews, Peter H. Syntax. Cambridge Textbooks in Linguistics. Cambridge: Cambridge University Press, 1981. Mittwoch, Anita. “Idioms and Unspecified N[oun] P[hrase] Deletion.” Linguistic Inquiry 2 (1971) 255–59. Pesetsky, David. Zero Syntax: Experiencers and Cascades. Current Studies in Linguistics 27. Cambridge, MA: MIT Press, 1995. Pinker, Steven. Learnability and Cognition: The Acquisition of Argument Structure. Cambridge, MA: MIT Press, 1987. Rijksbaron, Albert. The Syntax and Semantics of the Verb in Classical Greek: An Introduction. 3rd ed. Amsterdam: Gieben, 2002. Saeed, John I. Semantics. Introducing Linguistics 2. Oxford: Blackwell, 1997. Sag, Ivan, and Jorge Hankamer. “Toward a Theory of Anaphoric Processing.” Linguistics and Philosophy 7 (1984) 325–45. Speas, Margaret. Phrase Structure in Natural Language. Studies in Natural Language and Linguistic Theory 21. Nordrecht: Kluwer, 1990.

[BAGL 7 (2018) 31–80]

CATEGORIES OF CONTRAST IN NEW TESTAMENT GREEK Nicholas P. Lunn Wycliffe Bible Translators, High Wycombe, Buckinghamshire, UK Abstract: The following aims to provide something lacking in the field of New Testament Greek studies, which is an overview of the various forms in which the logical relation of contrast may be realized in the surface structure of the language. Here seven distinct categories are described, illustrated, and differentiated, with regard to both their inherent relation and their respective connectors. Variations, where such exist, within each basic category are included, along with any sub-categories. A final section demonstrates the relevance of the presentation for the related tasks of translation and exegesis, offering analyses of several texts where there has been some confusion or misunderstanding with respect to the contrasting relation. (Article) Keywords: Concession, replacement, exception, connector, translation.

1. Introduction To my knowledge nothing approaching a full or systematic treatment of contrast in New Testament Greek has been published. Even more recent works, many of which adopt a more linguistic perspective, fail in this respect. Wallace (1996) presents us with just half a page that offers examples of just two different kinds of contrast.1 Levinsohn (2000), Long (2015), and Mathewson and Emig (2016) merely contain isolated examples in passing, while Köstenberger and his co-authors (2016) give us almost nothing of substance. In his discourse approach Runge

1.

Wallace, Greek Grammar, 671–72.

32

Biblical and Ancient Greek Linguistics 7

(2010) offers treatments of replacing and restrictive contrast,2 which are supplemented by his online articles (2013), along with that of Brannan (2008). These latter are among the few attempts to tackle the subject in some depth, if not in extent, though certain aspects of their analysis are questionable. In view of this evident lack, it is hoped that the present article will go some way to offering the reader of the New Testament a concise overview of the various categories of contrast present in its original language, and perhaps to advance our knowledge of how contrast operates in certain aspects of the same language. Contrast, in essence, is what most linguists would class as a “logical relation.”3 The term “relation” requires there to be, of course, more than one element. A logical relation is one that exists between two utterances that are immediately juxtaposed, or at least in close proximity. We describe this relation as “logical” when the connection is not simply temporal, that is, relating to time (perhaps involving, for example, connectors such as “before,” “after,” “while,” “until”). A connection that is logical includes such relations as reason, condition, result, purpose, as well as contrast. The presence of contrast, it should be stressed, is grounded on the actual semantic content of the two utterances in question.4 If the two oppose each other in total, or in part, then contrast is created. Other linguistic features might also be present, whether phonological, lexical, or syntactic, yet these do not establish the contrast, but serve to attract attention to it. In association with the contrasting relationship there may also be some or all of the following: A characteristic intonation pattern specific to the expression of contrast. This is readily discernible in spoken English and 2. Runge, Discourse Grammar, 83–100. 3. Cf. Larson, Meaning-Based Translation, 305. Some would prefer the term “semantic relation.” 4. Lambrecht, Information Structure, 291; Runge, “Where Does Contrast Come From?” In the fifth paragraph of the article, Runge states, “certain conjunctions like but can constrain you to read the linked elements as contrastive. However, these words do not create contrast that wasn’t already there, they simply amplify it.” Cf. Runge, Discourse Grammar, 28, 199.

LUNN Categories of Contrast

33

numerous other languages, but is obviously not detectable in languages existing only in written form, such as New Testament Greek, in which such prosodic variation is not represented orthographically. A connector, or conjoining word or words, which form a link between the two contrasting utterances. This is primarily the role of but in English, and of δέ and ἀ$ά, besides other terms, in the Greek of the New Testament. That these latter do not in themselves establish the contrasting relationship is demonstrable from the fact that δέ frequently appears in contexts which are plainly non-contrastive, while ἀ$ά, although more specifically associated with the contrastive relation, actually has a broader function in which contrast need not be an integral part (see below). Further to this, as shall be shown, instances occur in which a contrasting relation exists without there being any connector whatsoever. Finally, the presence of contrast may give rise to a particular word order in one or both of the two utterances. In general, we observe a movement to the front of the clause of the principal element, or elements, exhibiting the contrast. This element may then be said to be “fronted” or “preposed.” In New Testament Greek, it is mostly in evidence in the second of the two utterances, though it sometimes takes place in both. As with the connector, this ordering does not of itself create the contrast, but assists in underscoring its presence. Also, this same ordering can occur with other logical relations or discourse functions.5 There is, therefore, nothing inherently contrastive about such an ordering. Other specific syntactic structures (see below) may also have a similar function to particular word orders. With respect to what has just been stated, some significance in the order of sentence constituents is more readily identifiable if a basic word order for the language is identified. On this issue, I am in agreement with a number of recent scholars who advocate that New Testament Greek was basically a VerbSubject-Object (VSO) language, or at least a verb-initial

5.

For instance, parallel topic, or point of departure.

34

Biblical and Ancient Greek Linguistics 7

language.6 While this article works on that assumption, the fact is that this is not an essential tenet for the linguistic description of contrast that follows. It is sufficient for the present purpose to observe that certain clause constituents appear in a fronted position when a specific contrasting relation is evident. In this overview of contrasting relations, since it describes how the logical relation impacts the language, the discussion is one that is essentially linguistic. The basic linguistic model that undergirds the terminology and concepts is that of functional grammar, but especially the development of this particular grammar advanced by Knud Lambrecht, which is generally known as “information structure.”7 We are necessarily bound by the subject matter to incorporate technical vocabulary from this approach, but I have endeavored to keep this to a minimum as far as is possible and to provide explanations for the uninitiated, since some readers will be from a biblical or theological background rather than linguistic. So before listing the various categories of contrast, I will explain certain important linguistic terms that will form an important part of the discussion. 1.1 Constituent In this article, we prefer the designation “constituent” to include the basic functional elements of the sentence. Here we include items such as verb, subject, direct object, indirect object, prepositional phrase, and adverb. These are often more than a single word. In Greek a grammatical object may, for instance, consist of a whole phrase comprising an accusative noun governing a genitive noun, each with its appropriate definite article, as in τὴν ἀγάπην τοῦ θεοῦ, “the love of God” (John 5:42). Here the whole phrase, though containing four words, forms a single direct object constituent within its clause.

6. E.g., Levinsohn, Discourse Features, 16–17; Bailey, “Thetic Constructions,” 113; Long, Koine Greek Grammar, 66. 7. See Lambrecht, Information Structure.

LUNN Categories of Contrast

35

1.2 Topic In plain terms a “topic” is the entity that a proposition is about. This entity may be a person, a creature, a thing, or something abstract, in the singular or plural. It may be represented by a full noun or noun phrase, by a pronoun, or implicitly in the bare verbal form. In the overwhelming majority of instances, the topic coincides with the grammatical subject, though this is not necessarily the case. In Greek, some topics display the accusative case, as when joined to the modal particle δεῖ (e.g., John 3:7, Δεῖ ὑµᾶς γεννηθῆναι ἄνωθεν, “You must be born from above”), or perhaps the genitive, as in the genitive absolute construction (e.g., Luke 3:21, καὶ Ἰησοῦ βαπτισθέντος, “and when Jesus had been baptized”), or the dative (e.g., Acts 25:27, ἄλογον γάρ µοι δοκεῖ, “For it seems unreasonable to me”).8 1.3 Focus We use “focus” to denote the element of information in an utterance which is not presupposed. In most instances, this will be something that is unknown or unpredictable. However, in some circumstances, such as that of reiteration, the focused information may already be known, in which case attention will be drawn to it in a new way. Since it is not the purpose of this article to become overly technical, it may suffice to think of the focus element of an utterance in terms of what is traditionally termed the predicate. The focus often consists of more than one constituent. In a sentence of the form SVO, the VO together comprise what we call the “focal domain,” while S refers to the topic. In certain instances, a speaker or writer may choose to assign one constituent within the focal domain more prominence than the other(s). This is then called the “dominant focal element,” abbreviated as DFE.9

8. For a fuller description of topic, see Lambrecht, Information Structure, 117–84. 9. For this phrase see, for example, Levinsohn, “Adverbial Participial Clauses,” 8. For a fuller description of focus, see Lambrecht, Information Structure, 206–86.

36

Biblical and Ancient Greek Linguistics 7

1.4 Marked An utterance may be either “unmarked” or “marked.” The former means that there is nothing in what is expressed that requires special treatment of any particular aspect of it. Such will generally consist of a topic and a focal domain, both uttered in an unremarkable way. Circumstances sometimes pertain, however, when the speaker or author wishes to assign a more than ordinary role to a particular constituent, and contrast is one such circumstance, though there are numerous others. Often the method of marking is purely phonological, that is to say, produced by the voice. So, with regard to English, in a contrastive sentence like Jack went to town, but Jill stayed at home, the intonation borne by Jill in this environment will be noticeably distinct from the simple statement Jill stayed at home. In this latter case the subject constituent Jill, functioning as an independent topic, bears no special tone, and remains unmarked. In the former example, Jill functions as a contrastive topic and is therefore marked by a specific rising and falling intonation pattern. Phono-logical marking such as this is common, if not universal, in spoken languages, but it offers no help with unspoken languages like New Testament Greek. In this latter case, markedness may be indicated by word order, where the marked constituent is moved to the initial position within the clause. Besides word order, markedness can be expressed through the presence of a semantically redundant independent pronoun or by special syntactic structures, such as extraposition (sometimes termed left- or right-dislocation), an example of which will be examined presently. 1.5 Asyndeton This word is taken from the Greek adjective ἀσύνδετον ‘unconnected’. As a linguistic term, it simply indicates that two clauses or phrases are placed side by side without the presence of any conjunction or connecting particle. 1.6 Ellipsis By this noun, and its cognate verb elide, we mean the omission of certain words on account of the fact that their meaning is

LUNN Categories of Contrast

37

already presupposed. In answer to the question, “Do you like X?” the answer can simply be “I do.” Here the main verb like and its object have been elided. Since two statements in contrast frequently have some elements in common, there is the opportunity for certain elements to be elided in the second of the two. An alternative term for ellipsis is “gapping.” 2. Categories of Contrast The contrasting relation can take on a surprising number of forms, existing as it does at both sentence level and the larger discourse level. From a linguistic perspective, the former of these two is more precisely definable, since the presence of contrasting elements is confined to a narrow verbal locality. This makes the nature of the contrast and the manner of its linguistic expression more readily accessible to analysis. Contrariwise, at the wider discourse level, the contrastive relation is more diffuse, being more thematic in nature, and does not therefore lend itself so readily to precise linguistic analysis in the same way as contrast at sentence level. For this reason, and due to the limitations of space, the scope of this article will be confined to sentence-level relations.10 From this, I should stress, the reader is by no means to understand that the discourse-level contrasting relation is of lesser importance. This is a subject matter deserving of its own analysis. Here we will offer a description of the following seven categories of sentence-level contrast in New Testament Greek: (1) Contrasting Focus; (2) Contrasting Topic; (3) Thetic Contrast; (4) Concessive Contrast; (5) Replacing Contrast; (6) Expanding Contrast; (7) Exceptive Contrast.

10. The findings of this article are based upon approximately 1,100 instances of sentence-level contrast within the Greek New Testament.

38

Biblical and Ancient Greek Linguistics 7

As those readers not so linguistically inclined may be unfamiliar with these designations, in order that they may appreciate at the outset what is intended by these, each of the above may be illustrated by the following concise phrases in English: (1) “A does X but A doesn’t do Y”; (2) “A does X but B does Y”; (3) “but there is/was Y”; (4) “but though X may be the case . . .”; (5) “not X but Y”; (6) “not only X but also Y”; (7) “no X but Y.”

While English might employ the connective but in all these categories, though other terms are equally possible, the situation in Greek is far different. As shall be evinced, several distinct connective items are involved, plus the frequent significant orderings of clause constituents. In the above list, some items may be grouped together. Categories 1–3 may all be classed as “simple” contrast. Here the manner of relationship is overtly antithetical, which is to say that some element within the secondary clause consists of the opposite of or is at variance with something in the primary clause. Categories 5–7 also all bear a resemblance to one another. These three may be termed “negative” contrast, since in each case the presence of a negating particle is mandatory. Some linguists prefer the description “corrective” contrast, since the effect in all three is to remove some aspect of the initial statement, some untrue or unwanted information, and then offer the correct data in what follows. This leaves the fourth category, that of concession, which, as will become evident, is semantically quite distinct from the items in the other two groupings. This is a large group which could, in a lengthier treatment, have readily been broken down into further subcategories. For the sake of brevity, however, these will be dealt with together. It is not the sole purpose of this article to be purely descriptive from a linguistic point of view. Rather the intention is to provide the linguistic underpinnings necessary for precision in translation and robustness in exegesis. To that end, the final

LUNN Categories of Contrast

39

section in this paper provides instances of application to specific New Testament texts where confusion regarding the nature of the manner of contrast has led to inaccuracies in understanding. 3. Contrasting Focus This category of contrast is when the utterance concerns actions or states predicated to a single topic. This means that the contrastive relation adheres wholly to constituents within the focal domain. So in the English sentence, I loved Jacob, but I hated Esau (Rom 9:13), there is one topic, expressed by the first person pronoun I. The first focal domain, loved Jacob, consists of a verb and direct object, as does the second focal domain, hated Esau. Here there is an obvious contrasting relation between the verbs hated and loved, as well as the expressed contrast between the proper noun objects Jacob and Esau. So the two clauses share an identical S constituent, while the VO in each stands in contrast to one another. Not all contrasting focus utterances are so neat or concise. The New Testament contains numerous instances of this particular category. Here we give several examples: (3a) καὶ διακαθαριεῖ τὴν ἅλωνα αὐτοῦ καὶ συνάξει τὸν σῖτον αὐτοῦ εἰς τὴν ἀποθήκην, τὸ δὲ ἄχυρον κατακαύσει πυρὶ ἀσβέστῳ (Matt 3:12 = Luke 3:17) And he will thoroughly clear his threshing floor and will gather the wheat into his barn, but he will burn the chaff with unquenchable fire. (3b) οὐ δύναται ὁ κόσµος µισεῖν ὑµᾶς, ἐµὲ δὲ µισεῖ (John 7:7) The world cannot hate you, but it hates me. (3c) ἔπειτα µετὰ ἔτη τρία ἀνῆλθον εἰς Ἱεροσόλυµα ἱστορῆσαι Κηφᾶν καὶ ἐπέµεινα πρὸς αὐτὸν ἡµέρας δεκαπέντε, ἕτερον δὲ τῶν ἀποστόλων οὐκ εἶδον εἰ µὴ Ἰάκωβον τὸν ἀδελφὸν τοῦ κυρίου. (Gal 1:18–19) Then after three years I went up to Jerusalem to become acquainted with Cephas, and stayed with him fifteen days; but I did not see any other of the apostles except James, the Lord’s brother. (3d) . . . ἔχοντες µόρφωσιν εὐσεβείας τὴν δὲ δύναµιν αὐτῆς ἠρνηµένοι (2 Tim 3:5) . . . having a form of godliness, but denying its power.

40

Biblical and Ancient Greek Linguistics 7

The contrasting relations, at the logical level, are apparent in each instance. So, taking example (3a), the sole topic throughout is the one to come who is greater than John the Baptist. In the act of threshing, here speaking figuratively, he treats the two products, wheat and chaff, differently; the one he gathers, the other he burns. So the contrast exists between the two verbs and their two direct objects, as in the Jacob/Esau example above. Looking now at the configurations in Greek, we observe the same features in all these utterances. First, there is the usage of the connective particle δέ, which appears postpositively according to the rule. As emphasized earlier, it is not the presence of this connector that creates the contrast, which lies inherently within the logical relations. In many instances, if not the majority, this particle appears in non-contrastive contexts as what Runge terms a “development marker.”11 It is often the case, however, that δέ is employed to connect two contrastive statements. The second feature we note is that in the second, contrastive clause, within the focal domain one constituent has been fronted. In each of these examples that constituent is the direct object. This then is an instance of a DFE within the broader focal domain. The constituent in question may therefore be said to be marked. The effect of this marking is to assign the constituent greater prominence than the remainder of the focal domain. The whole functions as focus, but within than focus, this particular element is pushed further to the fore. This is most evidently seen in (3c), where Paul states that he “did not see” certain people, but as it is of great import for his purposes to note that those people were the “other apostles,” he places this phrase first in the clause. The above are typical expressions of the contrasting focus category. Of course, there is no reason why the fronted constituent in the second clause should be limited to the direct object in the accusative case. In other instances, it is found to be the indirect or dative object (e.g., John 10:5), a prepositional

11. Runge, Discourse Grammar, 28. He later defines this particle as “a coordinating conjunction like καί, but it includes the added constraint of signaling a new development” (31).

LUNN Categories of Contrast

41

phrase (e.g., Acts 27:26), or an adverb or adverbial phrase (e.g., 1 Pet 2:10). When the focal domain of the contrastive clause consists merely of a verb, there is obviously no fronted constituent from any of the foregoing categories. Rather, in this case the verb itself occurs in the clause-initial position, followed by the postpositive δέ. This we see, for example, in Acts 5:23, ἀνοίξαντες δέ (“but having opened up”). While the examples given, exhibiting a fronted constituent in the second clause, reflect the most common realization of this category, there is some diversity of form. This should occasion no surprise since the use of the employment of a DFE is itself a matter of choice. The fact remains that if an object or other constituent is not moved to the front but occurs later in the clause, it still remains in focus. The difference is that it is not elevated to the position of marked focus, at least not as far as the word order is concerned. There is still, however, a real possibility that this same constituent may have carried phonological marking. So the variation evidenced with regard to the DFE is that on a comparatively small number of occasions the contrasting clause will not front any non-verbal constituent (e.g., Heb 6:8, ἐκφέρουσα δὲ ἀκάνθας καὶ τριβόλους, “but bearing thorns and thistles”), while on others, both the principal clause and the contrasting clause will equally display contrastive fronted DFEs (e.g., 2 Cor 12:5: ὑπὲρ τοῦ τοιούτου καυχήσοµαι, ὑπὲρ δὲ ἐµαυτοῦ οὐ καυχήσοµαι ‘On behalf of such a man I will boast, but on my own behalf I will not boast’). With this particular category of contrast ellipsis is possible, though not frequent: (3e) εἰ κακῶς ἐλάλησα, µαρτύρησον περὶ τοῦ κακοῦ· εἰ δὲ καλῶς, τί µε δέρεις; (John 18:23) If I have spoken wrongly, testify to the wrong, but if well, why do you strike me? (3f) καὶ ἀνεθέµην αὐτοῖς τὸ εὐαγγέλιον ὃ κηρύσσω ἐν τοῖς ἔθνεσιν, κατ᾽ ἰδίαν δὲ τοῖς δοκοῦσιν (Gal 2:2) And I set before them the gospel which I preach among the Gentiles, but in private to those who were of reputation.

In (3e) we find that the verb has been elided in the contrasting statement, in which the presence of ἐλάλησα (“I have spoken”) is

42

Biblical and Ancient Greek Linguistics 7

to be inferred. Similarly, in (3f) the whole of the verbal clause of the first proposition is implicit in the second. Hence several English versions make this explicit by adding the words to this verse “I did so” (NASB) or “I did this” (NIV).12 Asyndeton, the lack of any connector, is rare with this category. The following was the only evident occurrence: (3g) ἄλλους ἔσωσεν, ἑαυτὸν οὐ δύναται σῶσαι (Matt 27:42 = Mark 15:31) He saved others, [but] he cannot save himself.

Here but is present in NIV, NLT, NET, GNT, CEV, and in some less common versions.13 As with the previous example, the contrasting relation is evident and again the object constituent, ἑαυτόν ‘himself,’ is fronted, as is ἄ$ους ‘others’ in the first clause. Lastly, we point out that although in the vast majority of instances the two clauses showing this logical relation are immediately adjacent, it can be the case that some intervening material separates the two (e.g., Heb 9:25–26). The distance between them, however, is not extensive, otherwise the reader would have difficulty in perceiving the logical connection. This question of distance, of course, applies to all categories of contrast.14

12. Other examples of ellipsis in contrasting focus clauses include Eph 5:8 and Heb 4:15. 13. Such as the New Century Version, International Standard Version, and the Holman Christian Standard Bible. 14. Further instances of contrasting focus include: Matt 7:3; 13:32; 16:26; 21:29; 23:28; 26:11; 27:26; Mark 4:34; 14:7; Luke 6:41; 8:10; 9:32; 11:34; 12:56; 18:4; 21:23, 37; 23:25; 24:24; John 9:29; 12:8; 13:7, 36; 15:15, 22, 24; 16:7, 21; 19:33; 20:17; 21:18; Acts 2:34; 5:39; 9:7; 12:9, 14; 22:9; 23:29; 25:19; 27:39; Rom 2:25; 4:19–20; 7:25; 11:30; 1 Cor 7:25, 28; 13:1, 2, 3; 15:51; 2 Cor 6:12; 10:1, 15; 12:6; Eph 5:8; Phil 1:28; 3:1, 12, 13; 4:10; Col 1:26; Phlm 11; Heb 4:13; 6:8; 10:5; 12:13; Jas 2:11, 14; 3:8; 4:6; 1 Pet 1:20; 2:10, 23; 3:18; 1 John 1:7.

LUNN Categories of Contrast

43

4. Contrasting Topic On many occasions, a contrasting relation is formed by the opposing or diverse actions or states of distinct entities. So, unlike the previous category we are here dealing with two or more topics. These can be two persons, objects, or abstract ideas, or two groups of these. Identical to the foregoing category, it should be pointed out, is the fact that there are likewise two focal domains within this relation. As a simple illustration, we may take Isaac loved Esau, but Rebekah loved Jacob. The two contrasting topics are Isaac and Rebekah. The contrasting foci are loved Esau and loved Jacob. While the verb remains unchanged in both clauses, the direct object is equally part of the focal domain, and so the different objects, Esau on the one hand and Jacob on the other, establish the contrast between the two focal domains. So, although this category also involves contrasting focus, its principal distinction from the foregoing section is that the foci are also predicated on contrasting topics, hence the designation. Instances of this manner of contrast abound. A number of examples will serve to illustrate its principal characteristics: (4a) αἱ ἀλώπεκες φωλεοὺς ἔχουσιν καὶ τὰ πετεινὰ τοῦ οὐρανοῦ κατασκηνώσεις, ὁ δὲ υἱὸς τοῦ ἀνθρώπου οὐκ ἔχει ποῦ τὴν κεφαλὴν κλίνῃ (Matt 8:20) Foxes have holes, and birds of the air have nests, but the Son of Man has nowhere to lay his head. (4b) καὶ οὐκ ᾔδει πόθεν ἐστίν, οἱ δὲ διάκονοι ᾔδεισαν οἱ ἠντληκότες τὸ ὕδωρ (John 2:9) And he did not know where it came from, but the servants who had drawn the water knew. (4c) ἐκεῖνον δεῖ αὐξάνειν, ἐµὲ δὲ ἐλαττοῦσθαι (John 3:30) He must increase, but I decrease. (4d) γινέσθω δὲ ὁ θεὸς ἀληθής, πᾶς δὲ ἄνθρωπος ψεύστης (Rom 3:4) Let God be true, but every man a liar. (4e) δουλεύει γὰρ µετὰ τῶν τέκνων αὐτῆς. ἡ δὲ ἄνω Ἰερουσαλὴµ ἐλευθέρα ἐστίν (Gal 4:25–26) For she is in bondage with her children, but the Jerusalem above is free.

44

Biblical and Ancient Greek Linguistics 7 (4f) ἡ γὰρ σωµατικὴ γυµνασία πρὸς ὀλίγον ἐστὶν ὠφέλιµος, ἡ δὲ εὐσέβεια πρὸς πάντα ὠφέλιµός ἐστιν (1 Tim 4:8) For physical exercise profits a little, but godliness is profitable for all things.

Each of the above texts displays the same invariable feature of this category of contrast, and that is the fronted position of the topic constituent in the second clause. The topic is therefore being marked by this placement. No doubt the word order marking serves to highlight the fact that the action or state of the second topic is going to be different from that of the first, and it is equally certain that in speech the preposed topic would have been assigned its own particular intonation, again indicative of the fact that a contrast is to follow. However, as noted in the introduction, neither of these two features itself produces the contrast, but they rather accompany it and highlight its presence. The contrast, therefore, lies essentially in the logical relation between the two topics and their corresponding foci. It is often the case that the topic in the first clause of the contrasting pair is likewise fronted, as in (4a), (4c), and (4f).15 This is, however, an optional placement, unlike the obligatory positioning of the topic at the head of the second clause. We also observe the presence of the connecting particle δέ in all the examples. Contrasting topic therefore shares the same connector as contrasting focus, which is due no doubt to the fact that there is some overlap between them in that the focal domain of the two clauses stands in a contrasting semantic relationship in both categories. The distinction only lies in the number of topics. Another feature exhibited above is that of ellipsis, or gapping, in the second clause. This is evidenced in examples (4c) and (4d). In the latter of these the contrasting clause is verbless. The sense of the verb of the primary clause is to be taken as implicit in the secondary. In (4c) it is the modal of obligation, δεῖ ‘must,’ that has been gapped, yet the pronoun ἐµέ, indicating the topic, remains accusative nonetheless. Ellipsis occurs with reasonable frequency in this category.

15. See Bailey, “Thetic Constructions,” 29 n. 53.

LUNN Categories of Contrast

45

Further examples are now listed in which there is no nominal subject in the second clause. Here there is a significant development: (4g) καὶ ἤγαγον αὐτὸν ἕως ὀφρύος τοῦ ὄρους ἐφ᾽ οὗ ἡ πόλις ᾠκοδόµητο αὐτῶν ὥστε κατακρηµνίσαι αὐτόν· αὐτὸς δὲ διελθὼν διὰ µέσου αὐτῶν ἐπορεύετο (Luke 4:29–30) And they led him to the brow of the hill on which their city had been built, in order to throw him down the cliff. But he, passing through their midst, went on his way. (4h) . . . ὅτι οἶδα πόθεν ἦλθον καὶ ποῦ ὑπάγω· ὑµεῖς δὲ οὐκ οἴδατε πόθεν ἔρχοµαι ἢ ποῦ ὑπάγω (John 8:14) . . . because I know where I came from and where I am going, but you do not know where I come from and where I am going. (4i) ἐνόµιζεν δὲ συνιέναι τοὺς ἀδελφοὺς αὐτοῦ ὅτι ὁ θεὸς διὰ χειρὸς αὐτοῦ δίδωσιν σωτηρίαν αὐτοῖς· οἱ δὲ οὐ συνῆκαν (Acts 7:25) He supposed that his kinsmen would understand that God was granting them deliverance by his hand, but they did not understand.

In each of the foregoing utterances the verb in the contrasting clause requires no overt subject phrase since the bare verbal form, containing as it does the articulation of grammatical number and person, is of itself adequate to indicate the participant intended. In these examples, we observe the presence of an independent pronoun in (4g) and (4h), and the definite article functioning pronominally in (4i). Clearly none of these are actually required semantically, since the subject of the verb is unambiguous in each instance. The presence of the distinct pronominal form therefore has another function. Since each of these forms occupies the clause-initial position in the second, contrasting clause, we argue that its presence can readily be explained in terms of markedness. As with the preposed subject noun phrases in examples (4a) to (4f), these pronominal forms in the same position are placed there expressly to highlight the presence of a marked topic, which in these particular contexts stands in a contrasting relationship to that of the preceding clause. This use of a fronted pronoun, when no overt nominal subject is present, is a consistent feature of this category. Additional confirmation that we are looking at a marked word order in the foregoing fronted constituents is to be found in other

46

Biblical and Ancient Greek Linguistics 7

contrasting topic sentences that show a specifically marked syntax at the commencement of the second clause. This syntactic structure is variously designated by linguists as extraposition or left-dislocation. In classical grammatical terminology, it was referred to as a pendens (“hanging”) construction. Here is one instance: (4j) διὰ τὸ πληθυνθῆναι τὴν ἀνοµίαν ψυγήσεται ἡ ἀγάπη τῶν πολλῶν. ὁ δὲ ὑποµείνας εἰς τέλος οὗτος σωθήσεται (Matt 24:12–13) Because of the increase of lawlessness the love of many will grow cold, but he who endures to the end, this one will be saved.

From the logical relation, it is evident that the connection is one of contrast.16 The one who perseveres in faith to the end is contrasted with the many whose love grows cold. But here, instead of a mere fronted subject phrase, we have an instance of a left-dislocated phrase, ὁ δὲ ὑποµείνας εἰς τέλος. This stands outside the boundaries of the grammar of the following selfcontained clause, οὗτος σωθήσεται, which itself consists of a subject and verb.17 ὁ δὲ ὑποµείνας εἰς τέλος may therefore be said to be “hanging” or “extraposed” (meaning ‘placed outside’) the principal clause. Then within this latter clause the participant indicated by the extraposed phrase is referred to retrospectively by means of a resumptive pronoun, here οὗτος. This is a specific syntactic structure existing in many languages, including Greek and Hebrew, and in each instance, it expresses a form of markedness, which in this particular case is a marked topic. So, in this category of contrast we have observed three distinct means by which the contrasting topic is rendered marked. Either there will be a fronted subject phrase, a semantically redundant independent pronominal form, or an extraposed phrase before the actual clause, which contains a resumptive pronoun. Asyndeton is more common in the case of contrasting topic than with contrasting focus:

16. All the major English translations have but as the conjunction. 17. Many English versions, it should be noted, completely ignore the Greek syntax.

LUNN Categories of Contrast

47

(4k) Πᾶς ἄνθρωπος πρῶτον τὸν καλὸν οἶνον τίθησιν καὶ ὅταν µεθυσθῶσιν τὸν ἐλάσσω· σὺ τετήρηκας τὸν καλὸν οἶνον ἕως ἄρτι (John 2:10) Everyone serves the good wine first, and then the inferior wine when the guests have become drunk, [but] you have kept the good wine until now.

It is noteworthy that this absence of any connector particularly occurs in the writings attributed to John (cf. also John 1:17; 6:58; 8:35; 14:17; 1 John 2:23; 3:7–8; 4:8; 5:12; 2 John 9; 3 John 11). Finally, in this section, we note the occurrence of variant readings in the textual tradition. Certain contrastive clauses appear with δέ in some manuscripts, and with nothing in others (i.e. they exhibit asyndeton). Such is the case, for example, with the latter part of John 9:16. Here the Byzantine text has simply ἄ$οι ἔλεγον, whereas the other textual families display a mixture of ἄ$οι ἔλεγον and ἄ$οι δὲ ἔλεγον. John 14:17 shows the opposite trend. Here the Byzantine manuscripts have ὑµεῖς δὲ γινώσκετε αὐτό but others ὑµεῖς γινώσκετε αὐτό.18 5. Thetic Contrast An infrequent contrastive structure is that termed thetic contrast.19 By this term we denote a structure the purpose of which is to introduce or present something new within a discourse. This may be the presentation of new participants or new themes. A typical thetic construction involves an existential verb such as to be, usually in the sense of ‘there is/was.’ One

18. Further instances of contrasting topic include: Matt 8:12; 15:20; 18:6; 22:5; 24:22; 25:3–4, 46; 26:56; Mark 1:8; 3:4, 29; 4:11; 7:6; 8:35; 9:32; 10:40; 11:17; 13:31; 14:38; Luke 5:33; 6:7–8, 40; 7:44, 45, 46; 8:56; 9:58; 10:29; 11:39; 12:48; 18:34; 20:10; 21:33; 22:56; 23:41; John 2:23–24; 3:18; 7:6; 8:59; 9:16, 28; 10:6; 11:13; 14:10, 19; 17:25; 20:11; Acts 1:5; 12:5, 15; 13:13–14, 50–51; 15:39–40; 17:32; 19:15; 23:8; 27:42–43; 28:26; Rom 4:4–5; 5:16; 7:18; 8:5, 10; 11:7; 12:4; 1 Cor 2:15; 7:34; 8:1; 13:8; 14:22; 15:51; 2 Cor 3:6; 4:12; 7:10; 10:10; Gal 3:20; 5:22; 6:8; Phil 1:15, 24; Col 2:17; 1 Thess 5:4; 2 Thess 2:13; 1 Tim 2:14; 5:5–6; 6:9; 2 Tim 3:10, 13; Tit 2:1; 3:5; Heb 3:5–6; 7:28; 10:12; Jas 1:9–10; 1 Pet 2:25; 2 Pet 3:6–7; 1 John 2:17; Jude 9–10, 19–20; Rev 21:7–8. 19. Derived from the Greek adjective θετικός, meaning something “placed” or “set forth.”

48

Biblical and Ancient Greek Linguistics 7

non-contrastive example of this is John 1:6, Ἐγένετο ἄνθρωπος ἀπεσταλµένος παρὰ θεοῦ, ὄνοµα αὐτῷ Ἰωάννης ‘There was a man sent from God, whose name was John.’ Thetic contrast occurs in that very narrow context in which such a presentation is made of entities that bear a contrasting relation to other entities previously mentioned. This specificity means that the number of occurrences in the New Testament is extremely small. With respect to its form and content, thetic contrast stands alongside contrasting focus and contrasting topic in that its manner of contrasting relation, like theirs, is deemed “simple.” The connector is therefore likewise δέ, the usual particle occurring in a simple contrasting relation, which is not so in the remaining categories. Unlike categories 1 and 2, however, a thetic construction does not itself possess a topic, but presents something which may then become a topic in subsequent utterances. This is the distinction between presentation (of a previously inactive entity) and commenting (on an active topic). The focus-presupposition distribution of each is altogether distinct. The essential character of presentation is shown by the placement of an existential verb in the clause-initial position, which is then followed by the postpositive particle δέ. In such circumstances, it is usual to consider the verb as possessing what linguists sometimes refer to as an existential “dummy” subject,20 which is to say, it would appear in translation as “there is,” “there was,” etc. We observe the following instances of this manner of proposition in a contrasting relationship: (5a) ἔστιν δὲ συνήθεια ὑµῖν ἵνα ἕνα ἀπολύσω ὑµῖν ἐν τῷ πάσχα (John 18:39) But there is a custom of yours that I release someone for you at the Passover. (5b) ἔστιν δὲ καὶ ἄλλα πολλὰ ἃ ἐποίησεν ὁ Ἰησοῦς (John 21:25) But there are also many other things that Jesus did.

20. E.g., Bergs and Brinton, English Historical Linguistics, 1361.

LUNN Categories of Contrast

49

(5c) ἐγένοντο δὲ καὶ ψευδοπροφῆται ἐν τῷ λαῷ, ὡς καὶ ἐν ὑµῖν ἔσονται ψευδοδιδάσκαλοι (2 Pet 2:1) But there were also false prophets among the people, just as there will be false teachers among you.

Text (5a) relates the words of Pilate in which he introduces into the situation the matter of the custom observed at Passover time. This stands in contrast,21 albeit loosely, to the preceding attempts of Christ’s accusers to have him condemned. In the sentence coming before example (5b) the author has just declared, “This is the disciple who is testifying to these things and has written them,” a reference to the deeds of Jesus recorded in his Gospel. And then in the next verse he introduces those many other things not written in his book. There is therefore the potential for a contrasting relation between these two sets of things. While (5b) is not always translated as a contrast, since the structure fits exactly with this present category it is perhaps best taken as contrastive (as RSV and NRSV).22 In (5c) the writer has made mention in the immediately preceding verses of the prophets of the Old Testament, who of course to his mind uttered genuine prophecies. But then he further introduces, by way of contrast, the false prophets of the same era, who also claimed to speak from God. 6. Concessive Contrast In coming to this category, often overlooked in modern treatments, we are moving away from δέ as the most widely used connector. Here the principal conjoining word is ἀ$ά. The basic function of ἀ$ά is strongly discontinuous,23 that is to say it marks a break in the flow of thought. This particular marker of discontinuity is entirely appropriate for the manners of contrast

21. The connector is ubiquitously translated as “But” in the English translations of this verse. 22. Cf. Bailey, “Thetic Constructions,” 161, who is also of the view that the verse “evokes a feeling of contrast.” 23. Brannan (“The Discourse Function of ἀλλά,” 7) states that “ἀλλά indicates a high degree of discontinuity.”

50

Biblical and Ancient Greek Linguistics 7

that follow, though it can also occur in non-contrastive contexts.24 Up till now the contrasting relations discussed have involved simple oppositions or different potential options, such as “A didn’t do X, but he did Y,” and “A did X, but B did Y.” Rather than merely different or opposite actions, however, we are here looking at actions that in some way do not fit well together, where the second is not the expected outcome (concession), or where the second corrects or excludes the possibility of the first (replacement). By concessive contrast25 we mean sentences in which the connecting idea is one of “but though that is the case,” or “but despite the fact that is so.” A typical example, to quote a biblical proverb, would be, “Locusts have no king, but they all advance in ranks” (Prov 30:27). In its English version, the conjunction may remain as but (NLT, NET, CEV, NCV), as in the previously discussed categories, but it could reasonably be rendered as yet (ESV, NRSV, NIV, NJB). So here the contrast lies in the lack of leadership among locusts expressed in the initial clause and the orderliness presented by the second. For our purposes, we may note that two basic kinds of concession exist. Between these an important distinction exists that is, on occasion, actually realized lexically. There is concession in which the contents of the first proposition are acknowledged, and then in some way set aside, invalidated, or discounted in the second. This can happen in a number of ways. There are four possible configurations: positive-negative, negative-positive, positive-more positive, negative-more negative. It is crucial to grasp therefore that the discounting of an initial positive element may not be because it is simply being negated but may be because it is insufficiently positive, and vice versa. Moreover, the discounting is not absolute, but only for

24. A good example of non-contrastive ἀλλά is found in Acts 10:19–20: “Behold, three men are looking for you. But [Ἀλλὰ] get up, go down, and go with them.” Here the word seems to indicate a localised switch in topic, from the three men to Peter. 25. This is the designation employed in the detailed treatment of Rudolph, Contrast, 385.

LUNN Categories of Contrast

51

purposes relative to the second half of the utterance. For simplicity of reference, we shall label this as A-D (Acknowledged-Discounted). Sec-ondly, there is concession in which the second proposition does not contain the expected outcome of the first. This we shall label C-E (ContraExpectation).26 Many New Testament examples can be found of both the foregoing types, such as: (6a) Αββα ὁ πατήρ, πάντα δυνατά σοι· παρένεγκε τὸ ποτήριον τοῦτο ἀπ᾽ ἐµοῦ· ἀλλ᾽ οὐ τί ἐγὼ θέλω ἀλλὰ τί σύ [A-D] (Mark 14:36) Abba, Father, all things are possible for you; take this cup from me; yet not what I will, but what you will. (6b) καὶ γὰρ ἠσθένησεν παραπλήσιον θανάτῳ· ἀλλὰ ὁ θεὸς ἠλέησεν αὐτόν [C-E] (Phil 2:27) For he was indeed ill to the point of death, but God had mercy on him. (6c) δυνάµενοι ἐν βάρει εἶναι ὡς Χριστοῦ ἀπόστολοι· ἀλλὰ ἐγενήθηµεν νήπιοι ἐν µέσῳ ὑµῶν [A-D] (1 Thess 2:7) We could have made demands as apostles of Christ, but we were gentle among you. (6d) καὶ γάρ ἐσµεν εὐηγγελισµένοι καθάπερ κἀκεῖνοι· ἀλλ᾽ οὐκ ὠφέλησεν ὁ λόγος τῆς ἀκοῆς ἐκείνους [C-E] (Heb 4:2) For we have indeed had good news preached to us, just as they also did, but the word they heard did not profit them.

In text (6a) it is first acknowledged that the Father is able to do anything, but then this fact is discounted so that Christ, rather than have it removed from him as he prayed, is indeed made to drink the cup of suffering. Epaphroditus in (6b) was on the verge of death and might well have been expected to have died, yet the second clause tells us that the situation turned out otherwise. Text (6c) presents the situation in which the apostles could have imposed themselves upon the Thessalonians for their up-keep, out of deference to their apostolic status, and yet Paul and the others made no claims on such a basis in this particular circumstance. The writer to the Hebrews states in (6d) that the generation of Hebrews in the wilderness heard the message of

26. Also sometimes referred to as “counter-presupposition.”

52

Biblical and Ancient Greek Linguistics 7

God, as had the author and his readers with much benefit, but in the case of the former, whereas one might reasonably have expected them equally to have profited from such hearing, this was in fact not so. From the foregoing texts, we see that ἀ$ά serves as the most general connector for concessive contrast. However, two other connectors are also found, καί and πλήν, though these are not used in an identical manner. On the many occasions where καί functions as the connector where a concessive relation pertains, it is significant that in every case the same manner of concession is expressed: (6e) καὶ παρέλαβεν τὴν γυναῖκα αὐτοῦ, καὶ οὐκ ἐγίνωσκεν αὐτὴν ἕως οὗ ἔτεκεν υἱόν [C-E] (Matt 1:24–25)27 He took her as his wife, but did not know her until she gave birth to a son. (6f) ἰδοὺ τοσαῦτα ἔτη δουλεύω σοι καὶ οὐδέποτε ἐντολήν σου παρῆλθον, καὶ ἐµοὶ οὐδέποτε ἔδωκας ἔριφον [C-E] (Luke 15:29) Look! For all these years I have been serving you, and I have never disobeyed your command; yet you have never given me even a young goat. (6g) ἐξῆλθον καὶ ἐνέβησαν εἰς τὸ πλοῖον, καὶ ἐν ἐκείνῃ τῇ νυκτὶ ἐπίασαν οὐδέν [C-E] (John 21:3) They went out and got into the boat, but that night they caught nothing.

Each single instance where καί appears in concessive contrast the sense is that of contra-expectation. Text (6e) relates how Joseph married Mary, and yet, contrary to what all would suppose, he had no marital relations with her at that time. The elder son speaking in (6f) draws his father’s attention to all the years of hard service he had offered him, for which he might reasonably have expected a treat of some kind, but had received none meeting his expectation. Example (6g) speaks of a group of seasoned fishermen embarking on a fishing trip on a lake they knew intimately, but despite fishing the whole night they caught nothing.

27. The translation “but” appears in many standard English versions, such as ESV, NIV, NASB, NRSV, NLT.

LUNN Categories of Contrast

53

When we come to πλήν in concessive environments, again all instances fall within a single sub-category ([A-D]): (6h) οὐαὶ τῷ κόσµῳ ἀπὸ τῶν σκανδάλων· ἀνάγκη γὰρ ἐλθεῖν τὰ σκάνδαλα, πλὴν οὐαὶ τῷ ἀνθρώπῳ δι᾽ οὗ τὸ σκάνδαλον ἔρχεται (Matt 18:7) Woe to the world because of offenses! For offenses must come, but woe to that man through whom the offense comes! (6i) λέγω ὑµῖν ὅτι ποιήσει τὴν ἐκδίκησιν αὐτῶν ἐν τάχει. πλὴν ὁ υἱὸς τοῦ ἀνθρώπου ἐλθὼν ἆρα εὑρήσει τὴν πίστιν ἐπὶ τῆς γῆς; [A-D] (Luke 18:8) I tell you, he will quickly grant them justice. Yet when the Son of Man comes, will he find faith on earth? (6j) πάντα ἰσχύω ἐν τῷ ἐνδυναµοῦντί µε. πλὴν καλῶς ἐποιήσατε συγκοινωνήσαντές µου τῇ θλίψει [A-D] (Phil 4:13–14) I prevail in all things through him who strengthens me. Nevertheless, you have done well to share in my distress.

In example (6h) the fact is first granted that offenses will come, an obviously undesirable situation. But this is then laid aside to express the worse evil that will come upon the one who is the cause of the offense. So the relation is between something acknowledged and something which is worse. This is clearly AD, of the configuration negative-more negative (see above). Next (6i) states that God will speedily vindicate, in the eschatological sense, his people. This is unreservedly positive, but then the more serious issue is posed as to whether there will be any faithful upon the earth at this time, which is in effect to discount the positivity of the previous element. Here again A-D is in evidence. In (6j) the context is that of the apostle relating how he had not been overwhelmed by various hardships, and the reason he gives for this is the strength he receives from his Lord. For the purposes of the second proposition the first is laid to one side, and Paul focuses on the help he had received from the Philippians. So all occurrences of πλήν in concessive contrast relate to the Acknowledged-Discounted configuration. A rule seems to have been established with regard to this category of contrast whereby ἀ$ά may properly function in both A-D and C-E types of concession, whereas καί is restricted to CE and πλήν to A-D. Interestingly, within the synoptic Gospels

54

Biblical and Ancient Greek Linguistics 7

there are occasions when one writer uses the more general concessive connector ἀ$ά, while another uses one of the more specific ones (e.g. Mark 14:36 uses ἀ$ά whereas the parallel in Matt 26:39 has πλήν in a context of A-D). At this point it is necessary to point out an exceptional usage on the part of Luke. The fact is that in the two books attributed to this author the particle δέ appears, as well as the aforementioned connectors, in contexts of concessive contrast. The number of occasions is not considerable relative to the whole. The texts in question are Luke 5:5; 8:38; 9:32, 61; 12:27; 14:34; 22:27, 32; 23:9; Acts 3:6; 5:19, 22; 22:3, 28; 27:26. These comprise a mixture of both kinds of concession, and we therefore ought to understand δέ to function as a substitute for the more general connector ἀ$ά. Interestingly, in a variant reading located in Luke 17:1, a context clearly denoting concession of the A-D subcategory, where the Byzantine text-type reads δέ, the Alexandrian has the more specific πλήν.28 Another alternative connecting term, though rarely used, is deserving of some mention. This is the postpositive µέντοι. Its handful of appearances shows first that µέντοι may have the completely non-contrastive sense of ‘actually’ or ‘indeed.’29 Yet it clearly functions also to conjoin utterances expressing a concessive contrasting relation, in which case it is generally translated as but, yet, or however. It is most common in the Gospel of John: (6k) καὶ ἐπὶ τούτῳ ἦλθαν οἱ µαθηταὶ αὐτοῦ καὶ ἐθαύµαζον ὅτι µετὰ γυναικὸς ἐλάλει· οὐδεὶς µέντοι εἶπεν, Τί ζητεῖς ἢ τί λαλεῖς µετ᾽ αὐτῆς; [C-E] (John 4:27)

28. The variant appears in the saying: “It is impossible that no offenses should come, but woe to the one through whom they come!” As an aside, it may be reasonably argued in this instance that the Byzantine text-type preserves the more original reading. Not only does the Alexandrian text display the more expected form, making it the least difficult reading, and therefore more likely to be the product of alteration (according to the principle of lectio difficilior potior), but it also harmonizes with the parallel saying in Matt 18:7, where πλήν stands without a variant. 29. BDAG, s.v. “µέντοι.”

LUNN Categories of Contrast

55

At this point his disciples came, and they were amazed that he had been speaking with a woman, but no one said, “What do you want?” or, “Why are you speaking with her?” (6l) πρωΐας δὲ ἤδη γενοµένης ἔστη Ἰησοῦς εἰς τὸν αἰγιαλόν, οὐ µέντοι ᾔδεισαν οἱ µαθηταὶ ὅτι Ἰησοῦς ἐστιν [C-E] (John 21:4) Just as day was breaking, Jesus stood on the beach, but the disciples did not know that it was Jesus.

All occurrences of µέντοι in the Fourth Gospel are C-E (see also 7:12–13; 12:42; 20:5). The three non-Johannine instances are either A-D or non-contrastive (2 Tim 2:19; Jas 2:8; Jude 8). One sole case of asyndeton, where no connecting term is present in a concessive relationship, was found in the data: (6m) ἅτινά ἐστιν λόγον µὲν ἔχοντα σοφίας ἐν ἐθελοθρησκίᾳ καὶ ταπεινοφροσύνῃ καὶ ἀφειδίᾳ σώµατος, οὐκ ἐν τιµῇ τινι πρὸς πλησµονὴν τῆς σαρκός (Col 2:23) These have indeed an appearance of wisdom in promoting selfimposed piety, humility, and severe treatment of the body, [but] they have no value in restraining the indulgence of the flesh.

Ellipsis with concessive contrast is not in evidence within the data. This is probably due to the fact that the two propositions involved do not need to have any parallel elements, as is required in instances of simple contrast. In these latter there is always some correspondence between the first and second propositions to establish a point of contact to render the simple contrast meaningful. This is not so in concessive contrast. Before leaving this category, there is a comparatively little used manner of contrast in the New Testament, which is probably best incorporated under the heading of concession, although some might wish to keep it separate. In essence it takes the form of a contrastive statement that imposes some kind of qualification upon the contents of the preceding clause. The connector remains the discontinuous ἀ$ά, hence its inclusion here. And even the relation might be construed as a form of concession. It is not of the A-D or the C-E types treated above, but in not a too dissimilar fashion, a statement is allowed, and then immediately qualified, rather than discounted, by what can only be described as a constraint or restriction.

56

Biblical and Ancient Greek Linguistics 7 (6n) ἕτοιµοι ἀεὶ πρὸς ἀπολογίαν παντὶ τῷ αἰτοῦντι ὑµᾶς λόγον περὶ τῆς ἐν ὑµῖν ἐλπίδος, ἀλλὰ µετὰ πραΰτητος καὶ φόβου (1 Pet 3:15–16)30 Always be ready to give a defence to anyone who asks you for a reason for the hope that is in you, but with gentleness and reverence.

The first proposition consists of the injunction to offer a reasoned defence of the faith, while what follows places upon it the constraint of the manner in which it is to be enacted. We observe the ellipsis in the final phrase, which is made good in several English versions (e.g., NIV: “But do this”). There is also one instance of such constraining contrast in which asyndeton appears to be in evidence: (6o) τὸν δὲ ἀσθενοῦντα τῇ πίστει προσλαµβάνεσθε, µὴ εἰς διακρίσεις διαλογισµῶν (Rom 14:1) Accept the one who is weak in faith, [but] not for quarrelling over disputable matters.

We note that ESV, NRSV, NASB, NJB, NKJV, and others, all insert the contrastive conjunction but.31 7. Replacing Contrast Here again the connector is ἀ$ά, as in the foregoing. A significant difference, however, lies in the prior occurrence of a negation as a necessary requirement. The basic configuration is simply that of “not X but Y,” where the X element is negated and replaced by Y.32 It always entails some form of negative in the first element for the replacement to occur. This helps to distinguish the category from concession, in which ἀ$ά also 30. See the discussion in Runge, “Meaningful Distinction between ἀλλά and εἰ µή, Pt. 2.” 31. Other instances of concessive contrast include: Matt 6:26; 9:18; 12:34; 17:16; 21:30; 22:3; 23:37; 26:60; Mark 4:31–32; 13:20; 14:28, 29, 56; Luke 1:7; 4:26; 8:29; 9:40, 52–53; 12:6; 17:1; 21:18; 22:22; John 1:10; 2:20; 3:8, 11; 6:36; 7:44; 10:8, 39; 13:10; 14:9; 16:12; 20:29; Acts 5:28; 7:5, 9; 9:26; 10:28; 18:17; 20:23–24; 23:3; Rom 1:21; 4:2; 8:36–37; 10:2; 14:20; 1 Cor 3:6; 6:12 [2x]; 8:5–6; 10:23 [2x] 14:17; 2 Cor 4:8 [2x], 9 [2x], 16; 5:16; 11:6; 13:4; Gal 2:2–3; 4:17; 6:13; 1 Thess 2:18; 1 Tim 1:13, 16; 2 Tim 1:12; 2:9; Jas 3:5; 2 Pet 2:5; 1 John 1:6; 2:9; 3 John 13; Rev 2:4; 9; 3:4; 10:9. 32. Runge (Discourse Grammar, 92–93) prefers to speak of “correction.”

LUNN Categories of Contrast

57

commonly serves as connector, but where there is no obligation for a preceding negative. Another feature of replacement is the variation in the extent of the replaced element. At times this might be another topic with its own entire clause, at other times a clause with the same topic as the first clause, and sometimes just a single constituent. This particular category of contrast is exceedingly common in the New Testament. Runge holds the view that the rejection of one entity before the introduction of the other is a deliberate device, which he terms a “point-counterpoint strategy.”33 This, he claims, employs a preceding negated element to highlight what follows. And so it would, in his opinion, serve as a prominencegiving device to draw greater attention to the latter of the two. A bare statement without this manner of counterpoint would, to his mind, be lacking in such prominence. I consider that Runge may well be correct with respect to many occurrences of this form of contrast. This is arguably the case where the negated element is contextually explicit, inferable, or part of common knowledge. In such cases its entrance in the “not X but Y” is strictly redundant, as in example (7c) below, and so may well serve to confer prominence. However, in other instances the negated first element is itself a new proposition, which is then immediately replaced by what follows, as in example (7f) below. Here the opening proposition would appear to be saying something of itself, rather than merely functioning as a counterfoil to what comes next. The following are typical instances of replacing contrast: (7a) καὶ µὴ εἰσενέγκῃς ἡµᾶς εἰς πειρασµόν, ἀλλὰ ῥῦσαι ἡµᾶς ἀπὸ τοῦ πονηροῦ (Matt 6:13) And do not lead us into temptation, but deliver us from the evil one. (7b) ὥστε µηκέτι αὐτὸν δύνασθαι φανερῶς εἰς πόλιν εἰσελθεῖν, ἀλλ᾽ ἔξω ἐπ᾽ ἐρήµοις τόποις ἦν (Mark 1:45) So he could no longer openly enter the city, but he was outside in deserted places. (7c) οὐκ ἐψεύσω ἀνθρώποις ἀλλὰ τῷ θεῷ (Acts 5:4) You have not lied to men but to God.

33. Runge, “Teaching Them What NOT to Do,” 7–8.

58

Biblical and Ancient Greek Linguistics 7 (7d) οὐ γὰρ δήπου ἀγγέλων ἐπιλαµβάνεται ἀλλὰ σπέρµατος Ἀβραὰµ ἐπιλαµβάνεται (Heb 2:16) For indeed he does not give help to angels, but he does give help to the seed of Abraham.

All the foregoing texts contain an explicit negative particle in the opening clause. This, of course, can take on numerous realizations, typically formed with the negators (οὐ, µή). Yet other kinds of negation exist. One of these can be a question anticipating a negative answer, as in 1 Cor 10:19–20: (7e) τί οὖν φηµι; ὅτι εἰδωλόθυτόν τί ἐστιν ἢ ὅτι εἴδωλόν τί ἐστιν; ἀλλ᾽ ὅτι ἃ θύουσιν, δαιµονίοις καὶ οὐ θεῷ What am I saying then? That a thing sacrificed to idols is anything, or that an idol is anything? But that the things which they sacrifice, they sacrifice to demons and not to God.

Here Paul poses a question, immediately followed by a second question suggesting an answer. Yet the answer, about the nature of idols and what is offered to them, is obviously not the correct answer. This is not what the apostle is stating. So since the answer is implicitly negative, the subsequent ἀ$ά is contextually appropriate. The implicit “No” is actually supplied in several English translations (e.g., ESV, NRSV, NASB, NIV, NJB).34 The major proportion of replacing contrast sentences comprise clauses that manifest a single topic. When the second clause concerns a different topic the particular constituent denoting that topic is marked, that is, fronted. This is to say that the rules that we saw with regard to word order in the contrasting topic equally apply in this case. For example: (7f) ὃς δ᾽ ἂν πίῃ ἐκ τοῦ ὕδατος οὗ ἐγὼ δώσω αὐτῷ, οὐ µὴ διψήσει εἰς τὸν αἰῶνα, ἀλλὰ τὸ ὕδωρ ὃ δώσω αὐτῷ γενήσεται ἐν αὐτῷ πηγὴ ὕδατος ἁλλοµένου εἰς ζωὴν αἰώνιον (John 4:14) Whoever drinks of the water that I will give him will never thirst; but the water that I will give him will become in him a spring of water welling up to eternal life.

34. See the discussion in Brannan (“The Discourse Function of ἀλλά,” 13) on the similarly implicit negation in Matt 11:7–8.

LUNN Categories of Contrast

59

Ellipsis abounds in replacing contrast, and not only in the second clause. On occasion the ellipsis works in the opposite direction, where the gap occurs among the constituents of the first proposition, later to be filled in the second. (7g) οὐ χρείαν ἔχουσιν οἱ ὑγιαίνοντες ἰατροῦ ἀλλὰ οἱ κακῶς ἔχοντες (Luke 5:31) Those who are well have no need of a physician, but those who are sick. (7h) ἐν γὰρ Χριστῷ Ἰησοῦ οὔτε περιτοµή τι ἰσχύει οὔτε ἀκροβυστία ἀλλὰ πίστις δι᾽ ἀγάπης ἐνεργουµένη (Gal 5:6) For in Christ Jesus neither circumcision nor uncircumcision counts for anything, but faith working through love. (7i) οὐκ ἐξ ἔργων τῶν ἐν δικαιοσύνῃ ἃ ἐποιήσαµεν ἡµεῖς ἀλλὰ κατὰ τὸ αὐτοῦ ἔλεος ἔσωσεν ἡµᾶς (Titus 3:5) Not by works of righteousness which we ourselves have done, but according to his mercy he saved us. (7j) . . . εἰς τὸ µηκέτι ἀνθρώπων ἐπιθυµίαις, ἀλλὰ θελήµατι θεοῦ τὸν ἐπίλοιπον ἐν σαρκὶ βιῶσαι χρόνον (1 Pet 4:2) . . . so that no longer for human desires, but for the will of God one should live the rest of one’s time in the flesh.

The second clause of example (7g) is grammatically and semantically incomplete. It is simply a subject constituent, or topic, with no focus. Evidently the focal domain, consisting of a verb and object, χρείαν ἔχουσιν, has to be understood from the first clause. A verbal phrase has likewise been elided from the contrastive element of text (7h). The reader therefore needs to supply τι ἰσχύει from the foregoing clause. This is something that the NIV (“the only thing that counts is faith expressing itself through love”) and NRSV (“the only thing that counts is faith working through love”) do explicitly. In the last two examples ellipsis affects the first statement. In (7i) the initial proposition offers no complete sense until we reach ἔσωσεν ἡµᾶς in the following. It is interesting to note that several modern versions find this awkward or unnatural in English and so bring “he saved us” into the primary clause (e.g., ESV, NRSV, NASB, NIV). Similarly, (7j) has a gapped first clause with a partial meaning that is not fully comprehended until the end of the second. Again English versions bring the relevant words, here τὸν ἐπίλοιπον ἐν

60

Biblical and Ancient Greek Linguistics 7

σαρκὶ βιῶσαι χρόνον, forward (e.g., ESV, NRSV, NASB, NIV, NJB). A few instances exist in which an alternative connector appears in the place of ἀ$ά. It would seem that the combination of the particle δέ with the adverb µᾶ$ον ‘rather, instead’ creates a suitable substitution for ἀ$ά in contexts of replacement. This is so in Matt 10:6, 28; Eph 4:28; 5:11; Heb 12:13; and as a variant reading in Matt 25:9. Seven further places are to be found in the data in which δέ alone appears. In all of these instances apart from one, the second replacing element consists of a distinct clause, in which the grammatical subject is the same as that in the preceding element, as is often so, though not universally, in replacing contrast sentences. What we might be witnessing here, therefore, is a blurring of the distinction between the replacing contrast structure and that of contrasting focus, discussed earlier. In a sentence of the latter kind, where the first element is negated and the second not, there is some outward resemblance to the form of replacing contrast, though the inner logic is subtly different. Whereas replacement involves the removal of one in favor of another in a given situation, contrasting focus does not limit the choice to the two options mentioned. This is the difference between “I do not like tea but [I do like] coffee” (contrasting focus), and “I don’t want tea but coffee” (replacing contrast). In the former, unlike the latter, the first element is spoken of negatively, but cannot strictly be said to be replaced or corrected.35 Such a similarity in form, I suggest, along with this more subtle distinction in meaning, gave rise to this small number of instances in which the δέ used in contrasting focus clauses found its way into replacing clauses.

35. Jas 2:11 illustrates this well: “Now if you do not commit adultery, but [δέ] you do murder . . .” Taken superficially it might be mistaken for a replacing contrast, since the first element boasts a negative. However, the act of murder cannot be said in any way to replace that of adultery. The two belong to altogether different moral domains. It is rather a case of a single individual, the topic, does not do X, but does do Y, quite independently of each other, which is a definite instance of contrasting focus.

LUNN Categories of Contrast

61

The seven exceptions in question are interesting in that they may be grouped into three sets. The first consists of two texts found in synoptic Gospels, these being Matt 6:19–20 and Luke 10:20. Both share the same intrinsic features, each consisting of a long initial negated element, paralleled by an equally long second element using the exact same verb, but positively. So the first shows Μὴ θησαυρίζετε ὑµῖν θησαυροὺς ἐπὶ τῆς γῆς . . . θησαυρίζετε δὲ ὑµῖν θησαυροὺς ἐν οὐρανῷ (Do not store up treasure for yourselves on earth . . . but store up treasure for yourselves in heaven). The second reads µὴ χαίρετε ὅτι . . . χαίρετε δὲ ὅτι ‘Do not rejoice that . . . but rejoice that . . .’ What they also have in common is that after the negated verb there is a syntactic insertion of a subsidiary clause before coming to the positive verb. In the Matthean text we have the addition of the relative spatial clause ὅπου σὴς καὶ βρῶσις ἀφανίζει καὶ ὅπου κλέπται διορύσσουσιν καὶ κλέπτουσιν ‘where moth and rust destroy, and where thieves break in and steal,’ while in the Lucan text we find the complement clause ὅτι τὰ πνεύµατα ὑµῖν ὑποτάσσεται ‘that the spirits are subject to you’. It is possibly the insertion of these intervening clauses, with the resultant distancing of the positive verb from the preceding negative, that gives these sentences a different character in which the authors thought that the contrasting focus form was suitable even while the sense remained essentially that of replacing contrast. Four of the other passages are highly significant in that all are attributed to the same author. Indeed, all occur within the same short epistle: 1 Pet 1:12; 2:23; 3:9; and 4:16. Here none of the possible contributory factors adduced above are present. There is no intervening clause to create distance, and the positive verbs differ from the negated. Evidently what we are looking at in this case is something stylistic with regard to this particular writer.36 The outstanding passage of this kind is 1 Tim 1:9, which is strictly anomalous in that it is the sole one of these texts not to exhibit a complete clause in its replacing element.

36. One further instance may appear in 2 Pet 3:17, but this could be otherwise construed.

62

Biblical and Ancient Greek Linguistics 7

One single text exists in which it would seem that πλήν is employed in a replacing context. In Luke 23:28 Jesus is recorded to have said to the women of Jerusalem, µὴ κλαίετε ἐπ᾽ ἐµέ· πλὴν ἐφ᾽ ἑαυτὰς κλαίετε καὶ ἐπὶ τὰ τέκνα ὑµῶν ‘Do not weep for me, but weep for yourselves and for your children.’ What we could be looking at here is another case of linguistic blurring as was noted above with respect to replacement and contrasting focus. In this instance, it might just be that replacement is mixed with concession. The use of πλήν in concessive contrast has already been discussed. The other possibility for this term is the sense of exception (see below), which, according to the defined elements of that category, is patently not applicable here. Πλήν denotes that form of concession in which the initial element is acknowledged and then set aside for something else, while in replacement that which comes first is completely removed in favor of what is viewed as the correct information. Simple replacement here on the lips of Jesus might sound rather blunt. The use of πλήν, however, might impart to the utterance the connotation that the women’s weeping for him receives some acknowledgment before moving on to the more appropriate action—weeping for themselves and their children. With regard to replacing, contrast asyndeton is scarcely to be found. There is just one candidate among the data: (7k) καὶ τοῦτο --οὐκ ἐξ ὑµῶν, θεοῦ τὸ δῶρον (Eph 2:8) And this is not of yourselves, [but] it is the gift of God.

The preceding negative, together with the logical relation of contrast between something being “of yourselves” or “of God,” satisfies the essential requirements for replacement. A small number of English versions do indeed add but (e.g., NJB: “not by anything of your own, but by a gift from God”; GNT: “It is not the result of your own efforts, but God’s gift”), though most of the major versions do not (ESV, NRSV, NASB, NIV, NKJV). Amongst these latter the punctuation reveals two different ways of construing the text. NASB (“and that not of yourselves, it is the gift of God”) and NIV (“and this not from yourselves, it is the gift of God”) follow the replacing contrast reading and retain

LUNN Categories of Contrast

63

the asyndeton of the Greek. Others, such as NRSV37 (“and this is not your own doing; it is the gift of God—not the result of works”) and NKJV (“and that not of yourselves; it is the gift of God, not of works”), divide the text otherwise. These make a break after καὶ τοῦτο οὐκ ἐξ ὑµῶν and then connect θεοῦ τὸ δῶρον with the opening phrase of v. 9, οὐκ ἐξ ἔργων ‘not of works’. Both seem possible. However, the punctuation in printed Greek Testaments consistently favors replacing contrast. Lastly, we note that sometimes the replacement can take place after a sizeable span of text. We observe the three verse intervals, for example, between “You have not [οὐ] come to a mountain that can be touched and that burns with fire” (Heb 12:18) and “But [ἀ$ά] you have come to Mount Zion . . .” (v. 22).38 8. Additive Contrast Since this next category is to a large degree merely an extension of the previous, it will only require the briefest treatment. In essence this employs the contrasting relation, with some adaptation, to express the idea of addition. The replacing contrast structure, “not X but Y” (οὐ . . . ἀ$ά . . .) is further supplied with “only” (µόνος) in the first element and “also” (καί) in the second,

37. ESV adopts the same basic interpretation. 38. Other instances of replacing contrast include: Matt 4:4; 5:17; 7:21; 9:17; 10:20; 13:21; 15:11; 16:12; 19:6; 22:30; Mark 1:44, 45; 2:17, 22; 3:26; 4:17; 5:26, 39; 6:52; 7:5, 15; 9:37; 10:45; 12:25; 13:11; Luke 1:60; 4:4; 5:31, 32; 7:7; 8:16, 27, 52; 9:56; 11:33; 12:5; 14:13; 16:30; 18:13; 20:21, 38; 22:26, 42; 24:6; John 1:8, 13, 33; 3:16, 17, 28; 4:2; 5:22, 24, 30; 6:27, 38, 39; 7:10, 16, 22; 8:12, 16, 49; 9:3, 31; 10:1, 33; 11:4, 54; 12:6, 27; 14:24; 15:19; 16:25; 17:20; 19:24; 20:27; 21:8; Acts 1:4; 4:32; 7:39; 10:41; 16:37; 18:9; 21:24; 26:25; Rom 2:13, 29; 3:27; 4:4, 10, 13; 6:13, 14; 7:15, 17, 19, 20; 8:4, 26, 32; 9:7, 8; 11:18, 20; 12:2, 16, 21; 14:13; 16:18; 1 Cor 1:17; 2:4, 5, 12; 3:1; 4:14, 19, 20; 5:8; 6:13; 7:4 [2x], 10; 9:12; 10:24, 33; 11:8, 9; 12:14; 14:2, 22, 33, 34; 15:46; 2 Cor 1:9, 24; 2:4, 17; 3:3, 6; 4:5, 18; 5:15; 8:5, 8; 10:4, 13, 18; 12:14; 13:3, 7, 8; Gal 1:1, 12, 17; 4:1–2, 14, 31; 5:13; 6:13, 15; Eph 2:19; 4:29; 5:17, 18, 29; 6:4, 12; Phil 1:20; 2:3, 4, 6–7, 12; 3:9; 4:17; Col 2:5; 3:11; 1 Thess 2:8, 13; 4:8; 5:6, 9, 15; 2 Thess 2:12; 3:8, 11, 15; 1 Tim 3:3; 5:1, 23; 6:17; 2 Tim 1:7, 9; 2:24; Heb 7:16; 10:39; Jas 1:25; 1 Pet 1:19; 1 John 3:18; 4:1; Rev 2:9; 3:9.

64

Biblical and Ancient Greek Linguistics 7

to produce “not only X but also Y.” So although distinctly contrastive in its basic form, its overall function is to add, hence the designation additive contrast, also sometimes referred to as “expanding contrast.”39 This category may be illustrated by the following: (8a) . . . ὅτι οὐ µόνον ἔλυεν τὸ σάββατον, ἀλλὰ καὶ πατέρα ἴδιον ἔλεγεν τὸν θεόν (John 5:18) . . . because he not only broke the Sabbath, but also said God was his Father. (8b) ἐγὼ γὰρ οὐ µόνον δεθῆναι ἀλλὰ καὶ ἀποθανεῖν εἰς Ἰερουσαλὴµ ἑτοίµως ἔχω ὑπὲρ τοῦ ὀνόµατος τοῦ κυρίου Ἰησοῦ (Acts 21:13) For I am ready not only to be bound, but also to die in Jerusalem for the name of the Lord Jesus. (8c) οὐ µόνον αὐτὰ ποιοῦσιν ἀλλὰ καὶ συνευδοκοῦσιν τοῖς πράσσουσιν (Rom 1:32) They not only do the same things, but also approve those who practice them. (8d) ὅτι ὑµῖν ἐχαρίσθη τὸ ὑπὲρ Χριστοῦ, οὐ µόνον τὸ εἰς αὐτὸν πιστεύειν ἀλλὰ καὶ τὸ ὑπὲρ αὐτοῦ πάσχειν (Phil 1:29) For it has been granted to you on behalf of Christ not only to believe in him, but also to suffer for his sake.

Its form is self-evident and requires no further comment. As regards the meaning, as was seen with the foregoing category, the first element is sometimes assumed or predictable, while at other times it contains a new proposition. So in (8d) “to believe in him” is a known proposition, while in (8b) “ready . . . to be bound” is new information. In the former situation, Runge’s “point-counterpoint strategy” may well apply. In this latter situation, however, the second proposition must contain something that is to a greater degree uninferable, or more striking, than the first. In such cases one might justifiably render καί as “even.” So in both (8b) and (8c) we find “even” appearing in several modern translations (e.g., NRSV, NJB).40

39. Cf. Dik, Theory of Functional Grammar, 331. 40. The remaining instances of additive contrast are: Matt 21:21; John 11:52; 12:9; 13:9; 17:20; Acts 19:26, 27; 26:29; 27:10; Rom 4:12, 16; 5:3, 11; 8:23; 9:10, 24; 13:5; 16:4; 2 Cor 7:7; 8:10, 19, 21; 9:12; Eph 1:21; Phil 2:27; 1

LUNN Categories of Contrast

65

9. Exceptive Contrast In this final category of negated contrast we are thinking of contrast in the basic sense of “No X but Y.” This is, of course, quite a different relation to that expressed by “Not X but Y,” seen in replacing contrast discussed earlier, though common ground exists in that both must include a negative preceding the contrastive element. We note that whereas replacement prefers the connector ἀ$ά, we are now looking at εἰ µή or ἐὰν µή.41 A typical example would be a statement of the kind, “We have no king but Caesar” (John 19:15). Runge has offered a helpful explanation of the semantic distinction between replacement and exception.42 He proposes that in the former the two entities involved belong to two directly antithetical sets, which is indeed so, while in the latter the element in the second part belongs to the same set as that in the first but is being referenced as the sole element in that set that is pertinent to the proposition being stated. So in the above New Testament example, Caesar is taken as a king, and the Jews are made to declare that he is the sole member of the set labelled “kings” that they would own as theirs. Although this instance follows the principle, on this matter of sets the evidence, it must be said, does not entirely fit with what Runge advocates. It will be proven that in some exceptive clauses in New Testament Greek the subsequent excluded element does not belong to the same set as the first. A prominent component involved in the logical morpho-logy of exception is quantification. Exceptive phrases are most commonly found following propositions which contain a Thess 1:5, 8; 2:8; 1 Tim 5:13; 2 Tim 2:20; 4:8; Heb 12:26; 1 Pet 2:18; 1 John 2:2; 2 John 1. These references include the five appearances of the Pauline expression οὐ µόνον δέ, ἀλλὰ καί ‘Not only this, but also . . .’ occurring four times in Romans and once in 2 Corinthians. 41. These connectors are, of course, not restricted to the sense discussed here. Both also serve as conjunctive phrases introducing negative conditions (“If . . . not . . .” or “Unless . . .”), as in Matt 24:22 and Acts 27:31. This multivalence is similarly evident in other connectors that may appear in contrastive contexts, such as δέ and καί. 42. Runge, “Meaningful Distinction between ἀλλά and εἰ µή, Pt. 2.”

66

Biblical and Ancient Greek Linguistics 7

universal quantifier.43 Here we are thinking of quantifying terms that are absolute in their degree of inclusion or of ex-clusion. So in positive utterances we find such expressions as all . . . but/ except and every . . . but/except. Negative pro-positions consist of no . . . but/except, none . . . but/except, nobody . . . but/except, nothing . . . but/except, not any . . . but/except, and the like. Quantifiers such as a few, some, many, and most are inadmissible in exceptive clauses on self-evident logical grounds.44 In the language of the New Testa-ment all instances of universal quantification in exceptive contrast are negative in character. Here are several examples: (9a) γενεὰ πονηρὰ καὶ µοιχαλὶς σηµεῖον ἐπιζητεῖ, καὶ σηµεῖον οὐ δοθήσεται αὐτῇ εἰ µὴ τὸ σηµεῖον Ἰωνᾶ τοῦ προφήτου (Matt 12:39) An evil and adulterous generation seeks for a sign; yet no sign will be given to it but the sign of Jonah the prophet. (9b) τοῦτο τὸ γένος ἐν οὐδενὶ δύναται ἐξελθεῖν εἰ µὴ ἐν προσευχῇ (Mark 9:29) This kind cannot come out by anything but by prayer. (9c) οἴδαµεν ὅτι οὐδὲν εἴδωλον ἐν κόσµῳ καὶ ὅτι οὐδεὶς θεὸς εἰ µὴ εἷς (1 Cor 8:4) We know that an idol is nothing at all in the world and that there is no God but one.

Example (9a) has Jesus speak of signs as the principal set, from among which initially “no sign” is to be included, but then, with the addition of the exceptive phrase, from this same set one member is identified to be granted to the Jews. In (9b) the universal quantifier is the instrumental phrase “by nothing,” that is, by no means, following which “by prayer” is made the sole exception, that is, the only means by which this manner of exorcism may be accomplished. Finally, (9c) contains a negative proposition in which the existence of any member of the set “God” is first denied (“there is no God”), and then just a single

43. For this designation, see, for example, Steedman, Taking Scope, 147. 44. One cannot say “Many came to the wedding except John and Mary,” for if these two are the only exceptions one would need to say “They all came to the wedding except John and Mary.” If there are other exceptions, these would need to be included along with John and Mary.

LUNN Categories of Contrast

67

particular entity within that set is excepted—the God in whom Paul believes. In each of the above instances the exceptive element is preceded by a proposition that includes a universal quantifier, and it is in such circumstances that Runge’s rule with regard to membership of the same set applies. Nevertheless, a survey of all instances of the exceptive use of εἰ µή and ἐὰν µή in the New Testament reveals a number of cases where the excepted component is evidently not a member of the set previously identified. This we may observe in: (9d) οὐκ ἀνέγνωτε τί ἐποίησεν Δαυὶδ ὅτε ἐπείνασεν καὶ οἱ µετ᾽ αὐτοῦ, πῶς εἰσῆλθεν εἰς τὸν οἶκον τοῦ θεοῦ καὶ τοὺς ἄρτους τῆς προθέσεως ἔφαγον, ὃ οὐκ ἐξὸν ἦν αὐτῷ φαγεῖν οὐδὲ τοῖς µετ᾽ αὐτοῦ εἰ µὴ τοῖς ἱερεῦσιν µόνοις; (Matt 12:3–4) Have you not read what David did when he and those with him were hungry, how he entered the house of God and ate the bread of the Presence, which was not lawful for him or those with him to eat, but only for the priests? (9e) καὶ οὐ µὴ εἰσέλθῃ εἰς αὐτὴν πᾶν κοινὸν καὶ ὁ ποιῶν βδέλυγµα καὶ ψεῦδος εἰ µὴ οἱ γεγραµµένοι ἐν τῷ βιβλίῳ τῆς ζωῆς τοῦ ἀρνίου (Rev 21:27) And nothing unclean and no one who practices abomination and falsehood will enter it, but [only] those whose names are written in the Lamb’s book of life.

In (9d) it is patently obvious that the initial set and the subsequent set are totally distinct, speaking on the one hand of David and his men and of the priests on the other. The whole point of Jesus’s saying is that David and those accompanying him were not priests and therefore not permitted to eat the sacred bread. Text (9e) is speaking of the New Jerusalem. First, it is stated who will be barred from entering the city, and then, in the structure of an exceptive phrase, those who are permitted entrance are identified. Without doubt these latter do not belong to the same set as the former. Exclusion from the city pertains to the “unclean,” and the one “who practices abomination and falsehood.” The contrasting element relates to “those whose names are written in the Lamb’s book of life.” Plainly these are not members of the foregoing set. It is not at all the intention to declare that only those showing the ungodly attributes whose

68

Biblical and Ancient Greek Linguistics 7

names are in the book of life can enter the city. According to the plain message of the book, these are two totally distinct categories who share radically different fates. So what is entailed in this manner of exception is a contrast between two different sets, as described. What this involves, as a concomitant feature, is the important fact that the first set lacks universal quantification. Rather than impose a universality, it specifies a particular group, and it is this that allows for an altogether different set to be presented in the second element. Other passages where the exceptive element is not a member of the antecedent set are: Luke 4:25–26 (many widows in Israel/ the widow of Zarephath in Sidon), also v. 27 (many lepers in Israel/Naaman the Syrian); and Rev 9:4 (vegetation and trees/ those people not bearing the seal of God on their foreheads).45 A related exceptive structure is found where the connecting term is πλήν. In such instances πλήν is a preposition for which the governed element displays the genitive case. This serves to distinguish it grammatically, alongside the semantic distinctions, from the use of πλήν in concession, where it occurs as a conjunction.46 This prepositional use could be grouped in the category of exceptive contrast, though the grammar differs and the sense is not absolutely identical. There are just four47 instances among the data:

45. English exceptive clauses do in fact make a distinction where membership to the same set does and does not pertain. To say that “Nobody has been to the moon except Americans” is fine. However, a sentence such as “No Europeans have been to the moon, except Americans,” would be nonsensical, since it would logically entail that Americans belong to the set of Europeans. Natural English would, therefore, express it as “No Europeans have been to the moon, but only Americans.” It is interesting, in this context, to observe that in the instances in the New Testament where the excepted elements belong to a distinct set, many of the common modern versions render εἰ µή as “but only” (Matt 12:4, ESV, NRSV, NIV, NJB; Luke 4:26, NASB, ESV; 4:27, NASB, ESV, NJB; Rev 9:4, NASB, ESV, NRSV, NIV; 21:27, NASB, ESV, NRSV, NIV, NJB). 46. There is also the higher, discourse-level function of πλήν, which is not discussed in this article. 47. The Byzantine text-type displays a variant at John 8:10 (“he saw no one but [πλήν] the woman”), in the disputed pericope adulterae.

LUNN Categories of Contrast

69

(9f) ἐπ᾽ ἀληθείας εἶπες ὅτι εἷς ἐστιν καὶ οὐκ ἔστιν ἄλλος πλὴν αὐτοῦ (Mark 12:32) You truly said that God is one and there is no other but him. (9g) πάντες δὲ διεσπάρησαν κατὰ τὰς χώρας τῆς Ἰουδαίας καὶ Σαµαρείας πλὴν τῶν ἀποστόλων (Acts 8:1) All were scattered throughout the regions of Judea and Samaria except the apostles. (9h) . . . µηδὲν πλέον ἐπιτίθεσθαι ὑµῖν βάρος πλὴν τούτων τῶν ἐπάναγκες (Acts 15:28) . . . not to burden you with anything but these necessary things. (9i) ἀποβολὴ γὰρ ψυχῆς οὐδεµία ἔσται ἐξ ὑµῶν πλὴν τοῦ πλοίου (Acts 27:22) For there will be no loss of life among you, but [only] of the ship.

As regards the “set” rule laid out above with respect to εἰ µή, we here observe that in text (9g) the apostles are plainly to be included within the universal set ‘all,’ while in (9h) it is evident that the universal ‘anything’ of the first element covers the ‘necessary things’ of the second. However, in (9f) and (9i) the rule does not apply. Example (9f) gives the proposition ‘there is no other [God].’ This set is therefore, not ‘gods,’ but strictly ‘other gods,’ which has already created, through the qualification ‘other,’ a semantic distinction between these alternative deities and Paul’s God. If the second phrase, πλὴν αὐτοῦ, denoted an excepted member of the previous set, then the statement would be indicating that there is only one other god as well as the one true God, making two deities. This is certainly not the intention. In (9i) the first element speaks of loss of life, a quantification that is non-universal in scope, and the second the loss of the ship. So again, we find that the latter does not belong to same set as the former. We discover, then, that the same general rule about universal quantification and set membership applies as it did in the case of sentences constructed with εἰ µή. With regard to πλήν, in texts (9f) and (9i) the preposition may, we note, be given the meaning ‘besides,’48 one of its attested lexical senses when functioning as a preposition. 48. See LSJ, s.v. “πλήν.” This is in fact the translation given for πλήν in Mark 12:32 by the ESV and NASB.

70

Biblical and Ancient Greek Linguistics 7

There is one sole occurrence of ἐκτός functioning as a preposition with the genitive case in an exceptive sense. This is Acts 26:22, “saying nothing but [οὐδὲν ἐκτός] what the prophets and Moses said would happen.” Here it adheres to the same principles regarding set membership, as seen above.49 10. Application to Translation and Exegesis We come now to give some consideration to the practical application of the foregoing. Most generally, we may argue that an appreciation of the various distinct categories of contrast grants the translator and exegete an awareness that can only be advantageous when it comes to a close reading of a text. In English it is easy to pass over the distinctions involved in the contrastive conjunction but. In a cursory reading of the following verse, for example, one can readily observe the presence of two contrasts, but greater discernment reveals their distinct functions: Truly, truly, I say to you, that you will weep and lament, but the world will rejoice; you will grieve, but your grief will be turned into joy. (John 16:20 NASB, italics added)

An examination of the Greek discloses that the first but consists of a structure formed with δέ and with a fronted topic, ὁ δὲ κόσµος χαρήσεται. This is the standard realization of a contrasting topic clause, the contrast being between the ‘you’ that will weep and the ‘world’ that will rejoice. The second but in fact renders ἀ$ά. The immediate context shows, by the lack of any preceding negative, that this cannot be replacement or addition. It is therefore an instance of concession. Here it is unambiguously concession of the Acknowledged-Discounted type. First, the grief of the disciples is acknowledged and then set aside. Something negative, which is conceded, is later turned into

49. Other instances of exceptive contrast include: Matt 11:27; 12:4; 16:4; 21:19; 24:36; Mark 2:7; 5:37; 10:18; 13:32; Luke 5:21; 6:4; 11:29; 17:18; 18:19; John 3:13; 6:46; 14:6; 17:12; Acts 11:19; Rom 13:1; 1 Cor 1:14; 2:11; 10:13; 2 Cor 2:2; Gal 1:19; Heb 3:18; 1 John 2:22; 5:5; Rev 9:4; 19:12.

LUNN Categories of Contrast

71

something positive.50 The twofold occurrence of but in this verse, therefore, is hiding two differently nuanced contrasting relations. Some English versions, we note, even include three instances of but in a single verse, as in 1 Cor 14:22 (NASB, NKJV), reflecting ἀ$ὰ . . . δὲ . . . ἀ$ὰ . . . (replacing contrast, contrasting topic, replacing contrast). More significant is that an understanding of contrasting relations allows the exegete to detect errors in the renderings of the various English versions and in the analyses of commentators. If we take, in the first instance, the use of the independent pronouns, ἐγώ, σύ, αὐτός and so forth, one soon encounters misunderstandings. Above it was seen that in the second proposition of a contrasting topic relation, where there was no overt noun or noun phrase occupying the subject in the clause-initial position, then the independent pronoun or articular pronoun appeared in that same position. This we recall was strictly redundant from a purely semantic perspective (since the identity of the person indicated by the pronoun is evident from the bare form of the verb), but was necessary to bear the markedness of the contrasted topic. An altogether different use of the pronoun, as is well known, is to express the concept of “self,”51 as in Mark 12:36 (αὐτὸς Δαυίδ ‘David himself’), 1 Cor 15:28 (αὐτὸς ὁ υἱός ‘the Son himself’), Rev 22:3 (αὐτὸς ὁ θεός ‘God himself’), where special attention is drawn to the individual performing the action, as distinct from any other. The pronoun, of course, may stand alone, where there is no explicit nominal subject, as for example in Acts 17:25: “He [God] is not served by human hands, as though he needed anything, because he himself gives [αὐτὸς διδοὺς] all people life and breath and everything else.” In contrast, however, the fronted pronoun highlights not the individual himself or herself, but rather the fact that this topic is acting in an antithetical manner to the previous topic. So the two functions of the pronoun are semantically quite distinct. In translation, however, the two are sometimes confused. This we see, for example, in:

50. See the discussion in the earlier section on concession. 51. BDAG, s.v. “αὐτός.”

72

Biblical and Ancient Greek Linguistics 7 αὐτὸς δὲ ἐκάθευδεν (Matt 8:24) But Jesus Himself was asleep. (NASB) αὐτοὶ δὲ ἐπλήσθησαν ἀνοίας (Luke 6:11) But they themselves were filled with rage. (NASB) ἡµεῖς δὲ τῇ προσευχῇ . . . προσκαρτερήσοµεν (Acts 6:4) We ourselves will continue to devote ourselves to prayer. (NJB)

What we are looking at in all the above Greek texts is the contrastive use of the independent pronoun, as a marked topic, not the reflexive use. Renderings, therefore, that include -self or selves are unwarranted. In this the vast majority of English versions agree (cf., NRSV, ESV, NIV, NKJV), but those specifically cited above all err in this respect. An even more serious error is the assignment of a contrast to a completely wrong category, or seeing a contrast where none is intended. Note the following almost universally mistranslated sentence: Πάλιν ἠκούσατε ὅτι ἐρρέθη τοῖς ἀρχαίοις, Οὐκ ἐπιορκήσεις, ἀποδώσεις δὲ τῷ κυρίῳ τοὺς ὅρκους σου (Matt 5:33) Again, you have heard that it was said to those of old, “You shall not swear falsely, but shall perform your oaths to the Lord.”

The principal modern English versions all understand the words of Jesus’s citation here to be a contrast of the replacing kind (“not . . . but . . .”). It seems a reasonable and meaningful rendering of the words. Jesus is made to say basically that oaths should not be taken falsely, but should be enacted upon. What is problematic about this, however, is that the relation of replacement is expressed by ἀ$ά, as hundreds of other instances of this category of contrast attest. The particle δέ, on the other hand, when in contrasting contexts, principally serves as the connector for contrasting focus and contrasting topic. Also, as stated earlier, δέ is essentially a marker of development, not of contrast per se, and may be translated by and, so, and but. Furthermore, for replacement to exist in this context, the two elements have to be speaking of the same basic issue for the relationship to make sense. This is to say, the positive clause You shall perform your oaths to the Lord has to be a logical substitute for You shall not swear falsely. Although the English translations

LUNN Categories of Contrast

73

make this appear so, in Greek it is questionable that this is actually the case. It is more likely that the two verbs of the citation, though belonging to the same semantic domain, do not in fact bear the same meaning, and therefore the second positive utterance cannot be a replacement for the preceding negative one. One lexical sense of the verb ἐπιορκέω is to ‘commit perjury,’52 that is, to give false testimony under oath. This is in fact the only attested sense in biblical Greek of this verb and its cognates. In the LXX the phrase πᾶς ὁ ἐπίορκος (Zech 5:3) is rendered as “everyone who swears falsely” by several versions, and as “everyone who commits perjury” by NJB.53 Its cognate abstract noun, ἐπιορκία, is rendered as “perjury” in its sole LXX appearance (Wis 14:25) by its modern translators,54 and the nearby verb ἐπιορκοῦσιν (v. 28) as “[they] commit perjury” or “perjure themselves.” Closer to home, the only cognate to be found in the New Testament, ἐπιόρκοις (1 Tim 1:10), is translated in virtually all standard modern versions as “perjurers,” and occurs in association with ψεύσταις, “liars.” It would seem, then, that ἐπιορκέω indicates the taking of an oath within the context of giving testimony, most probably in the setting of a court of law, and then speaking falsely. The latter part of Matt 5:33 contains the clause ἀποδώσεις δὲ τῷ κυρίῳ τοὺς ὅρκους σου, literally ‘you shall render to the Lord your oaths.’ The noun ὅρκος ‘oath’ refers to any utterance of swearing. What determines the precise meaning of the clause is its verb ἀποδώσεις and the qualifying dative phrase τῷ κυρίῳ. The verb clearly denotes some manner of giving, with the dative indicating the recipient. There can be little doubt that this is using ‘oath’ in the sense of ‘vow.’ LXX texts affirm that this self-same verb and the accompanying dative are found in connection with the more precise Greek term for ‘vow,’ namely εὐχή:

52. BDAG, s.v. “ἐπιορκέω.” 53. Cf. NCV: “everyone who makes false promises”; GNT: “everyone who tells lies under oath.” 54. NRSV and NJB, recalling that the other major English versions do not translate the Apocrypha.

74

Biblical and Ancient Greek Linguistics 7 When you make a vow to the Lord [κυρίῳ] your God, you shall not delay to pay [ἀποδοῦναι] it. (Deut 23:21 [23:22 LXX]) I will pay [ἀποδώσω] my vows to the Lord [τῷ κυρίῳ]. (Ps 116:18 [115:9 LXX]) Today I pay [ἀποδίδωµι] my vows. (Prov 7:14) And they will make vows to the Lord [τῷ κυρίῳ] and perform [ἀποδώσουσιν] them. (Isa 19:21)

Also see Pss 22:25 (21:26 LXX); 49:14 (50:14 LXX); 61:8 (60:9 LXX); 66:13 (65:13 LXX); Eccl 5:3; Nah 1:15. Besides the same verb as Matt 5:33 and the same dative phrase in several of these texts, there is the further fact of the genitive of possession: ‘my vows,’ as compared with ‘your oaths.’ The evidence, then, for Jesus’s words being understood in terms of making vows is strong, strong enough for a number of English versions to place “vows” in the text (e.g., NASB, NRSV, NLT, NET).55 If this is so, we have moved from oath-taking in a legal setting, namely speaking the truth, for which commentators point to Lev 19:12, to a promissory vow to God, an act to be performed, for which the usual reference cited is Num 30:2. For such a development the connector δέ is very appropriate in its non-contrastive sense. A further distinction between the two kinds of swearing is highlighted by the fact that whereas the latter is ‘to the Lord,’ as shown, the former is typically ‘in the name of the Lord’ (as Lev 19:12; 1 Kgs 22:16; Isa 48:1; Zech 5:4). A final corroboration for the non-contrastive nature of Matt 5:33 is the parallel formula in 5:21. In explicating the law, Jesus uses the ‘You have heard . . .’ formula several times (vv. 21, 27, 33, 39, 43). The only other case that shows two elements connected by δέ is v. 21: Οὐ φονεύσεις· ὃς δ᾽ ἂν φονεύσῃ, ἔνοχος ἔσται τῇ κρίσει. The source for the first part is most certainly the fifth commandment, “You shall not murder” (Exod 20:13; Deut 5:17), while the source of the latter part is unknown, being possibly derived from oral tradition. No matter what the exact sources might be, what we are looking at here are two separate

55. Cf. Turner, Matthew, 172.

LUNN Categories of Contrast

75

sayings connected by δέ, and that connector is here invariably represented in modern versions by and. I would contend, therefore, that we are looking at the same basic features in 5:33, that is, two sayings of separate origin connected by and. In both contexts, v. 21 and v. 33, there is a development from the first element to the second, though remaining within the same general domain. Some commentators have perceived this relation. The Expositor’s Greek Testament rendered the connector as and,56 while in his classic New Testament commentary Lenski translated it as moreover,57 both of which are suitably noncontrastive. There would appear to be, then, some justification in concluding that the range of modern versions, as well as a good many commentators, is mistaken at this point. Both Brannan and Runge make serious and commendable attempts at getting to grips with the more detailed aspects of contrast in New Testament Greek. Their investigations into this area contain many new and helpful insights. However, there are also some significant deficiencies. Brannan, for example, has no notion of concessive contrast and so interprets a whole series of concessive sentences as replacing contrast, even when there is no preceding negative.58 While Runge’s study of εἰ µή and ἐὰν µή is enlightening, he does include a strange and unwarranted theological deduction of some considerable importance when analyzing the following Pauline statement: . . . εἰδότες δὲ ὅτι οὐ δικαιοῦται ἄνθρωπος ἐξ ἔργων νόµου ἐὰν µὴ διὰ πίστεως Ἰησοῦ Χριστοῦ (Gal 2:16) . . . knowing that a man is not justified by the works of the law but through faith in Jesus Christ. (NASB)

56. Bruce, “Matthew,” 110. 57. Lenski, Commentary on the New Testament, 1:235. 58. E.g. Brannan (“The Discourse Function of ἀλλά,” 18) offers an explanation of the ἀλλά in Mark 14:28 in terms of replacement or correction when it doubtless falls into our category of concession, discussed earlier. He makes the same error with regard to ἀλλά in 1 Cor 3:6 (p. 10), Mark 9:21–22 (p. 16), 1 Cor 6:11 (p. 23), and Phil 3:7 (p. 26).

76

Biblical and Ancient Greek Linguistics 7

Here Runge correctly identifies an exceptive contrast,59 as indicated by the connection using ἐὰν µή. If we had found ἀ$ά instead, then this would have been replacement, which would be readily comprehensible, and many English versions give the impression that this is what the Greek actually says. This would have made the apostle declare that justification is not by works of the law but through faith in Christ, a doctrine we are all familiar with. Yet it is plainly exceptive rather than replacing, as Runge rightly states, though the conclusion he grounds upon this is quite remarkable. This instance of exceptive contrast clearly contains a primary proposition that includes universal quantification. The topic covers every human being, to which the predicate applies in its negated form. We need therefore to consider the idea of what belongs to the conceptual set of the initial proposition. Looking at Paul’s words, the set in question has to be justification by the works of the law. This being so, ‘through faith in Jesus Christ’ has to belong to that set for the exceptive relation to be meaningful. In other words, in the apostle’s mind there is first the hypothetical set of ways to be justified by the law, and then the uniquely applicable member of that set, namely having faith in Jesus. The theological consequence of this is that it makes faith in Christ the way in which one can be justified by works of the law, not the alternative to justification by works of the law. In other words, justification by legal works is not replaced, but is rather rendered possible through faith.60 The problem is that Paul elsewhere explicitly excludes works of the law from the equation (e.g., Rom 3:28, ‘a man is justified by faith apart from works of the Law’; cf. Rom 9:11, 32; Gal 3:11; 2 Tim 1:9). So how is the difficulty resolved? While I believe Runge’s analysis of the logical relation holds good, in his reconstruction of the hypothetical set within the first proposition he overlooks

59. Runge, “Meaningful Distinction between ἀλλά and εἰ µή, Pt. 2,” (blog), January 20, 2013 60. On this Runge (“Meaningful Distinction between ἀλλά and εἰ µή, Pt. 2,” [blog], January 20, 2013) comments: “To be blunt, I think a lot of reformed folks would have preferred Paul has used ἀλλά, but he didn’t” (italics original).

LUNN Categories of Contrast

77

the crucial fact that the verb δικαιόω has various senses. Perhaps its dominant sense in the New Testament is ‘justify, acquit, declare righteous,’ but it also bears the meaning ‘make free.’61 This is how the verb is to be understood in Rom 6:7, where Paul states, “For anyone who has died has been set free [δεδικαίωται] from sin.” We also note Acts 13:39, in which we find the clause “through him everyone who believes is set free [δικαιωθῆναι] from all those things from which you could not be freed by the law of Moses.” Here the passive verb δικαιωθῆναι has been rendered “is set free,” a sense which is endorsed by numerous English versions (e.g., ESV, NRSV, NASB, NLT, CEV, GNT). And it is important to be aware that the author of Acts is here attributing these words to the same apostle, Paul, who authored Galatians.62 On the basis of these foregoing texts, therefore, we argue that ‘justify from’ can have the meaning ‘set free from.’ Such a sense fits admirably into the context of our text in Gal 2:16, along with its strictly exceptive contrast. So, rather than the conceptual set being how to be justified by works of the law, it is in fact the quite different matter of how to be set free from works of the law, that is, as a way of religious life and means of gaining divine acceptance. To this the exceptive clause offers the sole solution, itself now being part of the specified conceptual set according to the logical requirement, which is ‘through faith in Jesus Christ.’ Understood in this way, all makes good sense and is entirely in keeping with Pauline statements elsewhere. Further misconstruals of contrastive expressions surface elsewhere and the reader is invited to consider, among other texts, Rom 6:17 (not concession, as some English translations, but contrasting focus); 1 Cor 7:19 (sometimes interpreted as exceptive, but actually replacing contrast with ellipsis); 2 Cor 8:11 (contrasting focus, overlooked by the majority); 2 Cor 8:17

61. BDAG, s.v. “δικαιόω”; also Longenecker, Romans, 169. 62. Cf. also in the LXX: “nor will a trader be freed [δικαιωθήσεται] from sin” (Sir 26:29). It may further be observed that δικάζω, a close cognate of δικαιόω, bears the same sense, as in 1 Sam 24:15(16 LXX): καὶ δικάσαι µοι ἐκ χειρός σου, where such versions as ESV, NASB, NIV, NKJV, NET all translate the verb by “deliver,” and NJB, NLT, by “rescue.”

78

Biblical and Ancient Greek Linguistics 7

(not additive contrast, as in several major versions, but noncontrastive); Gal 2:20ab (not replacing contrast, but contrasting topic, with an extraposed Χριστός moved to the end of the second clause [right-dislocation], matching the marked position of ἐγώ in the first clause); 1 Pet 1:12 (often translated as replacement, but in fact contrasting focus, cf. Phil 3:1). Numerous other similar instances no doubt exist. By way of conclusion, then, it is hoped that the description of contrastive clauses presented here in some measure contributes towards a deeper comprehension of the language in which the New Testament is written. And this in itself can lead to a higher degree of precision in translation, and then in turn to greater accuracy in the exegetical task. Bibliography Bailey, Nicholas A. “Thetic Constructions in Koine Greek: With Special Attention to Clauses with εἰµί ‘Be,’ γίνοµαι ‘Occur,’ ἔρχοµαι ‘Come,’ ἰδού/ἴδε ‘Behold,’” Ph.D. diss., Free University of Amsterdam, 2009. Online: http://dare.ubvu.vu.nl/bitstream/handle/1871/15504/4727.pdf ?sequence=4&isAllowed=y. Bergs, Alexander, and Laurel J. Brinton. English Historical Linguistics: An International Handbook. Vol. 2. Handbooks of Linguistics and Communication Science 34.2. Berlin: de Gruyter, 2012. Brannan, Rick. “The Discourse Function of ἀλλά in Non-Negative Contexts.” Paper Presented at the Evangelical Theological Society’s 60th National Meeting, Providence, RI, November 20, 2008. Online: https:/ /www.academia.edu/535637/The_Discourse_Function_of_ ἀλλά_In_Non-Negative_Contexts. Bruce, A.B. “Matthew.” In Expositor’s Greek Testament, Vol. 1: The Synoptic Gospels and the Gospel of St. John, edited by William Robertson Nicoll, 61–340. Grand Rapids: Eerdmans, 2002. Dik, Simon C. The Theory of Functional Grammar, Part 1: The Structure of the Clause. Edited by Kees Hengeveld. 2nd and rev. ed. Functional Grammar Series 20. Berlin: de Gruyter, 1997. Köstenberger, Andreas J., et al. Going Deeper with New Testament Greek: An Intermediate Study of the Grammar and Syntax of the New Testament. Nashville, TN: Broadman & Holman, 2016.

LUNN Categories of Contrast

79

Lambrecht, Knud. Information Structure and Sentence Form: Topic, Focus and the Mental Representations of Discourse Referents. Cambridge Studies in Linguistics 71. Cambridge: Cambridge University Press, 1994. Larson, Mildred L. Meaning-Based Translation: A Guide to Cross-Language Equivalence. Lanham, MD: University Press of America, 1984. Lenski, R.H.C. Commentary on the New Testament: Vol. 1, Matthew’s Gospel. 1961. Reprint, Peabody, MA: Hendrickson, 2001. Levinsohn, Stephen H. “Adverbial Participial Clauses in Koiné Greek: Grounding and Information Structure.” Paper presented at the International Conference on Discourse and Grammar, University of Ghent, Belgium, May 23–24, 2008. Online: http://www-01.sil.org/~levinsohns/ GkParticipialClauses.pdf. ———. Discourse Features of New Testament Greek: A Coursebook on the Information Structure of New Testament Greek. 2nd ed. Dallas: SIL International, 2000. Long, Fredrick J. Koine Greek Grammar: A Beginning-Intermediate Exegetical and Pragmatic Handbook. Wilmore, KY: GlossaHouse, 2015. Longenecker, Richard N. The Epistle to the Romans: A Commentary on the Greek Text. NIGTC. Grand Rapids: Eerdmans, 2016. Mathewson, David L., and Elodie Ballantine Emig. Intermediate Greek Grammar: Syntax for Students of the New Testament. Grand Rapids: Baker Academic, 2016. Rudolph, Elisabeth. Contrast: Adversative and Concessive Expressions on Sentence and Text Level. Research in Text Theory 23. Berlin: de Gruyter, 1996. Runge, Steven E. “Meaningful Distinction between ἀλλά and εἰ µή, Pt. 1,” NT Discourse, January 6, 2013, http://www.ntdiscourse.org/2013/01/ meaningful-distinction-between-ἀλλά-and-εἰ-µή-pt-1/. ———. “Meaningful Distinction between ἀλλά and εἰ µή, Pt. 2,” NT Discourse, January 20, 2013, http://www.ntdiscourse.org/2013/01/ meaningful-distinction-between-ἀλλά-and-εἰ-µή-pt-2/. ———. “Where Does Contrast Come From?” NT Discourse, August 20, 2011, http://www.ntdiscourse.org/2011/08/where-does-contrast-come-from/. ———. Discourse Grammar of the Greek New Testament: A Practical Introduction for Teaching and Exegesis. Peabody, MA: Hendrickson, 2010.

80

Biblical and Ancient Greek Linguistics 7

———. “Teaching Them What NOT to Do: The Nuances of Negation in the Greek New Testament.” Paper Presented at the Evangelical Theological Society Annual Meeting, San Diego, CA, November 13–16, 2007. Steedman, Mark. Taking Scope: The Natural Semantics of Quantifiers. Cambridge, MA: MIT Press, 2012. Turner, David L. Matthew. BECNT. Grand Rapids: Baker Academic, 2008. Wallace, Daniel B. Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament. Grand Rapids: Zondervan, 1996.

[BAGL 7 (2018) 81–106]

COHESIVE HARMONY ANALYSIS FOR ANCIENT GREEK: SELPAP I:112 AND PMICH VIII:491 AS A TEST CASE John J.H. Lee McMaster Divinity College, Hamilton, ON, Canada Abstract: Ruqaiya Hasan’s Cohesive Harmony Analysis (CHA) is a useful tool to quantifiably predict the degree of the reader’s perception of the coherence of an English text. This work adopts and reconfigures her ideas to make them applicable to ancient Greek texts. This article then applies the modified version of Hasan’s CHA to investigate and compare the degrees of the perceived coherence of two family letters written in the second century AD. Based on the textual analyses, the conclusion is drawn that CHA is a promising tool to quantifiably predict the degree of coherence of ancient Greek texts. (Article) Keywords: Cohesion, coherence, cohesive tie, cohesive chain, cohesive harmony analysis, ancient Greek.

1. Introduction In this work, I explore the application of Ruqaiya Hasan’s Cohesive Harmony Analysis (CHA) to ancient Greek texts. Hasan proposed CHA in 1984 as a means of quantifiably predicting the perception of the coherence of texts by human readers. Being primarily geared toward modern languages and texts (e.g., English), CHA has not been actively applied to ancient languages such as Hellenistic Greek.1 In this work, therefore, I seek to provide an answer to the following question:

1. There do exist some works. See, for example, Reed, “1 Timothy,” 131–47; Land, Integrity of 2 Corinthians, 48–81.

82

Biblical and Ancient Greek Linguistics 7

Can CHA be used to predict the perceived coherence of ancient Greek texts? I argue, in the present work, that the notions of cohesive tie, cohesive chain, and cohesive harmony proposed within Systemic Functional Linguistics (SFL)2 provide a powerful tool, both in modern English and in ancient Greek, to quantifiably explicate how much coherence a reader will perceive in a given text. In what follows, I discuss the most fundamental tenets of CHA, although the presentation of the theory reflects my own modification of it to make it properly capture the unique features of ancient Greek. I then apply the methodology to two texts. Lastly, I conclude that CHA in its modified version is a promising interpretive tool for investigating and objectively predicting degrees of coherence for ancient Greek texts. 2. Methodology 2.1 Cohesion and Coherence Halliday proposes that there are three universal components that make it possible for human language to perform its main functions: the ideational, the interpersonal, and the textual metafunctions.3 A language’s textual metafunction is used to realize the textual shape of the language used in a discourse. Simply put, the textual metafunction is “the resources the language has for creating text”4 of which cohesion constitutes a part.5

2. Both M.A.K. Halliday and Ruqaiya Hasan have proposed these notions. 3. For good discussions of metafunction, see Halliday and Hasan, Cohesion, 26–30; Thompson, Functional Grammar, 28–29; Halliday and Webster, Essential Halliday, 308; Halliday, Functional Grammar, 30–31. Due to the scope of the present study, I exclude the discussion of the ideational and the interpersonal metafunctions. 4. Halliday and Hasan, Cohesion, 27. 5. Halliday and Hasan, Cohesion, 27. It should be noted that cohesion alone cannot properly create a text. It is “not a sufficient condition for the creation of text” (Halliday and Hasan, Cohesion, 298–99). The entire textual metafunction is responsible for the generation of the text, including cohesion,

LEE Cohesive Harmony Analysis

83

What then is cohesion? Halliday and Hasan define cohesion as “the set of semantic resources for linking a sentence with what has gone before.”6 Cohesion, therefore, mainly concerns sentence linking—its role is to “express the continuity that exists between one part of the text and another.”7 As for coherence, I employ Hasan’s definition that it refers to the property of “hanging together”;8 that is, if a text is perceived as coherent, “the patterns of language manifests—or realizes—the existence of semantic bonds.”9 Hasan also confirms that, if we examine the patterns (i.e., wordings) of language, we can find the semantic bonds that exist “between parts of his [or her] utterance.”10 So, the reader arrives at the ascertainment of the coherence of a text by observing the lexico-grammatical realizations of the semantic resources of cohesion in the text. Simply put, therefore, the fundamental goal of CHA lies in predicting the perception of the degree of coherence by human readers. 2.2 Formation of Cohesive Ties When cohesion occurs or is used, what we see at the semantic level is the occurrence of ties. For example, in I saw John. He was chasing a squirrel, the He in the second clause presupposes the preceding John. This semantic linking between John and He is a tie.11 There are two types of ties: componential tie and organic tie.12 Componential ties concern composing message(s) whereas

“information structure, thematic patterns and the like” (Halliday and Hasan, Cohesion, 299). 6. Halliday and Hasan, Cohesion, 10. 7. Halliday and Hasan, Cohesion, 299. The authors (Halliday and Hasan, Cohesion, 19–20) distinguish cohesion as a system and cohesion as a process. 8. Hasan, “Cohesive Harmony,” 181. 9. Hasan, “Cohesive Harmony,” 183. 10. Hasan, “Cohesive Harmony,” 185. 11. Halliday and Hasan (Cohesion, 3) speak of a tie as “one occurrence of a pair of cohesively related items.” 12. For a good treatment of these concepts, see Reed, Philippians, 89– 101. See also Halliday and Hasan, “Text and Context,” 49.

84

Biblical and Ancient Greek Linguistics 7

organic ties pertain to logically organizing the composed messages.13 Since the terms of a componential tie constitute message(s),14 the terms are normally one of the three types of semantic relations: co-reference, co-extension, and coclassification.15 Co-referential relations take place when two lexical items refer to the same entity. The most significant cohesive device by which co-referential relations are expressed is reference. Reference is “the relation between an element of the text and something else by reference to which it is interpreted in the given instance.”16 I propose the following as seven major reference devices in Greek: naming, equivalence, semblance, metaphoric expression, verbal morphology, pronouns, and the article. Naming is referring to a thing or person by a specific label or name (e.g., Ἀπίων “Apion”). Equivalence (i.e., A is B) in Greek is often expressed via εἰµί ‘to be.’ Semblance is like simile where the relationship between two lexical items is expressed via the use of, for example, ὡς ‘as, like.’ An example of metaphoric linking of two lexical items can be seen in Matt 5:13 Ὑµεῖς ἐστε τὸ ἅλας τῆς γῆς ‘You are the salt of the earth.’ The metaphoric linking of ‘you’ and ‘the salt’ is meaningful only in this context (instance), i.e., there is no guarantee that ‘you’ and ‘the salt’ are semantically related in the system of Greek. For this reason, Hasan treats it as instantial lexical cohesion. However, in the present work, all of Hasan’s instantial lexical cohesive devices (i.e., naming, equivalence, semblance, metaphoric expressions) are handled under co-reference because they assume the identity of referents. Greek allows more than one morpheme to occur per

13. Reed, Philippians, 89. Organic ties are normally related to conjunction. Due to the limited scope of the present paper, I do not discuss organic ties. For a systemic treatment of Greek conjunctions, see Porter and O’Donnell, “Conjunctions,” 3–14. 14. Halliday and Hasan, Language, Context, and Text, 81. See also Halliday and Hasan, “Text and Context,” 49. 15. Halliday and Hasan, “Text and Context,” 44–48. See also Reed, “1 Timothy,” 135; Reed, Philippians, 93–101. Note that I do not discuss coclassification in the present work. 16. Halliday and Hasan, Cohesion, 309.

LEE Cohesive Harmony Analysis

85

word, and, in Greek, one morpheme has more than one meaning attached to it.17 For example, the suffix -ω in the verb λέγω indicates the tense, voice, mood, and person and number of the grammatical subject (i.e., ‘I say’). This “morphological intensity”18 of Greek necessitates that we make a small modification to Hasan’s methodology for it to be adapted to Greek. Another reference device involves pronouns. I define a pronoun as a lexical item that is used as a reference device to create a co-referential tie with another exophoric or endophoric term. Lastly, the Greek article, too, is a significant device for expressing co-referential relations. The use of the Greek article, however, should not simplistically be equated with that of the English article for, as Porter rightly suggests, a Greek author can choose to use an article not because of the need to mark definiteness but for some other reasons. Porter claims in Greek an article means that “the substantive may refer to a particular item, or it may represent a category of items.”19 When co-extension occurs, both fully-grammaticalized lexical items (A and B) refer to “different things,” but the referents normally belong to “the same semantic field.”20 In the following example, Amy drives a Ford, but I don’t even have a car, the tie between a Ford and a car is co-extensional because it is apparent that both terms do not refer to the identical thing but to different entities that belong to the same semantic field, i.e., motorized vehicle. However, co-extensional ties may occur in a referential environment. For example, in Amy drove a Ford. But she wanted to sell the car, we know a Ford and the car share the same referent, i.e., Amy’s Ford. This tie may ostensibly seem to be a co-extensional tie because it has two fully-grammaticalized lexemes (i.e., Ford and car). The presence of the reference item (the), however, makes it a co-referential tie. In this work,

17. Greek is normally classified as a synthetic and fusional language (see Payne, Exploring Language Structure, 190–91). 18. Porter, “Further Modeling,” 10. 19. Porter, Idioms, 103–4. For a thorough treatment of the Greek article, see Porter, Idioms, 103–14. 20. Halliday and Hasan, “Text and Context,” 48.

86

Biblical and Ancient Greek Linguistics 7

therefore, I view this kind of tie as co-referential mainly because of the accompaniment of a reference device, the article. The representative device to express co-extensional relations is lexical cohesion.21 I define lexical cohesion as a cohesive device utilizing vocabulary and their sense relations in the linguistic system. Lexical cohesion subsumes the following sense relations: synonymy, antonymy, hyponymy, and meronymy. Synonyms are defined as phonologically different words that have “the same or very similar meanings,” e.g., car and vehicle.22 As for antonymy, Saeed points out that there are different types of opposition under antonymy: simple antonyms (e.g., dead/ alive), gradable antonyms (e.g., hot [warm tepid cool] cold), reverses (e.g., come/go), converses (e.g., employer/employee), and taxonomic sisters (e.g., red orange blue . . . brown).23 Hyponymy refers to “a relation that holds between a general class and its sub-classes.”24 The general class is the superordinate and the sub-class the hyponym(s). For example, the superordinate animal has cat, dog, donkey, etc. as its hyponyms. So, we can say that donkey is “semantically related” to animal as a hyponym.25 Meronymy is “a part-whole relation.”26 For example, head, torso, limbs are meronyms of body. Since I have discussed so far how cohesive ties are formed, I now turn, in the following section, to demonstrate how these cohesive ties form chains.

21. For Halliday and Hasan’s treatment of lexical cohesion, see Halliday and Hasan, Cohesion, 274–92. 22. Saeed, Semantics, 65. A blunt definition of synonymy, however, will negatively affect the analysis because an unguided use of the notion of synonymy can be so over-extended that it may hinder objective assessment of synonymy of lexemes. To obviate this problem, I suggest that the analysis be based within a pre-established framework (e.g., the Louw-Nida lexicon), instead of depending on the analyst’s subjective intuition. 23. Saeed, Semantics, 66–69. 24. Halliday and Hasan, Language, Context, and Text, 80. 25. Halliday and Hasan, Language, Context, and Text, 80. 26. Halliday and Hasan, Language, Context, and Text, 81.

LEE Cohesive Harmony Analysis

87

2.3 Formation of Cohesive Chains A cohesive chain refers to a cumulative occurrence of cohesive ties. In the following example, we can extract a cohesive chain that refers to Mr. Mitchell Y. McDeere, consisting of five coreferential ties: himself (8)~he (5) [tie 1], he (5)~He (4) [tie 2], He (4)~His (3) [tie 3], His (3)~He (2) [tie 4], and He (2)~M. Mitchell Y. McDeere (1) [tie 5]. 1: At 10 A.M. on a Friday, limo stopped on Front Street and Mr. Mitchell Y. McDeere emerged. 2: He politely thanked the driver, and watched the vehicle as it drove away. 3: His first limo ride. 4: He stood on the sidewalk next to a streetlight and admired the quaint, picturesque, yet somehow imposing home of the quiet Bendini firm. 5: It was a far cry from the gargantuan steel-and-glass erections inhabited by New York’s finest or the enormous cylinder he had visited in Chicago. 6: But he instantly knew he would like it. 7: It was less pretentious. 8: It was more like himself.27

There are two types of cohesive chain: identity chains (IC) and similarity chains (SC).28 The identity chain members’ semantic relation is that of “co-referentiality,” and the members of a similarity chain are bound together either by co-classification or by co-extension.29 2.4 Analysis of Cohesive Harmony Since a mere counting of IC’s and SC’s within a text does not shed much light on our understanding of its coherence, the notion of chain interaction has been proposed. Hasan writes, “Cohesive harmony consists not only in the formation of ICs and SCs but also in the creation of that additional source of unity which is provided by chain interaction.”30 We know there is chain interaction “when two or more members of a chain stand in an identical functional relation to two or more members of another chain.”31

27. Grisham, The Firm, 18. 28. Hasan, “Cohesive Harmony,” 205. See also Halliday and Hasan, Language, Context, and Text, 70–96. 29. Hasan, “Cohesive Harmony,” 205–6. 30. Hasan, “Cohesive Harmony,” 216. 31. Hasan, “Cohesive Harmony,” 212.

88

Biblical and Ancient Greek Linguistics 7

In order to discuss chain interaction, however, we first need to define the notion of token. A token refers to a lexical item (typically a word). There are, therefore, three tokens in the following clause: Samuel, returned, and home, and generally speaking, the total number of tokens refers to the total number of words. Unfortunately, when it comes to the cohesive harmony analysis of Greek texts, this rather simplistic notion of token needs to be refined due to the nature of Greek’s verbal morphology. For example, in !"#$% . . . "6/ µ7% "8%.$% &9:;µ2# ‘Apion . . . Most of all, I pray,’ since Greek marks the grammatical subject on the verb, both !"#$% “Apion” and the verb &9:;µ2< ‘I pray’ are tokens in the Apion chain. Apion chain An[wv

Pray chain -------

c s ux_c>µa ,

EUX9J~C!l :::=:::?_

In my proposed framework, therefore, the lexeme &9:;µ2< belongs simultaneously to Apion chain and to the Pray chain, and this is why the notion of a token as a single linguistic item does not guarantee a successful analysis in Greek. In order to capture that not the whole of the lexeme (&9:;µ2-"2-2< ?2"#.$%[2] ";@A ‘Greet Kapiton very much’). Another chain interaction occurs between the Apion chain and the Brother chain (.;B) CD&1E;F) [µ];G ‘my brothers’). 15 ,66H-I2# -& &9:;µ2< ‘I pray you are well’ is a prototypical way of ending Greco-Roman letters of that era, where two interactions are observable: first, the Apion and Wish chains interact with each other (‘I [Apion] pray’); second, the Epimachos and Healthy chains are seen in interaction (‘you are (be) well’). The chart below outlines the flow of CSI in SelPap I 112. CSIFlow:SelPap1112 3.5% 3.0%

2.5% 2.0%

1.5% 1.0% 0.5% 0.0%

-0.5% 0 -1.0%

l

4

6

8

10

12

14

16

18

Figure 5. CSI Flow in SelPap I 112

Below is the overview of the chain interaction among the seven major chains in the text. Each rectangular box indicates a chain. The name of the chain is written in each box. The relative difference in the vertical length of the box depicts each chain’s significance (i.e., CSI); the longer the vertical length of the rectangle, the more significant the chain is (e.g., Apion CSI = 3.9 percent; Wish CSI = 1.3 percent). The arrow indicates that there is chain interaction (e.g., see a in Figure 6 below). If the interaction takes place at clause level, the arrow is a solid-line. But if the interaction is found at word-group level, the arrow is a dashed line (e.g., see d in Figure 6 below).46

45. Note that, for convenience’s sake, I have divided the sentences into manageable units and have given them verse numbers. For a precise analysis, however, it is advised that the division be made according to clause boundaries. 46. This idea of distinguishing clause-level and word-group level

LEE Cohesive Harmony Analysis

97

Ap ion W ish

I ,:.... I Figure 6. Chain Interaction in SelPap I 112

It looks clear that the Apion chain is the most significant chain in the text. This chain is an identity chain that refers to the author of the letter. It actively interacts with the Wish chain (see a in Figure 6). The primary reason that Apion writes this letter is to wish his father well-being and this is well captured by this interaction. Also, this interaction not only opens the letter (2) but also (15) closes it. This interaction is diagrammed as below. Apion chain __ E_Qx_C?_µctl (#2)

£A7rl~W (# 11)

Wish chain

-~g_xCJ_µm (#2) £A7rl~W (# 11)

-~g_xCJ_µm (# 15)

There is also interaction between the Epimachos chain and the Healthy chain (see b in Figure 6). What Apion wishes or prays in the letter is the well-being of his father, Epimachos. His wish is reflected in this interaction where Epimachos functions as the Senser of the Phenomenon of being healthy and prosperous.

interactions is credited to Dr. Stanley E. Porter. I agree with him that wordgroup level interactions may not contribute to cohesion as much as clause-level interactions do for their span is too localized.

98

Biblical and Ancient Greek Linguistics 7 Epimachos chain

(#2) a-E(# ll)

a'E

Healthy chain

uytalVE[V (#2)

_§_epwa-0at (# 11)

Greeting, too, is an important topic in Apion’s letter, which is indicated in the chain interaction between the Much and Greet chains (see c in Figure 6). The tokens in the Much chain function as the Circumstance in which the act of greeting takes place; that is, this interaction evinces that Apion is trying to express that his Greet chain

.X~l_pE!V (# 1) cfo1raaa[ (# 12)

Much chain

7rAE'i'crra (# 1) 7rOAACI. (# 12)

greeting is sincere and wholehearted. I have so far identified three clause-level interactions (see a, b, c above). There are word-group level chain interactions in the text as well, e.g. .;J CD&1E;J µ;G ‘my brother’ (3); .H% CD&1EH% µ;G ‘my brothers’ (9);47 .;B) CD&1E;F) [µ];G ‘my brothers’ (12). This word-group level interaction, however, does not carry much Brother (#3) -rou aoEA