Biblical and Ancient Greek Linguistics [6 — 2017] 9781532642654, 1532642652

Citation preview

Biblical and Ancient Greek Linguistics Volume 6 — 2017 Contents JONATHAN M. WATT From Adams (1885) to Zimmermann (2009): In, With, and Under the Substance of Prepositions STANLEY E. PORTER Greek Prepositions in a Systemic Functional Linguistic Framework LAURENȚIU FLORENTIN MOȚ Semitic Influence in the Use of New Testament Greek Prepositions: The Case of the Book of Revelation JACOB BULLOCK Identifying Barriers to Understanding: Using Hill’s Matrix to Examine Contextual Mismatch in Acts 12:15 RYDER A. WISHART Monosemy in Biblical Studies: A Critical Analysis of Recent Work

5

17

44

67

99

STANLEY E. PORTER AND ANDREW W. PITTS The Perfect Tense-Form, the Son of Man, and John 3:13, Once More

127

Ancient Sources Index Modern Authors Index

137 143

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 that exists to further 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 spon-soring of conferences and symposia open to scholars and stu-dents researching in Greek linguistics who are interested in con-tributing to advancing the discussion and methods of the field of research. BAGL is a refereed on-line and print journal dedicated to distributing the

4

Biblical and Ancient Greek Linguistics 6

results of significant research in the area of linguistic theory and application to biblical and ancient Greek, and is open to all scholars, not just those connected to the Centre and the OpenText.org project. Accepted pieces are in the first instance posted on-line in page-consistent pdf format, and then (except for reviews) are published in print form each volume year. This format ensures timely posting of the most recent work in Greek linguistics with consistently referencable articles then available in permanent print form. 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 which can be 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 to the author for revision before being considered by the editors. 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 © 2016 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: 978-1-5326-4265-4 www.wipfandstock.com Manufactured in the U.S.A.

[BAGL 6 (2017) 5–16]

FROM ADAMS (1885) TO ZIMMERMANN (2009): IN, WITH, AND UNDER THE SUBSTANCE OF PREPOSITIONS Jonathan M. Watt Geneva College, Beaver Falls, PA, USA Abstract: A common closed-class feature of languages, prepositions connote spatial and logical relationships, often (though not always) preceding the noun to which they specify that relationship. Their use is highly idiomatic to a given language, such that their meaning may be best connoted by something other than a general translation equivalent. It is widely theorized that case-marking historically preceded the rise of prepositions, though in Hellenistic Greek (as in earlier forms of English) these have been employed simultaneously. Cross-linguistic consideration of this basic feature of language can be a helpful step toward understanding the role of prepositions in the Greek of the New Testament. (Article) Keywords: Greek, ad/pre-positions, cross-linguistic, diachronic, mono/poly-semy.

1. Introduction Consider the humble preposition: workhorse of spatial, temporal, and conceptual relations and possessing considerable diachronic consistency, it is nevertheless regarded by its many casual users as little more than conversational landfill, and despite its customary monosyllabic conciseness it can be upstaged by the insidious micro-precision of inflections. Undergraduates who can define nouns, verbs, and adverbs in their sleep actually stumble when it comes to defining a preposition—a plight hardly alleviated by one linguistically-informed English grammar this writer consulted a few years back which defined prepositions as “words that begin a prepositional phrase.” Perhaps just a little

6

Biblical and Ancient Greek Linguistics 6

more needs to be said about these compact artifacts of natural language. Prepositions find their place among what de Jonge and Ophuijsen call “the most influential linguistic doctrine to survive from antiquity,”1 namely, the eight µέρη λόγου ‘parts of speech,’ alongside nouns, verbs, participles, articles, pronouns, adverbs, and conjunctions. They proliferate throughout Indo-European languages, though other language families certainly have them as well. Dionysius Thrax (170–90 BCE) is credited with being the first to conceive of them as constituting a distinct category;2 what Greeks classified as πρόθεσις ‘fore-placement’ was rendered prae-positio in Latin. As sometimes happens in a world punctuated by delicious irony—in which an otherwise unknown Frenchman decodes Egyptian hieroglyphics, and an irascible Russian philosopher can smoke away an entire manuscript and still command the attention of generations of scholars—one of the handiest angles on prepositions today comes from an unlikely source: from practitioners of elementary education come such handles as Everywhere a cat can go—and then some and Everything you can do with a cloud. Wisdom flows from the mouths of babes and, if J.S. Stewart is right, even “religion resides in the prepositions.”3 Prepositions are part of the broader landscape of adpositions, whose three subtypes all express some sort of relationship between nominals. One subtype, postpositions, shows up in languages like Hindi, Turkish, Korean, and Japanese following the ‘head noun,’ i.e. the defining word of a phrase. A few languages employ the second subtype, actually dividing the adposition morpheme and placing one piece of it before and the other after the head noun to produce a circumposition. The third subtype, prepositions in the strictest sense (from Latin prae ‘before’ and ponere ‘to put’), is the most common form of adposition, being so ubiquitous that the term often is used

1. de Jonge and Ophuijsen, “Greek Philosophers on Language,” 495. 2. Bortone, Greek Prepositions, 4. 3. Stewart, Man in Christ, 154–55. He is commenting on Deissmann at that point.

WATT Substance of Prepositions

7

inclusively to designate all three types. Typologically, prepositional languages (in the strict sense) are head-initial and right-branching (cf. postpositional ones are head-final and therefore left-branching), and that tendency either way correlates with other syntactic features of the language in question. Whatever the syntactic tendency, prepositions (in the broader sense) typically pair up with one complement which is itself a noun/nominal or, in some cases, a “determiner” phrase, and the resultant prepositional phrase (PP) relates one nominal to another (e.g. He walked with the dog) and sometimes does so with adverbial implications (e.g. He walked with confidence), establishing a relationship on grammatical/structural and semantic levels.4 Prepositions are classified as closed class lexical items, along with definite articles, quantifiers and particles. Closed class items fluctuate minimally in number and meaning, appearing to be more resistant to diachronic semantic change than nouns, verbs, or adjectives. Prepositions in many languages, such as English, determine the case of the complement (e.g. to her instead of to she) or, like Greek, have meanings interconnected with the case of the head noun (e.g. διά + accusative = ‘according to,’ yet διά + genitive = ‘through’). In some languages a preposition can blend with a definite article (e.g. German: im < in dem); Greek has something similar, involving vowel deletion, e.g. διὰ αὐτόν → δι᾽ αὐτόν. Broadly construed, prepositions indicate relationships, being functional cousins to conjunctions, adverbs, particles, and relative pronouns, among others.5 Given these overlapping purposes, Greek prepositions are often said to have derived from early adverbs—a claim that finds a parallel, at least in English,

4. Some other types of adpositions include ambipositions, which can appear either before or after complements, as in English He worked right through the shift and He worked the shift right through, as well as inpositions, which occur between pieces of complex complements, and interpositions, as in They searched alley by alley, the last being roughly analogous to infixational morphology, which is not common to natural languages. 5. Beale et al., Interpretive Lexicon, 6–7.

8

Biblical and Ancient Greek Linguistics 6

where certain prepositions function as particles in two-word or three-word verbs (e.g. burn up, hunt down; run in to, meet up with). The opposite tendency, it seems, is for a preposition to become stranded (e.g. I hate being talked about, but I love being thought of), which Winston Churchill is said to have quipped was “a bad habit, up with which I shall not put.” Prepositions can be prefixed orthographically (e.g. Hebrew b- ‘in/with’, k- ‘like’ and l- ‘to’) and postpositions can be suffixed (e.g. Latin mecum ‘me-with’)—apparent insurance policies against becoming stranded. Nevertheless, this offers no guarantee against the inevitability of grammatical role shifts: an English preposition can function adverbially (e.g. ago), or both spatially and adverbially (He spoke before the court and It happened before the end of the week), or as a conjunction and spatially (He found it after she lost it and He laid down his Harley after the stop sign), while some Mandarin prepositions reportedly moonlight even as verbs. Furthermore, they may participate in crosslinguistic influence, a preposition from a speaker’s second language being calqued on the idiomatic use of a similar one present in the first language, as happened with ἀπό under the influence of Hebrew min in Deut 1:29 (LXX), “do not be afraid of them” (ἀπ᾽ αὐτῶν).6 As if that were not enough complexity troubling the world of prepositions, there remains the historic issue of case-marking ‘versus’ employment of a preposition. Synthetic languages (such as Old English/Anglo-Saxon, 449–ca. 1100 CE) tend to express relationships via case marking while analytical ones (such as what began to emerge in the Middle English period, ca. 1100–ca. 1500 CE) tend to defer the task to prepositions. Greek, like earlier forms of English, allows both options and in many specific instances permits interchangeable forms with no discernible difference of meaning (e.g. περιεπάτουν εἰς τὸν οἶκον / περιεπάτουν τῷ οἴκῳ)—and furthermore combines these two morphological features so that the meaning/function of a preposition is in some way tied to the case of the head noun. Additionally, a preposition may call for a specific case—a 6.

George, “Jewish and Christian Greek,” 270.

WATT Substance of Prepositions

9

functional redundancy seen in Septuagint Greek, which preserves both the use of dative case and also dative combined with various prepositions,7 as well as modern English pronouns (The visitor spoke with him, not with he). Conundrums populate this field like rabbits in a grassy paddock. Prepositional usage is notoriously idiomatic to each language and poses no end of headaches for second-language learners. In spite of this fact, or possibly because of it, prepositions are frightfully polysemous: a single one may carry a literal meaning and another by metaphoric extension (e.g. He lay under the table, being under the influence), or both spatial and temporal meanings (e.g. He stood behind the desk as he ran behind schedule), or spatial and conceptual meanings (e.g. He sat silently within the courtroom, his refusal to speak being well within his rights). And this is just the beginning of a convoluted picture. Prepositions may be directional (He walked in the door) or static (He sat in the chair), with clarification coming only from physical or conceptual context. Bortone observes that prepositional polysemy “verges on the chaotic.”8 However, neither syntax nor context offers any guarantee of semantic clarification: of may carry possessive, authorial, or legal implication in That is a book of his. And by could relay spatial or instrumental information (e.g. The victim was felled by the post), even as it fails to clarify whether the PP modifies verb or object. As if to add insult to injury, preposition usage sometimes varies between dialects: e.g. American English has on the weekend and in the hospital while British English has at the weekend and in hospital. Then, there are the many verbs that require prepositions even when their synonyms do not (e.g. He looked at me but He observed me) while some changes in preposition carry little discernible semantic difference (e.g. He spoke with/to me and He looked at/toward me).

7. Horrocks, Greek, 57–59; George, “Jewish and Christian Greek,” 271. 8. Bortone, Greek Prepositions, 41, citing Taylor, Linguistic Categorization.

10

Biblical and Ancient Greek Linguistics 6 2. Greek Prepositional Literature

In Greek, and some other languages, a distinction is often made between proper/essential prepositions in contrast with improper/ accidental ones, the latter kind appearing to have been borrowed from other parts of speech. However, when referring to Ancient Greek, ‘improper’ prepositions specifically are those which never appear as a verb prefix, being forty-two in number by most counts. Harris lists them,9 and cites Bortone in support of the observation that they tend to be longer than proper ones, tend to replace older ones, and typically bear spatial, rather than temporal or logical, senses.10 In contrast, ‘proper’ prepositions can be prefixed to verbs, and are commonly numbered at eighteen. However, these labels are cross-linguistically confusing, functionally irrelevant for the most part, and terminologically unfortunate as they seem to imply a preferential value judgment, at least for ‘good’ speech.11 Prepositions themselves, not to mention the terminology generated to codify and contain them, constitute a minefield of obstacles and opportunities. Grammarians who search for the heuristic key to this Pandora’s box must be rigorous and creative, and a certificate in alchemy wouldn’t hurt. Contributions to the field relevant to Ancient Greek include the following: F.A. Adams (1885) was one voice in the chorus that proclaims prepositions to be “suggestive primarily of notions of space,”12 explaining that “[t]he present work is an endeavor to clear somewhat this seeming jungle of the Greek Prepositions—to show that it is not a jungle, but a garden, whose alleys and paths have become overgrown through neglect, and lost to view.”13 With spatial notions established, one “seeks for the analogues of

9. Harris, Prepositions and Theology, 241–42. 10. See Bortone, Greek Prepositions, 194. 11. Moule offers this number, and others count similarly. He offers that “prepositions were originally adverbs, which in turn, may have originally been nouns crystallized indeclinably in one particular case” (Idiom Book, 48), and offers that ἀντί and ἐπί carry once-dative, and χάριν accusative, suffixes. 12. Adams, The Greek Prepositions, v. 13. Adams, The Greek Prepositions, v.

WATT Substance of Prepositions

11

these in human experience. Thus the whole field of human life, of thought, passion, and purpose, is laid open, and the Prepositions enter it in their own right.”14 In fact, Adams posits that all words, not just prepositions, “are derived largely from notions of things in space,”15 though he qualifies that “[n]o class of words in the Greek is more important than Prepositions; and none are more imperfectly understood . . . ”16 He sees something primal in them, surmising that “[a]s the ideas of space and the notions these carry with them were always present, it is reasonable to believe that they were operative in the formation of language from the first . . . ”17 Since verb tense and aspect might also comprise a formidable jungle, and nouns do seem to hold some semantic import, it can be acknowledged that Adams’s refrain echoes into the present day in the form of interest in primary cognitive categories. Herbert Weir Smyth’s Greek Grammar first appeared in 1920, yet even in its 1956 reprinting the editor’s preface claimed it was “by far the most complete reference grammar of ancient Greek to appear in English.”18 Identified there as “descriptive, not an historical, nor a comparative grammar,” the author states that “[o]riginally the preposition was a free adverb limiting the meaning of the verb, but not directly connected with it.”19 Smyth states that all prepositions specifically originated as adverbs of place. Subsequently, the adverb “was brought into closer connection with the noun . . . ”20 and the emergent preposition would “define the relations of a substantival notion to the predicate,” though he seems to overlook their potential relation specifically to a subject phrase. Smyth was an early advocate of the term ‘improper’ prepositions, i.e. those which “do not form compounds” and are never prefixed to verbs.21 14. 15. 16. 17. 18. 19. 20. 21.

Adams, The Greek Prepositions, v. Adams, The Greek Prepositions, iii. Adams, The Greek Prepositions, iii. Adams, The Greek Prepositions, vi. Smyth, Greek Grammar, iii. Smyth, Greek Grammar, 365. Smyth, Greek Grammar, 365. Smyth, Greek Grammar, 388.

12

Biblical and Ancient Greek Linguistics 6

A.T. Robertson’s (1934) 1454-page tome also seems to be a suitable possessor of the ‘most complete grammar’ moniker. His chapter on prepositions, approximating the length of a short monograph,22 contrasts with Adams in its assertion that cases preceded prepositions “both in time and at first in order” and historically did not govern the case of the head noun “but rather helped to define more precisely the distinctions indicated by the case forms.”23 The accusative originally denoted direction, the genitive marked separation, and the dative indicated location, according to Robertson, and what later developed was that “cases found in prepositions a convenient means of sharpening their significance”24—an interesting anthropomorphizing of grammatical ‘intent’ seemingly apart from cognitive, sociological, or pragmatic context (unless he had intended it only in a metaphoric sense). Stanley Porter (1992), who identifies himself as holding a monosemic bias, describes prepositions as “indeclinable fixed forms or particles used to enhance the force of . . . cases . . . ”25 and notes the stubborn fact that their diminutive number and immutability tendency do not prevent them from performing many different functions. He too comments on the suspiciously close relationship between prepositions and adverbs:26 the former tend to modify nouns and the latter, verbs and other modifiers— hence the inference that prepositions historically developed from adverbs.27 Porter cites ὑπὲρ ἐγώ ‘I more [than they]’ in 2 Cor 11:23 as an example of adverbial use of a preposition. It is best, he contends, to say that “a preposition is governed by its case, in some way helping the case to manifest its meaning and to perform more precisely its various functions.”28 He notes that in

22. Robertson, Grammar of the Greek New Testament, 553–649. 23. Robertson, Grammar of the Greek New Testament, 28. 24. Robertson, Grammar of the Greek New Testament, 28. 25. Porter, Idioms, 139. Emphasis mine. 26. Porter, Idioms, 125–27. 27. Murray Harris notes that “[i]n the parent Indo-European language, cases probably stood alone, but later some adverbs came to be used as prepositions” (Prepositions and Theology, 27). 28. Porter, Idioms, 140. Emphasis his.

WATT Substance of Prepositions

13

addition to comprising a prepositional phrase (customarily followed by a head noun), they may intensify meaning, be transformative of meaning, or just semantically retentive.29 He upholds a consensus to the effect that one-time spatial meanings became extended metaphorically over time, but may have developed so far from their earlier, literal usage that a speaker would use a particular preposition simply because it was customary, or idiomatic, and not because he necessarily made a correlation with an earlier or original denotation; he adds, “[m]any of these extensions are far removed from their basic sense.”30 Silvia Luraghi’s (2003) monograph, subtitled The Expression of Semantic Roles in Ancient Greek, offers a Cognitive Grammar approach which reaches beyond the ‘localistic approaches’ of the nineteenth century which “thought that the meaning of cases derived by abstraction from an original spatial meaning”31 and beyond European Structuralism with its ‘meaning-oriented’ approach. Cognitive Grammar posits that “grammatical forms are conceived as meaningful. The substance of their meaning is not different from the substance of lexical meaning: the difference lies in the degree of abstractness, rather than in substance. Furthermore, space is conceived as the basic domain of human experience, which serves as source for understanding other, more abstract domains.”32 The approach posits that semantic roles are prototypical categories, and assumes a localistic theory of the meaning of cases as well as prepositions.33 Luraghi’s study aims to provide “a synchronic account of the meaning of each case, considering cases as instances of structured polysemy that developed diachronically by means of motivated semantic extensions” in Ancient Greek.34 She continues her argument with a diachronic study of the

29. 30. 31. 32. 33. 34.

Porter, Idioms, 140–41. Porter, Idioms, 142. Luraghi, Prepositions and Cases, 11. Luraghi, Prepositions and Cases, 11–12. Luraghi, Prepositions and Cases, 18. Luraghi, Prepositions and Cases, 49.

14

Biblical and Ancient Greek Linguistics 6

eighteen ‘proper’ prepositions (noting that in Homer they can be used as free adverbs)35 and concludes, for example, that prepositions gradually become less spatially focused over time.36 Pietro Bortone’s Greek Prepositions: From Antiquity to Present (2010), reflecting previous work, reaches for “the evolution of the Greek prepositional system in its entire history”37 and specifically states that its “aim is not to formulate an abstract theory, nor to describe Greek usage strictly within a preconceived theoretical framework . . . ”38 Bortone argues that the history of Greek prepositions is “entirely congruent with the ‘unidirectionality hypothesis’ that spatial meanings evolve into non-spatial ones, but not vice-versa.”39 He continues: Considering the varying degrees of meaningfulness of case forms, from semantic emptiness to extreme polysemy, it can however be argued that polysemy is not unprincipled—there are clear and demonstrable links between, for example, locative, comitative, and instrumental/modal meanings. The ‘localist’ interpretation of this is that the various non-spatial senses of cases or adpositions are related in a chain of semantic extensions which starts from a spatial notion.40

Bortone suggests that “[t]he fact that cases and prepositions become less predictable and often less ‘meaningful’ when their sense is not spatial . . . fits well with the conjecture that the basic meaning is spatial and that other meanings are metaphors.”41 Murray Harris’s Prepositions and Theology (2012) expands upon his own previously published work and aligns in various ways with Bortone’s. Harris is heavily context-driven when it comes to their NT meanings,42 locating prepositions amongst what he calls the ‘big four’ zones of Greek grammar (along with the aorist, genitive case, and articles) “that produce the most

35. Luraghi, Prepositions and Cases, 76. 36. Luraghi, Prepositions and Cases, 315. 37. Bortone, Greek Prepositions, xii. 38. Bortone, Greek Prepositions, xii. 39. Bortone, Greek Prepositions, xii. 40. Bortone, Greek Prepositions, xiii. 41. Bortone, Greek Prepositions, 43. 42. Harris quotes Robertson as saying “prepositions in composition often best show their original import” (Prepositions and Theology, 14).

WATT Substance of Prepositions

15

handsome dividends when special attention is given to understanding them.”43 Harris construes a tight relationship between prepositions and theological concepts, and cautions against any assumption that Classical distinctions necessarily hold in the Hellenistic period, and against disregard for the NT writers’ stylistic uniquenesses,44 individually discussing the forty-two improper (though Porter numbers these at fifty) and what Harris numbers at seventeen, namely, the proper prepositions. Ilse Zimmermann (2003) investigates Russian case marking with respect to sound-meaning correlation, drawing on Roman Jakobson’s mid-twentieth century work on characteristics of cases and Dieter Wunderlich’s more recent Decomposition Grammar Theory. Her technical study of this Indo-European language concludes cautiously that “[c]ases of noun phrases can have semantic import. No case is a structure or semantic case per se.”45 To many theorists, case marking long prefigured the use of prepositions, though how Russian grammar may shed light on Greek prepositions is something yet to be identified; for the purposes of this paper, it at least facilitates handily an ‘A to Z’ scope of the literature survey. What is clear is the fact that the problematics of prepositions have been discussed in, with, and under the rubrics of the literature, and solutions to matters such as the relationship between case marking and prepositions, the productivity of the monosemy versus polysemy debate, and the potential role of linguistic frameworks, are all tasks awaiting the theorist. Bibliography Adams, F.A. The Greek Prepositions: Studied from their Original Meanings of Designations of Space. New York: D. Appleton, 1885.

43. Harris, Prepositions and Theology, 14. 44. See Harris, Prepositions and Theology, 39–44. 45. Zimmermann, “Semantics of Cases,” 371.

16

Biblical and Ancient Greek Linguistics 6

Beale, G.K., Daniel J. Brendsel, and William A. Ross. An Interpretive Lexicon of New Testament Greek. Grand Rapids: Zondervan, 2014. Bortone, Pietro. Greek Prepositions: From Antiquity to the Present. Oxford: Oxford University Press, 2010. George, Coulter H. “Jewish and Christian Greek.” In A Companion to the Ancient Greek Language, edited by Egbert J. Bakker, 267–80. Malden, MA: Wiley-Blackwell, 2010. Harris, Murray J. Prepositions and Theology in the Greek New Testament. Grand Rapids: Zondervan, 2012. Horrocks, G.C. Greek: A History of the Language and Its Speakers. London: Longman, 1997. Jespersen, Otto. Essentials of English Grammar. Tuscaloosa, AL: University of Alabama Press, 1964. de Jonge, Casper C., and Johannes M. van Ophuijsen. “Greek Philosophers on Language.” In A Companion to the Ancient Greek Language, edited by Egbert J. Bakker, 485–98. Malden, MA: Wiley-Blackwell, 2010. Luraghi, Silvia. On the Meanings of Prepositions and Cases. Philadelphia: Benjamins, 2003. Moule, C.F.D. An Idiom Book of New Testament Greek. Cambridge: Cambridge University Press, 1982. Porter, Stanley E. Idioms of New Testament Greek. Biblical Languages: Greek 2. Sheffield: Sheffield Academic, 1992. Robertson, A.T. A Grammar of the Greek New Testament in Light of Historical Research. Nashville: Broadman, 1934. Smyth, Herbert Weir. Greek Grammar. 4th ed. Cambridge: Harvard University Press, 1966. Stewart, James Stuart. Man in Christ: The Vital Elements of St. Paul’s Religion. 1935. Reprint, London: Forgotten Books, 2012. Taylor, J.R. Linguistic Categorization: Prototypes in Linguistic Theory. Oxford Textbooks in Linguistics. Oxford: Clarendon, 1989. Zimmermann, Ilse. “On the Semantics of Cases.” In Syntactic Structures and Morphological Information, edited by Uwe Junghanns and Luka Szucsich, 341–80. Berlin: Mouton de Gruyter, 2003.

[BAGL 6 (2017) 17–43]

GREEK PREPOSITIONS IN A SYSTEMIC FUNCTIONAL LINGUISTIC FRAMEWORK Stanley E. Porter McMaster Divinity College, Hamilton, ON, Canada Abstract: Greek prepositions belong to a class of words that are usually called particles. These function words are morphologically invariable and enable their function by indicating some kind of relationship between larger units. This means that prepositions are part of a larger category of words that include not only prepositions but conjunctions, adverbs, and possibly other lexemes. Systemic Functional Linguistics does not have an explicit theory of the preposition. However, prepositions are important within both syntagmatic and paradigmatic structure, and function at various ranks and as components of various structures at those ranks. In this paper, I discuss five topics regarding prepositions: word groups and phrases, types of prepositions, prepositions and other relators, the meaning of prepositions, and the function of prepositional groups within SFL architecture. (Article) Keywords: Preposition, Greek, Systemic Functional Linguistics, conjunction, adverb.

1. Introduction1 Systemic Functional Linguistics (SFL) does not have an explicit theory of the preposition. This does not mean that prepositions are not interesting or important in a SFL framework. To the contrary, prepositions provide an important test case for illustrating the importance of balancing form and function within

1. I wish to thank my colleague, Dr. Christopher Land, for discussion of this topic that has been extremely helpful in formulating my thoughts.

18

Biblical and Ancient Greek Linguistics 6

context. Prepositions are also important within both syntagmatic and paradigmatic structure, and function at various ranks and as components of various structures at those ranks. Within the Greek language, prepositions belong to a class of words that are usually called particles. These are function words—as opposed to content words (a distinction to which I will return below, if only to explore its ambiguity)—that are morphologically invariable, even if they are situationally phonologically variable, and play a role in enabling the function by indicating some kind of relationship between larger units.2 In that sense, prepositions are part of a larger category of words that include a number of sub-categories. In Greek, these subcategories include so-called (see below) proper prepositions, improper prepositions (these two categories the ostensive subject of this paper), conjunctions, and even some adverbs, among possibly some other lexemes. In order to discuss prepositions—or at least to begin such a discussion—I will treat five topics: word groups and phrases, types of prepositions, prepositions and other relators, the meaning of prepositions, and the function of prepositional groups within SFL architecture. 2. Word Groups and Phrases In SFL, one of the ranks is that of the group or, sometimes called, the phrase.3 This distinction—or not—is part of the discussion of the place of the preposition. In standard theory SFL, prepositions are not treated independently of being parts of prepositional phrases, with the preposition as the head and its complement or completive (typically, though not always, a noun group), as well as any modifiers.4 These prepositional phrases function similarly and at roughly the same rank as do word groups, that is, between the word and the clause. Halliday has

2. I realize that the notion of relationship is a problematic one. See Huddleston, Introduction to the Grammar of English, 336. 3. See Morley, “On Group,” esp. 217–20. 4. See Morley, Explorations in Functional Syntax, 80–81.

PORTER Greek Prepositions

19

traditionally made a distinction between prepositional phrases and (other) word groups, however, on the basis of what he contends is their distinct and even opposite origins (he also distinguishes a prepositional group as consisting of two or more prepositional phrases).5 For Halliday, the group is the expansion of the word, and hence a word group.6 As an example in English, the lexeme hope can be the headterm of a nominal group consisting of a minimum of one word, itself. An example within a clause, with the nominal group consisting of one word constituting the subject, would be: Hope is never to be abandoned. Many nominal groups consist of more than one word. These nominal groups consist of not only their headterm, but various other modifying words.7 English examples using hope might include:8 My hope, which consists of a headterm and a single definer premodifier My unfailing hope, which consists of a headterm and two definer premodifiers The unfailing hope of the ages, which consists of a headterm, a specifier, a definer (both premodifiers) and a postmodifier prepositional unit.

5. Halliday, IFG4, 423. 6. This has apparently been part of Halliday’s architecture from early on, and is retained in Halliday, IFG4, 362–63: “A phrase is different from a group in that, whereas a group is an expansion of a word, a phrase is a contraction of a clause. Starting from opposite ends, the two achieve roughly the same status on the rank scale, as units that lie somewhere between the rank of a clause and that of a word.” For a much earlier treatment, see Muir, Modern Approach to English Grammar, 50. 7. See Aarts and Aarts, English Syntactic Structures, 60–78, esp. 61–67, 67–68, and 68–71. 8. The terminology used here and throughout is from OpenText.org. We define the following terms: specifier (sp)—modifier that classifies or identifies, such as an article; definer (df)—modifier that attributes features or further defines, such as adjectives and appositives; qualifiers (ql)—modifier that limits or constrains, such as genitives or datives; and relators (rl)—a word specified by a preposition. These are recognizably different from those in standard theory SFL, of which there are several. See Morley, Explorations in Functional Syntax, 75–79, where the categories are modifiers (premodifiers) consisting of deictic (article, demonstrative, possessive, interrogative, relative), numerative, epithet, or classifier, and (postmodifier) qualifier.

20

Biblical and Ancient Greek Linguistics 6

The same occurs in the Greek of the New Testament: κτίσις (‘creation’), with the lexeme alone as headterm of the word group ἡ κτίσις (‘the creation’), with the headterm having a specifier πάση ἀνθρωπίνῃ κτίσει (‘all human creation’; 1 Pet 2:13), with the headterm having two definers as premodifiers πᾶσα ἡ κτίσις (‘all the creation’; Rom 8:22), with the headterm having a definer and a specifier as premodifiers πάσῃ κτίσει τῇ ὑπὸ τὸ οὐρανόν (‘all creation under heaven’; Col 1:23), with the headterm (within a prepositional unit) having one definer as premodifier and one postmodifier, consisting of a prepositional unit following a specifier.

One can see from these examples that the nominal group is the expansion of a word. However, Halliday sees the prepositional phrase as having different origins than the nominal or other groups, to the extent that prepositions, he contends, are functionally more closely related to verbs than to adverbials.9 Whereas the word group is the “expansion of a word,” the prepositional phrase (and hence the ostensive reason for the descriptive difference) is the “contraction of a clause.” This contracted unit consists of two elements, the preposition and the element that is related to the preposition, that is, the preposition and its complement or completive. Even though the prepositional phrase ends up being placed on the rank scale between the word and the clause—and hence roughly at the rank of the group—the configuration and hence function according to Halliday are different. The preposition serves as a minor verb or predicate, with the nominalized element as its complement, and hence its treatment as a contracted clause (that is, a minor clause, one without a finite verb).10 This theory of the preposition, though widely adopted in SFL because of the influence of Halliday, is not accepted by all, even within SFL. The results are several. Some, such as David Morley, wish to reject altogether the notion of group, because group seems to indicate the bringing together into a larger configuration a number of smaller units—just the opposite of

9. Halliday, IFG4, 423. 10. See Halliday and Webster, Text Linguistics, 111.

PORTER Greek Prepositions

21

how Halliday theorizes regarding the group (apart from the prepositional group).11 Others, such as Robin Fawcett, wish to use the terminology of group for both phenomena discussed above, even if they examine their structures similarly as do others in SFL.12 Fawcett is probably right that the term “group” is satisfactory for labeling the units that rank between word and clause, and that consist of both the prepositional group and (usually, though not invariably) the nominal group (note that I adopt the language of prepositional group, on the basis of this decision). There are a number of reasons for accepting this terminology, besides simplicity. One reason is that the placement on the rank scale is admittedly so similar, even for Halliday, so as to make no difference in structural significance. If the results were quantifiably different on the rank scale, then it would be useful to distinguish the two structures. A second reason is that the nominal group and the prepositional group are similar in structure. Both of them consist of a headterm with its appropriate modifiers. The headterm is modified in various ways by other constituents (SFL is considered a constituency grammar, but this raises questions regarding constituency and dependency; I am not convinced that there is a significant difference for the purposes of this paper). In the case of the prepositional group, this consists of a minimum of one required modifier, the preposition, with any of its own modifiers (in English: directly to the rear). In the Greek of the New Testament, this modifier is now pre-positioned (hence preposition), although in earlier Greek it could be postpositioned. In the OpenText.org model, the headterm of the prepositional group is the nominalized element, not the 11. Morley, “On Group,” 218–19, with the term ‘group’ reserved for a ‘word complex.’ See further his Explorations in Functional Syntax, 80–81; cf. 52–54. 12. Fawcett, Theory of Syntax, 204–206. Fawcett suggests that the prepositional group be called the ‘relator group,’ but that the term ‘preposition’ is so firmly enshrined as to resist this. Since we use relator to describe a wider range of words, this term is not useful for our description of Greek (especially when conjunctions are given group status in standard-theory SFL).

22

Biblical and Ancient Greek Linguistics 6

preposition (referred to as the preposition heading the phrase)—a difference from many if not most other linguistic frameworks, and discussed further below.13 A third reason is that the nominal group and the prepositional group are similar in function. I will take clausal structure as an example of how the cases and prepositions are functionally similar, but the same could be said of their functioning at group level (as elements of groups). Within clausal structure, some nominal groups function as the subjects of their clauses, while others function as complements. Those that function as complements in Greek are (except in relational clauses) found in the so-called oblique cases, indicating a functional difference within the clause as shown by the change of case. In the history of Greek, for reasons that we do not need to discuss here, these nominal groups functioned differently on the basis of their context, with the alteration in case indicating the different function. However, the case system was restricted, and expression of these differences in function was simply by means of morphology. Prepositions (many if not most of them perhaps originally adverbs) were used as modifiers (usually but not always as premodifiers) of the nominal group to enhance the use of the cases and to indicate different types of relations that went beyond those of case and to make it clearer how the nominal group functions within its larger (complement) structure. This is the function performed by the preposition, to a greater extent than cases.14 When the nominal group has this prepositional modification, it forms a prepositional group. In this way, the prepositional group is, like the nominal group, an expansion of the word. (I also note that, at least at the clausal level, the formal differences between the prepositional and nominal group are also indicated by a distinction in semantics, because the prepositional

13. See Huddleston, Introduction, 336. This includes SFL. See, e.g., Hudson, English Complex Sentences, 292. 14. See Porter, Idioms, 139; cf. 80–100 on case. Cf. Sgall, Bémová, and Hajcová, “Remarks on the Semantic Features,” esp. 71–73, where from a different framework they at least acknowledge the relationship between cases and prepositions.

PORTER Greek Prepositions

23

group is no longer a complement but an adjunct of the clause. This has semantic implications, in that complements indicate direct participants, but adjuncts indicate circumstances and indirect participants. This distinction is not true at the group level.) Therefore, the headterm of the prepositional group remains the nominal group, with the preposition serving as a relator, that is, a type of modifier that relates the function of the prepositional group within its appropriate structure. The prepositional group—which we have seen is best treated as similar to a nominal group—relates the nominal group (as head) to its structural environment whether it functions at the group or clause level. The unfailing hope of the ages, in which the prepositional group is a postmodifier (relator) of the headterm hope in the nominal group He entered into the room, in which the prepositional group is the adjunct of the clause. ὄλεθρον αἰώνιον ἀπὸ προσώπου τοῦ κυρίου καὶ ἀπὸ τῆς δόξης τῆς ἰσχύος αὐτοῦ (‘eternal destruction from the face of the lord and from the glory of his strength’; 2 Thess 1:9), in which there are two postmodifiers (prepositional groups) of the headterm ὄλεθρον καθίσατε ἐν τῇ πόλει (‘sit in the city’; Luke 24:49), in which the prepositional group is the adjunct of the clause.

In at least one respect, Halliday’s argument regarding the prepositional phrase as a minor clause is worth considering, because he recognizes that there is an important relational function of the preposition similar to that of other words labeled relators. Conjunctions are a type of relator. Conjunctions relate components at various ranks to each other. Similar functions occur in both English and Greek for the use of conjunctive relators—they relate elements at various levels. These include conjunctive relators that link wordgroups, clauses, clause complexes, and even larger units, such as paragraphs (or the equivalent). This function of relators may be found at the word level: big and boisterous person, where the two modifiers are joined by the relator πολλὰ καὶ ἄλλα σηµεῖα (‘many and other signs’; John 20:30), where the two modifiers are joined by the relator

24

Biblical and Ancient Greek Linguistics 6

or it may function at the group level: the big man and the tall woman καὶ ὁ ἄνεµος καὶ ἡ θάλασσα (‘both the wind and the sea’; Mark 4:41)

or it may function at the clause level: he bought and ate the meat τινες ἐσκληρύνοντο καὶ ἠπείθουν (‘certain ones were hardened and unpersuaded’; Acts 19:9)

or it may function at the clause complex level: he saw the boy and he smelled the food ἡ χαρὰ ἡ ἐµὴ ἐν ὑµῖν ᾖ καὶ ἡ χαρὰ ὑµῶν πληρωθῇ (‘my joy might be in you and your joy might be fulfilled’; John 15:11)

or it may function at the paragraph level (and possibly beyond): However we begin…, with a normal use of the conjunction that might be used at the beginning of a paragraph καὶ ἐγένετο ἐν ἐκείναις ταῖς ἡµέραις (‘and it came about in those days’; Mark 1:9), beginning the episode of John the Baptist after the opening section of Mark’s Gospel.

However, these conjunctive relators, whether in English or Greek, function at various ranks, only some of which are clausal. For Halliday, he must posit that his prepositional phrases as minor clauses function at the clause level but must rankshift to function at the group level (I am not sure how he sees them functioning at higher than the clause level). I believe that group expansion provides a better explanation, with the prepositional group functioning at various ranks as it fills higher levels of structure. There is the further problem for Halliday that minor clauses (clauses without a finite verb) are expandable with other elements of clause structure, but prepositional groups are not. Finally, whereas Halliday sees a parallel between prepositions and non-finite verbs, I find more compelling the parallel between prepositions and other types of relators, even if it is a cline of similarity and difference.15

15. Halliday, IFG4, 424–25.

PORTER Greek Prepositions

25

This placing of the prepositional group together with the nominal group is in distinction to the adjectival group (which is now being called the quality group in more recent OpenText.org discussion, following Fawcett’s Cardiff grammar) and to some extent the adverbial group (as discussed below).16 Whereas the prepositional group and nominal group have more in common with each other, they have some distinctions from the adverbial group and more distinctions from the adjectival group. Adjectival and adverbial groups do not take specifiers (or deictics in SFL or determiners in other frameworks), and hence do not, without such specifiers or within a noun group, function as subjects (etc.) of clauses.17 When an adjectival or adverbial group takes a specifier, it functions as a nominal group (or complex) with the group as headterm of a larger structure. Thus, we may have: The book is largely unknown, in which the adjectival group, largely unknown, consists of the headterm adjective, unknown, and the modifying adverb, largely ὁµολογουµένως µέγα ἐστὶν τὸ τῆς εὐσεβείας µυστήριον (‘confessedly great is the mystery of the godliness’; 1 Tim 3:16), in which the adjectival group, ὁµολογοθµένως µέγα, consists of the headterm adjective, µέγα, and the modifying adverb, ὁµολογουµένως. He ran quite quickly, in which the adverbial group, quite quickly, consists of the headterm adverb, quickly, and the modifying adverb, quite πολυµερῶς καὶ πολυτρόπως πάλαι (‘in diverse fashions and in diverse ways from long ago’; Heb 1:1), with the use of three adverbs, perhaps in two adverbial groups.

But then we might also have: The small and the mighty perform the greatest deeds, with the adjectival group filling the headterm of a nominal group with the specifier ὁ δὲ µικρότερος ἐν τῇ βασιλείᾳ τοῦ θεοῦ µείζων αὐτοῦ ἐστιν (‘the smaller one in the kingdom of God is greater than he’; Matt 11:11), where an adjectival group, µικρότερος ἐν τῇ βασιλείᾳ τοῦ θεοῦ, fills the role of the headterm of a nominal group.

16. Fawcett, Theory of Syntax, 206–207. 17. See Aarts and Aarts, English Syntactic Structures, 61–67, 67–68, and 68–71; Halliday, IFG4, 391.

26

Biblical and Ancient Greek Linguistics 6 In the dark backward and abysm of time (Tempest, I.2.50), where these two adverbs are used as the headterms of nominal groups within a prepositional group τὰ ἄνω ζητεῖτε (‘seek the above’; Col 3:1), where the adverb is the headterm of the nominal group with the specifier.

The use of the specifier changes the function of the group. Prepositions may appear with adverbs without the article, in which instances they are sometimes considered as separate words (prepositional groups with the adverb as the headterm) or sometimes as adverbial groups with a prefixed preposition. Examples include: ὑπὲρ ἐκ περισσοῦ or ὑπερεκπερισσοῦ (‘more than abundantly’; Eph 3:20 as conjunctive relator, but as adverb in 1 Thess 3:10; 5:13).18 I incidentally note that the verbal group in Greek is probably best analyzed as consisting of the simple or complex verb as head, whether that consists of an auxiliary/finite and head (as in periphrastic or catenative constructions) or simply the head (a single finite verb). The OpenText.org project has found it useful to have no further elements in the verbal group (and hence as the only element of a predicator), so that prepositional groups at clause structure are adjuncts of the clause, used to indicate circumstances and indirect participants (see below). Further, OpenText.org does not consider relators (or conjunctions) to be a separate word group, as do some other SFL frameworks, for many of the reasons that have already been intimated in the discussion of prepositions.19 However, we do recognize that prepositions can form groups or complexes of their own, in the sense that there are prepositional groups that consist of more than one preposition, and these prepositional groups form the relator element in a larger (complex) prepositional group.20 We could specify these as prepositional

18. See Robertson, Grammar, 547–48, although some of his discussion of adverbs appears to be highly misleading (e.g. on adverbs with other adverbs, where he appears to be speaking about adjectives with adverbs). 19. See, e.g., Halliday, IFG4, 77; Benson and Greaves, Language People Really Use, 13. 20. There is not much significant research on these types of prepositions or prepositional groups. See Morley, Explorations in Functional Syntax, 53,

PORTER Greek Prepositions

27

complexes, but do not do so. However, we do recognize that these prepositions form complex relational units. Thus, in conclusion to this first section, prepositions are relators, that is, words that relate one element to another, and they occur as relators with nominal groups to form prepositional groups, similar to nominal groups. 3. Types of Prepositions There are typically said to be two types of prepositions in Greek, proper (or essential) and improper (or accidental) prepositions.21 So-called proper prepositions are those that are prefixed to verbs, of which there are eighteen such prepositions in the Greek New Testament, and they consist of: ἀµφί, ἀνά, ἀντί, ἀπό, διά, εἰς, ἐκ, ἐν, ἐπί, κατά, µετά, παρά, περί, πρό, πρός, σύν, ὑπέρ, and ὑπό. The so-called improper prepositions are those that are not prefixed to verbs, of which there are around 50 or so in the Greek New Testament. They consist of: ἅµα, ἄνευ, ἄντικρυς, ἀντιπέρα, ἀπέναντι, ἅτερ, ἄχρι(ς), ἐ=ύς, ἐκτός, ἔµπροσθεν, ἔναντι, ἐναντίον, ἕνεκα, ἐντός, ἐνώπιον, ἔξω, ἔξωθεν, ἐπάνω, πέκεινα, ἕσω, ἕως, κατέναντι, κατενώπιον, κυκλόθεν, κύκλῳ, µέσον, µεταξύ, µέχρι(ς), ὄπισθεν, ὀπίσω, ὀψέ, παραπλήσιον, παρεκτός, πλήν, πλησίον, ὑπεράνω, ὑπερέκεινα, ὑπερεκπερισσοῦ, ὑποκάτω, χάριν, and χωρίς. Some, if not most, of these so-called improper prepositions are considered adverbs, as well as any number of other forms (sometimes compound forms, as noted above). Many believe that the preposition originated with the adverb, as evidenced by earlier usage (e.g. independent prepositions used for circumstances in Homer).

who gives four lines to “complex prepositions” consisting of “more than one word”; and Hoffmann, Grammaticalization, 1–3, 166–69, although in a very limited way and using grammaticalization theory to explain the phenomenon in English. 21. Cf. Morley, Explorations in Functional Syntax, 52–54. See also Watt, “From Adams (1885) to Zimmermann (2009),” 10, for the use of this alternative terminology. As the discussion illustrates, none of the terms is entirely satisfactory, as there is nothing either proper or essential to one set or that is improper or accidental to the other.

28

Biblical and Ancient Greek Linguistics 6

What is the difference between a proper and improper preposition? Nothing much. They are both relators, with the only difference being that some prepositions are preposed on verbs and others are not. There is a change in function of the preposition that occurs when it is prefixed to a verb, however, so that it is sometimes difficult to analyze the function of the verbal (or nominal) prefixed preposition. I will discuss the meaning of the prepositions, including the prefixed prepositions, below. Otherwise, their distribution and function is similar to each other and generally to the function of other relators. 4. Prepositions and Other Relators Prepositions have a number of similarities with other relators. As a result, the boundary between prepositions and other relators is not an easy one to find. Some of their similarities are as follows. First, prepositions and other relators function at different ranks of the language structure but in similar ways, as already noted above.22 There are both prepositions and other relators that conjoin material at the word level, group level, clause level, clause complex level, and possibly even higher. An example of a preposition functioning at the word level might include: ὑπὲρ ἐγώ (‘indeed I’; 2 Cor 11:23), with clear similarities to an adverbial function.

22. See Huddleston, Introduction, 336, who under “Functional potential, II,” states that (English) prepositional phrases “have a considerable variety of functions in larger constructions, notably: in clause structure, complement . . . or adjunct; in AdjP structure, complement . . . or modifier; in NP structure, complement . . . or modifier . . . ” Greek does not use the prepositional group in predicate structure, as the Greek verbal group consists only of the verbal element (whether simple or complex). The Greek prepositional group also extends beyond clause structure to relate clauses within clause complexes, as discussed below. See also Thompson, Introducing Functional Grammar, passim.

PORTER Greek Prepositions

29

An example of a preposition conjoining at the group level includes a prepositional group embedded within a nominal group as a relator, a very common usage: χάριν ἀντὶ χάριτος (‘grace instead of grace’; John 1:16), with the nominal group consisting of its headterm and the prepositional group.

An example of a preposition conjoining at the clause level includes a variety of examples. One example includes an embedded clause linked by a prepositional relator, such as a relative clause: ὁ τόπος ἐν ᾧ ἦσαν συνηγµένοι (‘the place in which they were gathered’; Acts 4:31), in which the prepositional conjunction relates the relative clause to the noun group to form a complex subject.

An example of a preposition conjoining at the clause complex level includes two clauses linked by a prepositional conjunction: . . . κυρίου ἡµῶν Ἰησοῦ Χριστοῦ, δι᾽οὗ νῦν τὴν καταλλαγὴν ἐλάβοµεν (‘. . . our Lord Jesus Christ, through whom now we receive reconciliation’; Rom 5:11).

An example of a preposition conjoining at the paragraph level involves opening a paragraph with a prepositional conjunction that relates to a previous paragraph: διὰ τοῦτο . . . (‘because of this’; e.g. Rom 5:12; 2 Cor 4:1), where the reference is anaphoric.

Second, there are a number of words for which classification is very difficult. Sometimes these words are called prepositions (usually improper prepositions), other times conjunctions, and still others adverbs, but the significance is that they have similar types of functions, regardless of what they are called. I will distinguish here between what might be called prepositional relator and conjunctive relator function and adverbial function. Some examples include:

30

Biblical and Ancient Greek Linguistics 6 πᾶσαι οὖν αἱ γενεαὶ ἀπὸ Ἀβραὰµ ἕως Δαυίδ (‘all the generations from Abraham to David’; Matt 1:16), the prepositional relator function23 ἕως ἐλθὼν ἐστάθη ἐπάνω οὗ ἦν τὸ παιδίον (‘until, arriving, it [the star] stood above . . . ’; Matt 2:9), the conjunctive relator function. ἄχρι τοῦ νῦν (‘until now’; Rom 8:22), the prepositional relator function ἄχρι ἧς ἡµέρας εἰσῆλθεν Νῶε εἰς τὸ κιβωτόν (‘until the day Noah entered the ark’; Matt 24:38), the conjunctive relator function. εἴτε ἐκτὸς τοῦ σώµατος οὐκ οἶδα (‘if I do not know apart from the body’; 2 Cor 12:2; cf. 1 Cor 6:18), the prepositional relator function ἐκτὸς εἰ µὴ διερµηνεύῃ (‘except if he interprets’; 1 Cor 14:5; cf. 1 Cor 15:2), the conjunctive relator function, with a conjunctive group. ἕνεκεν δικαιοσύνης (‘on account of righteousness’; Matt 5:10), the prepositional relator function οὗ εἵνεκεν ἔχρισέν µε εὐαγγελίσασθαι πτωχοῖς (‘on account of whom he anointed me to proclaim good news to the poor’; Luke 4:18), the conjunctive relator function. ὁ καθήµενος ἐπάνω αὐτοῦ (‘the one sitting above him’; Rev 6:8), the prepositional relator use ἐστάθη ἐπάνω οὗ ἦν τὸ παιδίον (‘it [the star] stood above where the child was’; Matt 2:9), a possible conjunctive relator use (although admittedly susceptible to other analyses) ἵνα ὦσιν ἓν καθὼς ἡµεῖς (‘so that they might be one as we’; John 17:11), with prepositional relator function, although this may be an instance of ellision καθὼς γέγραπται (‘as it stands written’; Matt 26:24), with conjunctive relator function. τὴν κώµην τὴν κατέναντι ὑµῶν (‘the village opposite you’; Matt 21:2; Mark 11:2; but cf. Luke 19:30: εἰς τὴν κατέναντι κώµην, ‘into the opposite village,’ where the word is the modifier within a nominal group), the prepositional relator function κατέναντι θεοῦ ἐν Χριστῷ λαλοῦµεν (‘oppositely, we speak of God in Christ’; 2 Cor 2:17; 12:19), a possible conjunctive relator use; cf. also Rom 4:17: κατἐναντι οὗ ἐπίστευσεν θεοῦ (‘opposite of which he believed God’), another possible conjunctive relator function. περιῆγεν τὰς κώµας κύκλῳ (‘he went around the villages in a circle’; Mark 6:6), adverbial function εἰς τοὺς κύκλῳ ἀγροὺς (‘into the surrounding fields’; Mark 6:36; Luke 9:12), a modifying function in a nominal group

23. I note that the Moulton-Geden-Marshall concordance differentiates ἕως conjunction from ἕως preposition.

PORTER Greek Prepositions

31

κύκλῳ τοῦ θρόνου (‘around the throne’; Rev 4:6; 5:11; 7:11), a prepositional relator function. µέχρι τῆς σήµερον (‘until today’; Matt 11:23), prepositional relator function µέχρις οὗ ταῦτα πάντα γένηται (‘until when all these things might occur’; Mark 13:30; cf. Gal 4:19), conjunctive relator function. κράζει ὄπισθεν ἡµῶν (‘cried after us’; Matt 15:23), prepositional relator function γέµοντα ὀφθαλµῶν ἔµπροσθεν καὶ ὄπισθεν (‘being covered with eyes before and after’; Rev 4:6), adverbial function. ὀψὲ δὲ σαββάτων (‘after sabbath’; Matt 28:1), prepositional relator function ἔρχεται ἢ ὀψὲ ἢ µεσονύκτιον (‘he comes either later or in the middle of the night’; Mark 13:35), adverbial function. οὐκ ἔστιν ἄλλος πλὴν αὐτοῦ (‘there is no other except for him’; Mark 12:32), prepositional relator function πλὴν λέγω ὑµῖν (‘except I say to you’; Matt 11:22, 24; 26:64), conjunctive relator function. νυκτὸς καὶ ἡµέρας ὑπερεκπερισσοῦ δεόµενοι (‘praying night and day far more’; 1 Thess 3:10), adverbial function ὑπὲρ πάντα ποιῆσαι ὑπερεκπερισσοῦ ὧν αἰτούµεθα (‘above all things to do far more than which things we ask’; Eph 3:20), probably conjunctive relator function. ὡς ἕνα ἕκαστον ὑµῶν ὡς πατὴρ τέκνα ἑαυτοῦ (‘as each one of you as a father his children’; 1 Thess 2:11), with prepositional relator function. ὡς ἐπαύσατο (‘as he finished’; Luke 11:1), with prepositional conjunctive function.

Third, prepositions and conjunctions have similar scope in relation to other elements. Prepositions, as noted above, have scope over minimal to maximal units. There are numerous places where the scope of the preposition extends over more than one unit within the language: κατὰ τὸ εὐαγγέλιόν µου καὶ τὸ κήρυγµα Ἰησοῦ Χριστοῦ, κατὰ ἀποκάλυψιν µυστηρίου χρόνοις αἰωνίοις σεσιηµένου φανερωθέντος δὲ νῦν . . . (‘according to my good news and the proclamation of Jesus Christ, according to the revelation of the mystery silent for eternal times but now manifested and . . . ’; Rom 16:25–26), with the second use of κατά extending its scope over several embedded clauses.

32

Biblical and Ancient Greek Linguistics 6

This extended scope is also found in the use of conjunctions, where the scope of the conjunction may extend over minimal units up to maximal units connecting paragraphs within a discourse: ἀποκαλύπτεται -γὰρ ὀργὴ θεοῦ ἀπ᾽οὐρανοῦ ἐπὶ πᾶσαν ἀσέβειαν καὶ ἀδικίαν ἀνθρώπων (‘for the wrath of God is revealed from heaven upon all ungodliness and unrighteousness of humans’; Rom 1:18), with the scope of the conjunction linking at least the opening of the body of the letter on the guilt of humankind to the rest of the letter.

Instances of this sort require differentiation of the scope of the function of the relator, whether prepositional or conjunctive. Fourth, some grammarians try to distinguish conjunctive and prepositional relators on the basis of clausal configurations. Thus, it has been argued that English subordinating conjunctions that conjoin a following clause and prepositions that do the same have different clausal patterning. As an example, an English clause such as because he had trained hard can become he had trained hard, but from what we understand cannot become what we understand and stand on its own.24 This explanation is not satisfactory in Greek, where at least some, although not all, prepositions and conjunctions are found in similar clausal environments up to the clause complex level, as the examples above demonstrate. 5. Meanings of Prepositions As mentioned above, prepositions are labeled not content words but function words—they perform the function of relating one unit to another. However, there is some ambiguity regarding the differentiation between function and content words. In the LouwNida lexicon, domain 89 is concerned with relations. This domain includes mostly particles such as prepositions and conjunctions, as Louw and Nida indicate.25 However, throughout 24. See Morley, Explorations in Functional Syntax, 56–57. Morley does entertain other examples that show their similarity in English, although he rejects these. 25. Louw and Nida, Greek-English Lexicon, 777 n. 1.

PORTER Greek Prepositions

33

the domain, there are other words included as well that would normally be thought of as content words, such as: αἰτία, κρέµαµαι, ἀναλογία, µετασχηµατίζω, θεµέλιον, ἀρχή, ῥίζα, λόγος, ἄλογος, δωρεάν, αὐτόµατος, ἀφορµή, ἔκβασις, τέλος ἀποβαίνω, ὀψώνιον, ἔρχοµαι, κενός, προστίθεµαι, πάλιν, περαιτέρω, σύµφυτος, and κοινός, besides some phrasal units. These so-called content words can be used in functional ways, ways in which they resemble function words. Within the lexicon, a number of prepositions and conjunctions are also included within other semantic domains that are usually thought of as content domains (e.g. the preposition ἐν is found in domains 13, 67, 83, 84, and 90, besides its being found in 22, 23, 28, 33, 65, 70, 67, 68, 87, 89, and 90 when used phrasally). These uses of the prepositions and conjunctions indicate that they have a certain amount of content when used in these contexts. There are, however, at least two specific environments where prepositions appear not to be function words but to be content words. One of these is when the preposition is itself the headterm of a nominal group. The other of these is when the preposition is prefixed to a verb. As relators, prepositions are words that convey positional status (a form of the localist hypothesis). Most prepositions are spatial locative indicators, such as toward, into, out of, away, above, below, upon, through, in front of, behind, and many more.26 Spatial location situates the element in a particular position in relation to another element or elements. There is also a close relationship between location in space and location in time, so much so that the sense of the preposition is modulated by context to indicate not simply spatial location but temporal location. As a result, such notions as in front become before, behind become after, through space become through time, and the like. Similarly, location in space leads to location in time, and

26. See Porter, Idioms, 142. Robertson, Grammar, 568, referring to Kühner and Gerth, Ausführliche Grammatik, I:451, believes in a ‘groundmeaning’ of the prepositions. He also articulates an analysis using the meaning of the case, then the preposition, and then the context, similar to the one in Porter, Idioms, 81–82, for case.

34

Biblical and Ancient Greek Linguistics 6

this leads to location in relation to another, that is, the relator function. In other words, space leads to time leads to relation. Positional before becomes temporal before becomes relational before (in priority or cause, etc.). In the same way that a conditional clause indicates a relationship between protasis and apodosis, the relator can indicate a relationship between two units, whether they are words, groups, clauses, or beyond. This is why prepositions, conjunctions, and even some adverbs have so much in common lexicogrammatically and semantically. They perform very similar functions in relating elements to each other on the basis of some type of location. This explanation of the meaning of prepositions may appear to have relations to various proposals in cognitive linguistics, and so it does. Cognitive linguistics has reinforced what we have already realized about how various elements are conceptualized in relation to each other, and how we metaphorically transfer or extend core meanings.27 However, this conception of the semantic relationship between Greek prepositions predates cognitive linguistics, as is seen in the conceptual understanding of the semantic space of prepositions in Bruce Metzger’s Lexical Aids for Students of the New Testament and in my own depiction of their semantic relations.28 The semantic overlap and interconnection among prepositions does not need to rely upon cognitive linguistics, however.29 They can also be explained as

27. Other frameworks have arrived at similar conclusions. For example, see Lindstromberg, English Prepositions Explained, esp. 7, who acknowledges his debt to Lakoff and Johnson, Metaphors We Live By and Lakoff, Women, Fire and Dangerous Things, the latter incorporating prototype theory; Bennett, Spatial and Temporal Uses; Herskovits, Language and Spatial Cognition, drawing on cognitive linguistics and artificial intelligence; Luraghi, Meaning of Prepositions and Cases, 82–314; Bortone, Greek Prepositions, 47–53; and (following Bortone) Harris, Prepositions and Theology, 28–30. An important essay in the history of cognitive linguistics is Lakoff, “Cognitive Semantics.” For the notion of space from a general cognitive and linguistic perspective, see Regier, Human Semantic Potential and Bloom et al., eds., Language and Space. 28. See Metzger, Lexical Aids, 80 (a chart that may go back as early as 1954 or even 1946); and Porter, Idioms, 143–79. 29. Cognitive linguistics, as well as prototype theory, relies upon theories of polysemy. See Taylor, Linguistic Categorization, 99–121, with 109–16

PORTER Greek Prepositions

35

lexical metaphorical (in the SFL tradition) expansion of the core meaning if viewed from a monosemic bias. The spatial, temporal, and hence relational elements of the preposition help us to understand that the prefixed preposition takes on some characteristics of a content word by how its locational, temporal, and even relational senses relate to the verb to which it is prefixed. This is clearly seen in verbs of motion, where we have εἰσέρχοµαι (‘go in’), ἐξέρχοµαι (‘go out’), etc., which add the locational sense of the prefixed preposition to the verb. The temporal and relational senses are less clearly seen, but are present nevertheless, even if the semantic features of the individual prefixed prepositions are difficult to estimate, although there are often signs that the locative sense is still present (e.g. κατεσθίω, ‘eat up’ or ‘chow down’; καταδιώκω, ‘hunt down,’ in which the preposition still maintains its idea of ‘ground’). Intensification and transformation of the meaning of the verb are two of the ways in which these other senses are expressed, although still often with local meaning.30 Instances in which the preposition has some characteristics of a content word are found when a preposition is itself the headterm of a nominal group.31 There are a number of prepositions that function in this way. λαµψάτω τὸ φῶς ὑµῶν ἔµπροσθεν τῶν ἀνθρώπων (‘let your light shine before people’; Matt 5:24), prepositional relator function προδραµὼν εἰς τὸ ἔµπροσθεν (‘running into the lead’; Luke 19:4), content function τὰ µὲν ὀπίσω ἐπιλανθανόµενος τοῖς δὲ ἔµπροσθεν ἐπεκτεινόµενος (‘losing sight “upon” the things “after” but reaching “out” “upon” the things “before”’; Phil 3:13), with several instances to consider. There are two instances of prepositions as the headterms of nominal groups (the things ‘after’ and the things ‘before’) and several uses of prefixed prepositions—all indicate the tendency for prepositions to modulate their spatial and temporal locations to indicate their content.

devoted to the preposition over; and 264–81, with recognition of the two-levels problem and treatment of in and round. On monosemy in a New Testament Greek framework, see Porter, Linguistic Analysis, 51–53. 30. See Porter, Verbal Aspect, 66–70. 31. See Peters, The Greek Article, 219–25.

36

Biblical and Ancient Greek Linguistics 6 Λάζαρε, δεῦρο ἔξω (‘Lazarus, come outside’; Matt 13:48), adverbial function προφήτην ἀπολέσθαι ἔξω Ἰεροσαλήµ (‘a prophet to die outside of Jerusalem’; Luke 13:33), prepositional relator function ἐν σοφίᾳ περιπατεῖτε πρὸς τοὺς ἔξω (‘in wisdom walk with those outside’; Col 4:5), content function ἕως ἔξω τῆς πόλεως (‘until outside of the city’; Acts 21:5; cf. also Acts 26:11: ἐδίωκον ἕως καὶ εἰς τὰς ἔξω πόλεις, ‘they were pursuing until and up to those outside cities,’ where the adverb functions as a modifier within a nominal group), content function ἔξωθεν µάχαι ἔσωθεν φόβοι (‘outside wars inside fears’; 2 Cor 7:5), adverbial function ἐπατήθη ἡ ληνὸς ἔξωθεν τῆς πόλεως (‘the wine press was trampled outside the city’; Rev 14:20), prepositional relator function καθαρίζετε τὸ ἔξωθεν τοῦ ποτηρίου καὶ τοῦ παροψίδος (‘purify the outside of the cup and the lip’; Matt 23:25; cf. Luke 11:39, 40), content function εἰσελθὼν ἔσω (‘entering inside’; Matt 26:58), adverbial function οἱ δὲ στρατιῶται ἀπήγαγον αὐτὸν ἔσω τῆς αὐλῆς (‘the soldiers took him away into the hall’; Mark 15:16), prepositional relator function ἠκολούθησεν αὐτῷ ἕως ἔσω εἰς τὴν αὐλήν (‘they followed him until inside into the hall’; Mark 14:54; cf. Eph 3:16: εἰς τὸν ἔσω ἄνθρωπον, ‘into the inside person,’ as the modifier within a nominal group; and Rom 7:22), content function µεταξὺ τοῦ ναοῦ καὶ τοὺ θυσιαστηρίου (‘between the temple and the altar’; Matt 23:35; cf. Luke 11:51), prepositional relator function ἐν τῷ µεταξύ (‘in the between place’; John 4:31), content function εἰς τὸ µεταξῦ (‘into the between place’; Acts 13:42), content function ῥυσάσθω νῦν εἰ θέλει αὐτόν (‘he should rescue him now if he wants him’; Matt 27:43), adverbial function νῦν δὲ πολλὰ µὲν µέλη (‘now [therefore?], there are many members’; 1 Cor 12:20), conjunctive relator function32 νῦν οὖν πορεύεσθε ἐν εἰρήνῃ (‘now therefore, go in peace’; Acts 16:36), conjunctive relator function καὶ τὰ νῦν λέγω ὑµῖν (‘and the present things I speak to you’; Acts 5:38; cf. Acts 4:29; 17:30; 20:32; 24:25; 27:22), content function ὀπίσω σαρκὸς ἑτέρας (‘after other flesh’; Jude 7), prepositional relator function ὁ ἐν τῷ ἀγρῷ µὴ ἐπιστρεψάτω ὀπίσω (‘don’t let the one in the field return

32. This and the following example are argued for by Thrall in her Greek Particles, 30–34. See Porter, Idioms, 213–14.

PORTER Greek Prepositions

37

after’; Matt 24:18), adverbial use; cf. Mark 13:16: ὁ εἰς τὸν ἀγρὸν µὴ ἐπιστρεψάτω εἰς τὰ ὀπίσω (‘don’t let the one in the field return into the after places’), content functions βλέπων εἰς τὰ ὀπίσω (‘looking at the after things’; Luke 9:62; cf. Luke 17:31; John 18:6; 20:14), content function παρεκτὸς λόγου πορνείας (‘apart from a word of adultery’; Matt 5:32; 19:9), prepositional relator function χωρὶς τῶν παρεκτός (‘apart from the exceptional things’; 2 Cor 11:28), content function πέραν τῆς θαλάσσης τῆς Γαλιλαίας τῆς Τιβεριάδος (‘opposite the sea of Galilee of Tiberias’; John 6:1), prepositional relator function ἦλθεν εἰς τὸ πέραν τῆς θαλάσσης (‘he entered into the otherside of the sea’; Mark 5:1), content function Συχὰρ πλησίον τοῦ χωρίου ὃ ἔδωκεν Ἰακώβ (‘Sychar, near the land which Jacob gave’; John 4:5), prepositional relator function ἀγαπήσεις τὸν πλησίον σου (‘you will love your nearperson [kinsperson]’; Matt 5:43; 19:19; 22:39; Mark 12:31, 33; Luke 10:27; Rom 13:9; Gal 5:14; Jas 2:8), content function εἰς τὰ ὑπερέκεινα ὑµῶν εὐαγγελίσασθαι (‘to preach the good news to those beyond you’; 2 Cor 10:16, but see above), content function τῶν παραβάσεων χάριν προσετέθη (‘on account of transgression it was added’; Gal 3:19), prepositional relator function δι᾽οὗ ἐλάβοµεν χάριν καὶ ἀποστολήν (‘through whom we receive grace and apostleship’; Rom 1:5), content function, as this is a noun.

6. Functions of Prepositional Groups within SFL Architecture In the fifth and final section of this paper, I wish to show how prepositional groups may variously function within SFL architecture. Prepositional groups can function at numerous ranks within the lexicogrammar, as already discussed and evidenced. The question here is what is the semantic potential of prepositional groups that other groups do not have. On the basis of what has been said above, it is clear that there is strong “functional overlap” to some extent with the adverbial group, but certainly between conjunctive relators (what Halliday calls the “conjunction group,” hence treatment with the other two groups) and the prepositional group. “They have the same

38

Biblical and Ancient Greek Linguistics 6

general functional potential.”33 However, having said that, there are also some differences between them that merit further attention regarding the prepositional group. There are two major distinctives of the prepositional group as opposed to the others that are worth mentioning as to their semantic potential. The first distinctive is that, “since prepositional [groups] include a nominal group” as the headterm (but as the complement for Halliday, who takes the preposition as the head), “they have greater expressive potential than adverbial groups.”34 This is because of the expressive potential of the nominal group. The adverbial group, by contrast, consists of the adverbial headterm and its modifiers, which are limited in number and scope. In fact, there are relatively few adverbial groups within the Greek of the New Testament that instantiate complex modification of the headterm. By contrast, the nominal group, as noted above, consists of the headterm and a number of potential pre- and post-modifiers. These include specifiers (such as articles, demonstrative pronouns, and the like), qualifiers (such as adjectives), definers (such as defining genitives and datives), and relators (such as prepositional groups). As in English, such usage in Greek is found in “the more elaborated registers” of writing, and constitutes one of the predominant ways in which prepositional groups are used.35 The expressive potential, therefore, of the prepositional group is immense, as the semantic potential contained within these elements is almost inexhaustible, especially with the prepositional group forming groups of nested units. Examples of such constructions that demonstrate the instantiation of such potential can be found in Eph 1:5–14, in which there are numerous prepositional groups, some of them serving as adjuncts within clausal structure and others as components of nominal groups expanding their semantic scope.

33. Halliday, IFG4, 363. 34. Halliday, IFG4, 363. 35. Halliday, IFG4, 331.

PORTER Greek Prepositions

39

The second distinctive is that prepositional groups “can construe more experientially complex circumstances.”36 The prepositional group may function within the textual metafunction as providing thematic material and within the interpersonal metafunction as providing participant status. However, the prepositional group, because of the complexity already described, has greater expressive potential than the adverbial group. The adverbial group tends to realize the circumstances of time, location, and manner,37 and contributes this to clausal semantics. However, the prepositional group, as adjunct, construes “other, experientially more complex circumstances.”38 (I already noted the distinction between the prepositional group and the nominal group in clause structure.) These circumstances include all of the functions that might be performed by the relators of the prepositional group. These include location or position (ἀπό), time (ἐκ), and manner (ἐν), to be sure, but also the kinds of relations indicated by the range of relators, and so also direction (ἀνά), cause (διά), instrumentation (ὑπό), distribution (ἐν), extension (εἰς), purpose or result (εἰς), control or power (sphere) (ἐν), standard or basis (κατά), accompaniment (µετά), focus (περί), benefit (ὑπέρ), and substitution (ἀντί). Certain types of participants, which can be specified with a prepositional group as part of its nominal group that serves as head, are sometimes represented, unlike with adverbial groups, as “indirect participants,”39 such as those who are involved in these various functions, especially primary (ὑπό), secondary (διά) or tertiary/ instrumental (ἐν) agents. These indirect participants, again unlike with adverbial groups, can be elevated within transitivity structure to direct participant status, to become the agent or recipient of an action. The function of prepositional groups in causality (realized by the voice system and other

36. Halliday, IFG4, 363. 37. See Porter, Idioms, 125: time: αὔριον, νῦν, πέρυσι, πρωΐ, σήµερον, τότε; location: ἄνω, ἐκεῖ, ἐκεῖθεν, ἐνθάδε, ἐντεῦθεν, κάτω, πόρρω, ὧδε; manner: ἅπαξ, εἰκῇ, ἐξαίφνης, ἡδέως, οὕτω(ς), παραχρῆµα, ταχέως. 38. Halliday, IFG4, 364. 39. Halliday, IFG4, 364.

40

Biblical and Ancient Greek Linguistics 6

morphosyntactical systems), has clear semantic overlap with the function of the nominal group in indicating participants.40 7. Conclusion Whereas there is no single theory of the preposition in SFL, once we examine some of the particular features of the preposition in Greek, we see that we can theorize at least preliminarily about its systemic function within the Greek language. The prepositional group is most like the nominal group, and most unlike the adjectival/quality group, and somewhat like the adverbial group. Its functional potential falls within a category called relators, in which case it has similarities to other relators, such as conjunctions. However, there are a number of similarities that the prepositional group has with other elements of the Greek language, including both some adverbs and content words. In other words, the prepositional group has great expressive potential, whether used to elaborate a nominal group, in which it has an almost infinite nesting capacity for expression, whether used as an adjunct to provide greater circumstantial expressiveness than an adverbial group, or whether used as part of the relator system, in which case it functions similarly to conjunctions and other connectors. Bibliography Aarts, Flor, and Jan Aarts. English Syntactic Structures: Functions and Categories in Sentence Analysis. Oxford: Pergamon; Utrecht: Bohn, Scheltema & Holkema, 1982. Bennett, David C. Spatial and Temporal Uses of English Prepositions: An Essay in Stratificational Semantics. Longman Linguistics Library. London: Longman, 1975. Benson, James D., and William S. Greaves. The Language People Really Use. Agincourt, ON: Book Society of Canada, 1973. Bloom, Paul, et al., eds. Language and Space. Cambridge: MIT Press, 1996.

40. Halliday, IFG4, 364.

PORTER Greek Prepositions

41

Bortone, Pietro. Greek Prepositions: From Antiquity to the Present. Oxford: Oxford University Press, 2010. Fawcett, Robin P. A Theory of Syntax for Systemic Functional Linguistics. Amsterdam: Benjamins, 2000. Halliday, Michael A.K. Halliday’s Introduction to Functional Grammar (IFG4). Revised by Christian M.I.M. Matthiessen. 4th ed. London: Routledge, 2014. Halliday, M.A.K., and Jonathan J. Webster. Text Linguistics: The How and Why of Meaning. Sheffield: Equinox, 2014. Harris, Murray J. Prepositions and Theology in the Greek New Testament. Grand Rapids: Zondervan, 2015. Herskovits, Annette. Language and Spatial Cognition: An Interdisciplinary Study of the Prepositions in English. Studies in Natural Language Processing. Cambridge: Cambridge University Press, 1986. Hoffmann, Sebastian. Grammaticalization and English Complex Prepositions: A Corpus-Based Study. Routledge Advances in Corpus Linguistics. London: Routledge, 2005. Huddleston, Rodney. Introduction to the Grammar of English. Cambridge Textbooks in Linguistics. Cambridge: Cambridge University Press, 1984. Hudson, R.A. English Complex Sentences: An Introduction to Systemic Grammar. North-Holland Linguistics Series 4. Amsterdam: NorthHolland, 1971. Kühner, Raphael, and Bernhard Gerth. Ausführliche Grammatik der Griechischen Sprache. Part 2, 2 vols. Hannover: Hahnsche, 1898–1904. Lakoff, George. “Cognitive Semantics.” In Meaning and Mental Representations, edited by Umberto Eco et al., 119–54. Bloomington: Indiana University Press, 1988. ———. Women, Fire and Dangerous Things: What Categories Reveal about the Life of the Mind. Chicago: University of Chicago Press, 1987. Lakoff, George, and Mark Johnson. The Metaphors We Live By. Chicago: University of Chicago Press, 1980. Lindstromberg, Seth. English Prepositions Explained. Amsterdam: Benjamins, 1998. Louw, Johannes P., and Eugene A. Nida. Greek-English Lexicon Based on Semantic Domains. 2 vols. New York: United Bible Societies, 1988.

42

Biblical and Ancient Greek Linguistics 6

Luraghi, Silvia. On the Meaning of Prepositions and Cases: The Expression of Semantic Roles in Ancient Greek. Studies in Language Comparisons Series. Amsterdam: Benjamins, 2003. Metzger, Bruce M. Lexical Aids for Students of New Testament Greek. New ed. Princeton, NJ: Privately published, 1969. Morley, G. David. Explorations in Functional Syntax: A New Framework for Lexicogrammatical Analysis. London: Equinox, 2004. ———. “On Group and Phrase in Functional Grammar.” Word 55.2 (2004) 217–26. Muir, James. A Modern Approach to English Grammar: An Introduction to Systemic Grammar. London: Batsford, 1972. Peters, Ronald D. The Greek Article: A Functional Grammar of ὁ-items in the Greek New Testament with Special Emphasis on the Greek Article. Linguistic Biblical Studies 9. Leiden: Brill, 2014. Porter, Stanley E. Idioms of the Greek New Testament. Biblical Languages: Greek 2. 2nd ed. Sheffield: Sheffield Academic, 1994. ———. Linguistic Analysis of the Greek New Testament: Studies in Tools, Methods, and Practice. Grand Rapids: Baker, 2015. ———. Verbal Aspect in the Greek of the New Testament with Reference to Tense and Mood. Studies in Biblical Greek 1. New York: Peter Lang, 1989. Regier, Terry. The Human Semantic Potential: Spatial Language and Constrained Connectionism. Cambridge, MA: MIT Press, 1996. Robertson, A.T. A Grammar of the Greek New Testament in the Light of Historical Research. 4th ed. Nashville: Broadman, 1934. Sgall, Petr, Alla Bémová, and Eva Hajcová. “Remarks on the Semantic Features of Cases and Prepositions as Related to Syntax.” In Toward a Calculus of Meaning: Studies in Markedness, Distinctive Features and Deixis, edited by Edna Andrews and Yishai Tobin, 71–82. Studies in Functional and Structural Linguistics 43. Amsterdam: Benjamins, 1996. Taylor, John S. Linguistic Categorization: Prototypes in Linguistic Theory. 2nd ed. Oxford: Oxford University Press, 1995. Thompson, Geoff. Introducing Functional Grammar. 2nd ed. London: Hodder Education, 2004. Thrall, Margaret E. Greek Particles in the New Testament: Linguistic and Exegetical Studies. NTTS 3. Leiden: Brill, 1962.

PORTER Greek Prepositions

43

Watt, Jonathan M. “From Adams (1885) to Zimmermann (2009): In, With, and Under the Substance of Prepositions.” BAGL 6 (2017) 5–16.

[BAGL 6 (2017) 44–66]

SEMITIC INFLUENCE IN THE USE OF NEW TESTAMENT GREEK PREPOSITIONS: THE CASE OF THE BOOK OF REVELATION Laurențiu Florentin Moț Adventist Theological Institute, Cernica, Romania Abstract: Semitic influence on New Testament Greek prepositional use has been proposed by various scholars. At times, it turns out that the examples these scholars emphasize are quite unconvincing, many times because their methodologies seem unclear. This article proposes the use of the Second Language Acquisition approach in assessing the degree of Semitic influence on the New Testament Greek prepositions uses and applies it in the case of the prepositional irregularities found in the book of Revelation. Error Analysis is a method whereby the source of a linguistic irregularity is identified and the irregularity is explained. The question of this research is, what is the source of Revelation’s prepositional irregularities? The paper discusses the usage of prepositions such as εἰς, ἐν, ἐκ, µετά, ἀπό, and ἐπί in the book of Revelation, the New Testament, and the Greek language at large. Unclear terminology and inaccurate methodology are two factors that led to the conclusion that the source of the irregular prepositional use in Revelation is mainly Semitic. This paper uses the terminology of Second Language Acquisition and its findings drawn from empirical studies about linguistic transfer and facilitation from the mother tongue into the second language. In light of Second Language Acquisition, there seem to be strong arguments that confirm the Greek hypothesis and inform the Semitic explanation for virtually all of John’s peculiar prepositions. (Article) Keywords: Prepositions, Greek, Semitic influence, second language acquisition, Revelation.

MOȚ Semitic Influence

45

1. Introduction The issue of the Semitic influence on the New Testament Greek prepositions1 can be traced in modern times back to Henry Gehman who argues that the Hebrew language pervades the LXX Greek syntax and vocabulary.2 His article entitled “The Hebraic Character of the Septuagint Greek” points to several Greek prepositions which seem to render their Hebrew counterparts quite literally.3 Gehman highlights (1) ἐν which seems to assume meanings of ‫בּ‬, ְ such as instrumental and accompaniment,4 and (2) ἐκ which appears to denote the partitive sense of the Hebrew ‫מן‬. ִ 5 Nigel Turner extends Gehman’s hypothesis over the Greek of the New Testament.6 When it comes to prepositions, Turner finds several anomalies due apparently to Hebraic influence.7 These include the higher frequency of εἰς in place of local ἐν and the preposition εἰς having the Semitic causal sense of ‫ל‬,ְ 8 or replacing the classical περί.9 In 1. Prepositions make the verbal action or state more precise as they bring in new emphases and nuances about the verb and its substantive, that is, the one that produces the action. Webb and Kysar, Greek for Preachers, 67; Black, Learn to Read New Testament Greek, 37; Harrison, Greek Prepositions, 3–4. 2. Gehman, “The Hebraic Character,” 81–90. Gehman denies the notion of a Jewish-Greek jargon, but argues for a Jewish-Greek register used in religious contexts around the synagogue. 3. Gehman, “The Hebraic Character,” 83–84. 4. For example, the instrumental ἐν is visible in the way the Hebrew phrase ‫֖את‬ ָ ‫ ִמ ֵלּ‬-‫וּביָ ְד‬ ְ -‫ וַ ְתּ ַד ֵ ֥בּר ְבּ ִ ֛פי‬was translated with καὶ ἐλάλησας ἐν τῷ στόµατί σου καὶ ἐν χερσίν σου ἐπλήρωσας in 1 Kgs 8:24. It appears that the Greek preposition ἐν expresses accompaniment in the translation of ‫?־בּ ָ֕שׂר ְבּנַ ְפ ֥שׁוֹ ָד ֖מוֹ‬ ָ ‫ַא‬ with πλὴν κρέας ἐν αἵµατι ψυχῆς (Gen 9:4). 5. For illustration, Gehman refers the reader to ‫יאים‬ ִ֔ ‫יכ ֙ם ִלנְ ִב‬ ֶ ֵ‫אָקים ִמ ְבּנ‬ ֤ ִ ָ‫ו‬ ‫חוּר ֶיכ֖ם ִלנְ זִ ִ ֑רים‬ ֵ ‫וּמ ַבּ‬, ִ whose rendition in Greek is καὶ ἔλαβον ἐκ τῶν υἱῶν ὑµῶν εἰς προφήτας καὶ ἐκ τῶν νεανίσκων ὑµῶν εἰς ἁγιασµόν (Amos 2:11). 6. Lee, Jesus and Gospel Traditions, 232–34. 7. Turner, Syntax, 254–57. 8. E.g., ἐβαπτίσθη εἰς τὸν Ἰορδάνην (Mark 1:9). Turner calls τοῖς εἰς µακράν (Acts 2:39) a Semitism and ἐγένετο ἡ φωνὴ εἰς τὰ ὦτά µου (Luke 1:44) “especially Semitic.” 9. E.g., ὁ δεχόµενος προφήτην εἰς ὄνοµα προφήτου (Matt 10:41), allegedly, instead of ἐν ὀνόµατι. Δότε δακτύλιον εἰς τὴν χεῖρα αὐτοῦ (Luke

46

Biblical and Ancient Greek Linguistics 6

a similar manner, Wilbert Francis Howard builds his view of New Testament Greek on Henry John Thackeray’s remarks regarding LXX Greek.10 According to Thackeray, “Hebrew is responsible for the extensive use of a large number of prepositional phrases in place of an accusative after a transitive verb.”11 Howard mentions, among others examples from the New Testament, that ἀπό and ἐκ are found in constructions which have the Hebrew ‫ ִמן‬behind them and the idiom πολεµεῖν µετά τινος.12 C.F.D. Moule, who seems to rely heavily on Howard, adduces several other examples of apparent Semitic influence over the use of prepositions.13 Moule refers to the following idioms: ἔσονται οἱ δύο εἰς σάρκα µίαν, ὁµολογέω ἐν, ὀµνύναι ἐν or εἰς, θέλειν ἐν, ἔλεος µετά τινος.14 It is beyond the scope of this paper to evaluate each individual case mentioned above. The overview was meant to show, rather, that the syntax of prepositions in the Greek of the New Testament is considered by some authors to have a strong Semitic tinge. However, there is no clear and solid methodology in the approaches of these authors and many of the examples put forward are assumed, suspected, and possible but not necessarily real Semitisms. For example, to say that ‫ ְבּ‬stands sometimes behind the instrumental ἐν or that the partitive sense of ἀπό and ἐκ conveys the function of the Hebrew ‫ ִמן‬is very elusive, because ἐν does have an instrumental function, and both ἀπό and ἐκ can take a partitive role. Therefore, this study is

15:22), where Turner states that εἰς replaces περί. 10. Moulton and Howard, Accidence and Word-Formation, 254–57. 11. Thackeray, Grammar of the Old Testament Greek, I:46. 12. E.g., προσέχειν ἀπό (Luke 20:46), βλέπειν ἀπό (Mark 8:15; 12:38), ἐσθίειν ἀπό (Mark 7:28; Matt 15:27). Rev 2:16; 12:7; 13:4; 17:14 (also ποιῆσαι πόλεµον µετά, Rev 11:7; 12:17; 13:7; 19:19). 13. Moule, Idiom Book, 183–84. 14. E.g., the idiom ἔσονται οἱ δύο εἰς σάρκα µίαν (Matt 19:5) is instead a quotation of the literal translation of ‫( וְ ָהי֖ וּ ְל ָב ָ ֥שׂר ֶא ָ ֽחד‬Gen 2:24); ὅστις ὁµολογήσει ἐν ἐµοὶ ἔµπροσθεν τῶν ἀνθρώπων, ὁµολογήσω κἀγὼ ἐν αὐτῷ ἔµπροσθεν τοῦ πατρός µου (Matt 10:32; cf. Luke 12:8); ὃς ἂν ὀµόσῃ ἐν τῷ ναῷ, οὐδέν ἐστιν· ὃς δ᾽ ἂν ὀµόσῃ ἐν τῷ χρυσῷ τοῦ ναοῦ, ὀφείλει (Matt 23:16; cf. also 5:34–36; 23:18, 20–22; Heb 3:11; 4:3; Rev 10:6); µηδεὶς ὑµᾶς καταβραβευέτω θέλων ἐν ταπεινοφροσύνῃ καὶ θρησκείᾳ τῶν ἀγγέλων (Col 2:18); and ἐµεγάλυνεν κύριος τὸ ἔλεος αὐτοῦ µετ᾽ αὐτῆς (Luke 1:58).

MOȚ Semitic Influence

47

meant to apply a clear methodology to a particular group of irregular prepositions—the ones found in the book of Revelation, which are mainly explained as Semitisms. 2. Methodology and Question of Research Scholars who perceive a Semitic influence in the Greek of Revelation assume that the author was a Jew and that Aramaic was his first language, whereas Greek he acquired later in life.15 John was probably a bilingual Jew who knew an acceptable or intermediate level of Greek.16 That is the reason why John’s Greek can be studied from the perspective of Second Language Acquisition research. Error Analysis is a fundamental component of this discipline, which aims at exploring grammatical error in terms of its cause and the linguistic law that the syntactical construction breaks. From this perspective, the main question of the present research is what is the source of Revelation’s prepositional irregularities? Is it Semitic (Hebrew or Aramaic) or Greek? It will also be observed what difference a prepositional irregularity makes in the process of interpretation. As far as methodology is concerned, I will implement the following methodological guidelines of Error Analysis:17 (a) the collection of data,18 (b) the identification of errors (what is the grammatical law that a

15. E.g., Barr, “The Apocalypse of John,” 640. 16. For further details see my monograph on the solecisms of Revelation. Moț, Morphological and Syntactical Irregularities, 40, 227, 233–36. 17. See Ellis, Second Language Acquisition, 48–60, and Gass and Selinker, Second Language Acquisition, 103. 18. Ellis warns that spontaneous productions are more persuasive than the careful ones (which are not applicable to a text) and that longitudinal data is preferable to the cross-sectional. Ellis, Second Language Acquisition, 46–47.

48

Biblical and Ancient Greek Linguistics 6

particular construction breaks),19 (c) the classification of errors,20 (d) the quantification of errors (how many items each class of errors contains), (e) the analysis of errors, and (f) the remediation. The sixth point is relevant for teaching and is not applicable to a written text whose author is not present. The fifth point is worth further elaboration. What Gass and Selinker call “analysis” becomes “explanation and evaluation” for Ellis. The purpose of the “explanation” is to identify the source of the grammatical error. This source may be fourfold. The first cause may lie in psycholinguistics, which refers to the level of proficiency in the second language, particularly to how well the things known are delivered. The second cause relates to sociolinguistics, and it refers to the conformation of language to the social standard or context. The last two causes may be epistemological, when learners lack a world knowledge, or pertaining to discourse structure, in which case the incoherence of the text may account for many of its awkward constructions.21

19. The most important issue here is that of a linguistic standard, against which one should compare the error. Ellis argues that the “colonial” varieties of English are not to be viewed as erroneous. A more profound aspect is that a form may look grammatically correct but may not be the solution a native speaker would be in favor of. In order to discriminate between fine differences such as these, a corpus of native speakers is needed, against which the researcher can compare the problematic grammatical forms in a text suspected to have been influenced by a foreign language. Here Ellis follows Lemmon’s definition of error: “A linguistic form or combination of forms which, in the same context and under similar conditions of production, would in all likelihood, not to be produced by the speaker’s native speaker counterparts” (Second Language Acquisition, 48–50). 20. Ellis finds three types of taxonomies of errors. The first one is according to the linguistic or grammatical category. The second one originates with Corder in 1974 and differentiates between pre-systematic errors (i.e., there is no rule awareness), systematic errors (i.e., an incorrect rule is consistently applied), and post-systematic errors (i.e., the correct rule is inconsistently applied). The third classification is put forward by Dulay, Burt, and Krashen who see errors as omissions, additions, misinformations, or misorderings. Ellis, Second Language Acquisition, 50–52. 21. Ellis, Second Language Acquisition, 53.

MOȚ Semitic Influence

49

3. Semitic and Greek Explanations In order to collect the prepositional irregularities in Revelation I draw on my own reading of its Greek text and the extensive studies on its grammar made by G. Winer, G. Ewald, S. Davidson, F. Lücke, L. Cowden, W. Bousset, H. Swete, A.T. Robertson, R.H. Charles, E. Allo, BDF, S. Thompson, N. Turner, K. Newport, E. Dougherty, and D. Aune.22 The general opinion before Winer was that the grammatical irregularities in Revelation, like those in the rest of the New Testament, are due to the Semitic linguistic background of its author and of the sources he used. The victory of the Hebraists over the Purists in the opening of the nineteenth century led to a new development that New Testament Greek was one of its kind,23 a “Jewish Greek” totally apart from the Greek in use of the first century.24 Aside from Winer and Robertson, all the writers mentioned above offer Hebraic explanations for the linguistic peculiarities in the Apocalypse of John. Winer was brave enough to oppose this tendency and state in 1886 that the constructions that involved irregular government and apposition in the book of Revelation are partly intended, and partly traceable to the writer’s negligence. From a Greek point of view they may be explained as instances of anacoluthon, blending of two constructions, constructio ad sensum,

22. Winer, “De Solecismis,” 144–58; Ewald, Commentarius in Apocalypsin Johannis, 37–46; Davidson, Introduction to the Study of the New Testament, 201–4; Lücke, Offenbarung des Johannes, 2:448–64; Cowden, “Solecisms of the Apocalypse,” 5–20; Bousset, Die Offenbarung Johannis, 159–79; Swete, The Apocalypse of St. John, cxv–cxxv; Robertson, Grammar of the Greek New Testament, 413–16; Charles, Commentary, 1:cxvii–clix; Allo, Saint Jean L’Apocalypse, cxxxv–cliv; BDF, 75–76; Turner, Syntax, 314–15; Turner, Style, 146–48; Thompson, The Apocalypse and Semitic Syntax; Newport, “The Use of EK in Revelation,” 223–30; Newport, “Semitic Influence in Revelation,” 249–56; Dougherty, “The Syntax of the Apocalypse;” Aune, Revelation 1–5, clxxvii–clxxxiv. 23. An old work that collects essays from both camps is Rhenferdius, ed., Disertationum Philologicum-Theologicarum. For a modern research of the same see also Léonas, Recherches sur le langage, 4–20. 24. Janse, “The Greek of the New Testament,” 647.

50

Biblical and Ancient Greek Linguistics 6 variatio structurae, as should always have been done, instead of attributing them to the ignorance of the author, or pronouncing them to be mere Hebraisms, since most of them would be anomalies even in Hebrew, and in producing many of them Hebrew could have had only an indirect and incidental influence.25

Robertson avoids Semitic explanations as well. He shows that more proficient writers like Paul and Luke commit the same kinds of departures, the point of difference being that Revelation contains far more instances than the rest of the writers.26 The probability that John makes an irregular use of prepositions in Greek because of his mother tongue or because of the language of composition (Greek) should be analyzed in light of modern SLA empirical studies. 4. Second Language Acquisition Approach As far as the source is concerned, there are two types of linguistic errors: interlingual (this kind of error is also labeled as ‘transfer’ from the first language into the second language) and intralingual (in this case, the error is caused by the second language level of acquisition).27 The contribution of the mother tongue and the second language in the causation of linguistic errors has been a concern for decades. Ellis’s synthesis of various studies is hereby presented in six points.28 (a) More often than not, the great majority of the errors that the learners produce are not due to transfer, but are intralingual.29 (b) In the 1980s,

25. Winer, Idiom of the New Testament, 534–35. 26. Robertson, Grammar of the Greek New Testament, 414. 27. According to Gass and Selinker, other authors prefer to call the interlingual error as “interference” and the intralingual error as “developmental.” Gass and Selinker, Second Language Acquisition, 108, cf. 103. 28. Ellis, Second Language Acquisition, 55. 29. Patsy Lightbown and Nina Spada affirm that before identifying the source of an error as being the native language, the researcher must be sure that there is no foreigner of a different ethnic origin who uses the same type of irregular construction. On the contrary, if this is so, then the cause must be identified as intralingual. Lighbown and Spada, How Languages Are Learned, 187. Cook provides an example, which involves Spanish speakers of English but not speakers of a different native language: “In winter snows a lot in

MOȚ Semitic Influence

51

scholars considered that transfer errors occur more often at an elementary level,30 whereas intralingual errors prevail at an intermediate and advanced level of proficiency. However, in the 1990s, the thesis that transfer errors are prevalent with the beginners was challenged. (c) The degree of transfer and the number of intralingual errors is dependent on the task. For example, while translations favor transfer, it was found that free compositions do not. (d) Phonology and vocabulary, but not grammar, are the most common linguistic areas where transfer errors take place.31 In a study in 1971, there were recorded 25 percent lexical errors, 10 percent syntactical errors, and no morphological errors caused by transfer or interference with the native language.32 (e) Adults tend to produce more transfer errors than children. (f) Errors can derive from more than one source (e.g., intralingual, transfer). Empirical studies generally argue that syntactical irregularities are not due to the mother tongue, but seem to be developmental. The following questions are to be considered in the quest for the source of a syntactical (in our case prepositional) peculiarity.33 Is the construction in question possible in Hebrew/Aramaic and impossible in Greek? Is a prepositional peculiarity also present in non-Semitic linguistic backgrounds? Is an irregular construction awkward in literary κοινή, but quite common in non-literary κοινή? Did an irregular

Canada.” Spanish, as opposed to French, tolerates the lack of ‘it’ as subject of ‘snows.’ Cook, Second Language Learning, 35. 30. Brown states that the learner’s errors in the second language in the beginning levels are influenced by “the learner’s assumption that the target language operates like the native language.” Brown, Teaching by Principles, 76. 31. So Parker and Riley, Linguistics for Non-Linguists, 216; Fromkin et al., Introduction to Language, 381; Spada and Lightbown, “Second Language Acquisition,” 116. 32. However, language transfer does occur “at the level of pronunciation, morphology, syntax, vocabulary, or meaning” (Omaggio, Teaching Language in Context, 276). 33. Similar questions are addressed in Porter, “Language of the Apocalypse,” 582–603.

52

Biblical and Ancient Greek Linguistics 6

construction change from being awkward into becoming accepted later, as the language evolved? 5. Case by Case Analysis The prepositional irregularities identified in this paper are basically of three kinds. The first has the preposition followed by the wrong case. The second kind has the preposition apparently used with the wrong verb. The third irregularity is identified as one preposition replacing another (expected) preposition. The classification I opted for is one which derives from the very prepositions analyzed. Accordingly, there are four classes of prepositions seemingly used in an irregular manner. The first and the third contain one irregular instance each. The second category is found in seven verses, whereas the last category is illustrated in six places. In total, there are fifteen individual cases of alleged irregular prepositions in the book of Revelation. What follows is the analysis (explanation and evaluation) of each of the four classes. 5.1 Ἀπό Followed by the Nominative Revelation 1:4 is probably the most common verse to illustrate solecisms in Revelation and it happens to involve a preposition. John sends greetings ἀπὸ ὁ ὢν καὶ ὁ ἦν καὶ ὁ ἐρχόµενος, literally “from He is, He was, and He is coming.” The preposition ἀπὸ is followed by a nominative phrase, not by the expected genitive case. This rendition has the consensus of !18, ‫א‬, A, C, P, and 2050 and there is no doubt that it is original. Later scribes tried to save the grammar by two types of corrections: ἀπὸ Θεοῦ34, ὁ ὢν (the Majority Text, GOC, RPT, and BYZ) and ἀπὸ τοῦ ὁ ὢν (Textus Receptus, STE, TBT, SCR, and MGK). The first solution makes the expression ὁ ὢν καὶ ὁ ἦν καὶ ὁ ἐρχόµενος a nominative in

34. There are also 11th- to 15th-century manuscripts containing the abbreviation of θεοῦ as θυ (see 69, 424, 1006, 1854, 2493, 2494, 2495, and 2845).

MOȚ Semitic Influence

53

apposition to a genitive. The second solution suggests that the same collocation is a title phrase.35 The oldest remark on the problem in Rev 1:4 is that of the Italian humanist and rhetorician Lorenzo Valla (c. 1407–1457). In his Annotationes to the New Testament, Valla asks the reader to decide if John conveys God’s immutable attribute here,36 which presumably led John not to alter the grammar as he should. The most common explanation is that ἀπὸ ὁ ὢν καὶ ὁ ἦν καὶ ὁ ἐρχόµενος is an intentional deviation from the rules of grammar whereby John treats the nominative formula as an indeclinable title.37 Charles states that ἀπὸ ὁ ὢν καὶ ὁ ἦν καὶ ὁ ἐρχόµενος is an example of the Jewish respect for the holy name of God.38 Similarly, Büschel perceives an intention to “preserve the sanctity of the divine self-predication,”39 Robertson sees that John wants “to accent the unchangeableness of God,”40 while Porter glimpses that what we have here is a poetical license.41 In the fashion of rabbinical exegesis on the phrase ἐγώ εἰµι ὁ ὤν in Exod 3:14, BDF identifies the construction as an unpolished nominative used to introduce names.42 William Guillemard holds that Rev 1:4 contains an “anomalous

35. Ewald disagrees with the insertion of τοῦ before the nominative by stating that prepositions never appear in grammatical connection with the nominatives. Ewald, Commentarius in Apocalypsin Johannis, 46. 36. “An uoluit Ioannes in deo significare immutabilem proprietam?” Vallae, Viri Tam Graecae Quàm Latinae Linguae Doctissimi, 339. 37. Winer, on the other hand, considers the collocation to be a serious sin, arguing that when John wants to decline the indeclinable name of God he even inflects it ungrammatically, cf. ὁ ἦν. Winer, “De Solecismis,” 156. 38. Charles, Commentary, 1:clii. Bousset names the expression in question “a solemn declaration.” Bousset, Die Offenbarung Johannis, 159. Cowden argues that the phrase ἀπὸ ὁ ὢν καὶ ὁ ἦν καὶ ὁ ἐρχόµενος “is the Greek equivalent for the Hebrew name Jehovah” (“Solecisms of the Apocalypse,” 11– 12). 39. Büschel, “eimí, ho ṓn,” 206. 40. Robertson, Grammar of the Greek New Testament, 414. 41. “Perhaps the entire phrase, rather than being solecistic, reflects a conscious use of poetic license: cf. ‘from the Is, the Was and the Coming One’” (Porter, Idioms, 146). 42. BDF, 79.

54

Biblical and Ancient Greek Linguistics 6

construction, clearly traceable to the absence of inflexion in Hebrew nouns, which made such a violation of grammar less startling to a Jew writing in Greek.”43 On the other hand, Allo does not see here a Hebrew transfer, for in his opinion, ἀπὸ ὁ ὢν καὶ ὁ ἦν καὶ ὁ ἐρχόµενος is a notorious solecism, which cannot be explicated either through Hebrew, or the construction according to sense, or the vulgar Greek.44 Moulton labels the same rendition a “tour de force”45 as he finds examples of nominative in apposition to oblique cases in the papyri.46 Moulton’s explanation that Rev 1:4 may be an example of vulgar Greek may not fit very well in Rev 1:4. This is because John is aware of the rule and implements it effectively. He never uses the preposition ἀπό with a case other than the required genitive in 35 instances, apart from 1:4a. This consistency in applying the rule also reduces the chances of a Hebrew transfer to a minimum. Rather, the explanation must be sought in John’s Greek

43. Guillemard, Hebraisms in the Greek Testament, 116. 44. Allo, Saint Jean L’Apocalypse, cxlviii. 45. Moulton, “Grammatical Notes From the Papyri,” 151–52. 46. I found the following examples in support of his contention: (a) P.Tebt. I 41.8–11: ποιουµένου τινῶν ἡµ̣ῶν καὶ ἑτέρων γυναικῶν διασείειν, οὐ στοχασάµενος (“of making some of us and other women to tremble, in truth aiming at”). Στοχασάµενος, a nominative participle, qualifies a genitive absolute. (b) BGU III 910.2.11: τοῦ ἀνδρός µου Ὀννῶφρις (“of my husband Onnofris”). The proper noun appears as an indeclinable nominative modifying a genitival referent. (c) The Christian inscription, Egypt and the Cyrenaica [Chr.], Philai II document 197.9–14, ἦλθα ἐνταῦθα καὶ ἐποίησα τὸ ἔργον µου ἅµα καὶ τοῦ ἀδελφοῦ µου Σµητό, διάτοχος [for διάδοχος] τοῦ προφήτου (“I came here and I did my work at once and of my brother, Smeto, a successor of the prophet”). Διάτοχος, a nominative qualifier of an indeclinable proper noun, should have appeared in the genitive. An identical case is (d) P.Cair. 15.7: παρὰ Τασοῦτος µητρὸς µητὴρ Ταυρίνου (“from the mother of Tasutos, mother of Taurinus”). (e) P.Oxy. III 527.r.2–4: περὶ Σερήνου τοῦ γναφέως τοῦ ὁ συνεργαζόµενος µετὰ Φιλέου (“about Serenos the fuller, the one which is working together with Phileos”). Συνεργαζόµενος, which is a participial nominative, qualifies the genitive τοῦ γναφέως. (f) The Christian inscription, Egypt and the Cyrenaica [Chr.], Philai II document 197.17–19, ε[ὐχ]αριστο̣ῦ̣µεν̣ τ̣ῇ [δ]εσποίνῃ ἡµῶν Ἶσις [καὶ τ]ῷ δεσπότῃ ἡµ[ῶν Ὄς]ιρις (“we thank to our queen Isis and to our master Osiris”). Isis and Osiris are in the nominative case and yet apposed to datives.

MOȚ Semitic Influence

55

language. There is also little room for the explanation put forward by some of the scholars above, that John’s respect for the sacred name and God’s immutability made him to intentionally write the name in the nominative. The most probable explanation is to come out of the marks John leaves in the text. The phrase in question is part of the greetings section in Rev 1:4–6. John sends grace and peace from the Trinitarian Godhead. It can be observed that in the case of the Holy Spirit and Jesus, after the proper names, there is an appositional qualification of these names.47 But in the case of the first person (God the Father), instead of writing the proper noun (τοῦ Θεοῦ) and then the apposition,48 John seems more attracted by the importance of the apposition, for which reason he omits the name that was supposed to precede “the one who is, who was, and who is coming.” This ellipsis is actually an instance of a nominative in apposition to an oblique case, a feature found quite often in Revelation. The omission may signify that the apposition prevailed in the mind of John over the name itself.49 In light of pragmatics, the more two speakers have in common the less explicit they are towards one another. In the present case, the writer hints to a word without writing it and he should not be considered mistaken in connecting a preposition that requires the genitive to a nominative expression.50

47. The examples are: ἀπὸ τῶν ἑπτὰ πνευµάτων ἃ ἐνώπιον τοῦ θρόνου αὐτοῦ καὶ ἀπὸ Ἰησοῦ Χριστοῦ, ὁ µάρτυς, ὁ πιστός, ὁ πρωτότοκος τῶν νεκρῶν καὶ ὁ ἄρχων τῶν βασιλέων τῆς γῆς (Rev 1:4b–5). When it comes to the seven spirits, NA28 follows !18 but both ‫ א‬and A have the subject qualified by an apposition in the same case, ἀπὸ τῶν ἑπτὰ πνευµάτων τῶν ἐνώπιον τοῦ θρόνου αὐτοῦ. 48. It is significant that John uses the combination ἀπὸ τοῦ θεοῦ four times (Rev 3:12; 12:6; 21:2, 10). 49. This would be in tune with John’s description of God in Rev 4, when he focuses on the details in the vision and not in his name (ἐπὶ τὸν θρόνον καθήµενος, καὶ ὁ καθήµενος, Rev 4:2–3). 50. Richard Young states that the combination ἀπὸ plus a nominative “can only be a violation if grammar is viewed prescriptively. With a descriptive view of grammar, it merely illustrates the range of expression that koine Greek tolerates. Thus John’s use of the nominative is not a mistake in grammar” (Intermediate New Testament Greek, 13). Young affirms that if something

56

Biblical and Ancient Greek Linguistics 6

The result may be considered solecistic, albeit not because of the combination ἀπὸ and the nominative, but because of the absent τοῦ Θεοῦ51 followed by nominative apposition. The explanation of this irregularity (nominative in apposition to a genitive) is not part of the present topic and I have detailed it extensively elsewhere.52 5.2 Πολεµεῖν µετά Guillemard argues that the use of µετά in the phrase πολεµήσω µετ᾽ αὐτῶν ἐν τῇ ῥοµφαία (2:16) “is against all good Greek usage” and its place should have been taken by ἐπί.53 Indeed, when ἐπί is followed either by an accusative or a dative, its meaning is that of “against.” To Guillemard, this is a Hebraism (cf. 2 Kgs 14:5), just as it is also for Newport.54 When the lexicons define the preposition µετά,55 one of the suggested meaning is that of conflict against, albeit it is acknowledged that its primary sense is associative. The

occurs in a language then it is not a mistake. However, if there are not rules (as in the prescriptive grammar), there should be some regularities (as descriptive grammar professes) that distinguish between what is regular and what is not. In Young’s view, the language (i.e., the general system) is equal with the idiolect (i.e., the personal reproduction of that system). In contrasting with Young’s view, I argue that since there is no other example in the Greek language, ἀπό followed by a nominative is irregular. However, if τοῦ Θεοῦ is implied there is nothing irregular about ἀπό followed by a nominative. 51. For an alternative view one must also consult Mussies, Morphology of the Koine Greek, 93–94. Mussies identifies several parallels in the Septuagint and some Qumran-scrolls which have the Divine Name replaced by four dots (Q.S. VIII 14 in a quotation from Isa 40:3). Mussies conjectures that the autograph of Revelation may have contained the same four dots for the Divine Name, and thus been easy to get dim by thumbing of the scroll or its decaying. Eventually, through the later scribes, ἀπὸ . . . . ὁ ὢν became ἀπὸ ὁ ὢν and finaly ἀπὸ ὁ ὢν. 52. See Moț, Morphological and Syntactical Irregularities, 108–34. 53. Guillemard, Hebraisms in the Greek Testament, 116–17. 54. Newport, “Semitic Influence in Revelation,” 250. 55. Including Friberg et al., Analytical Lexicon; BDAG; Louw and Nida, Greek-English Lexicon; Swanson, Dictionary of Biblical Languages; Newman, Concise Greek-English Dictionary; Lust et al., Greek-English Lexicon of the Septuagint.

MOȚ Semitic Influence

57

lexicographers, however, tend to be descriptive, that is, to suggest that by finding a certain syntax in the language that particular syntax is normal. But is the conflicting sense of µετά regular in the wider range of the Greek language? In order to answer this one must look at how µετά has been used in Greek in general.56 In Classical Greek, “fighting with [µετά]” meant only “joining in war with,” and never “fighting against.” I will use two examples. One is from Thucydides’s Historiae 1.59.2.4–5, where the Greek writer states that the Athenians ἐπολέµουν µετὰ Φιλίππου καὶ τῶν Δέρδου ἀδελφῶν. By this he means that the Athenians joined Phillip and the brothers of Derdas in a battle against the Macedonians. Pausanias also writes that the men of Asine declined πολεµῆσαι µετὰ Λακεδαιµονίων (“fighting with the Lacedaemonians”) against the people of Nauplia.57 Hellenistic Greek follows the classical with the exception of the Septuagint. All fifteen occurrences of µετά58 in the LXX have a conflictual sense, being a literal rendition of ‫ל ַחם ִעם‬.ָ In the New Testament, “to fight against [µετά]” is exclusively found in Revelation (2:16; 12:7, 17; 13:4, 7; 17:14; 19:19).59 Perhaps through Christian writers, the idiom made its way into the Byzantine times.60 One example is ἐπολέµησεν µετ’ (against) found in Historia Alexandri Magni.61 Another example is

56. For a similar methodology, see Sollamo, “Some ‘Improper’ Prepositions,” 781. Sollamo acknowledges that though the method may be laborious “there is no other way to go.” 57. Pausanias, Graeciae Description 4.27.8.4. See also Scholia in Aeschylum (scholia recentiora) Th.635.6, συµπολεµῆσαι µετὰ σοῦ. Strabo, Geographica 11.5.2.5–6, πολεµεῖν µετὰ Θρᾳκῶν καὶ Εὐβοέων τινῶν. 58. Judg 5:20; 11:20; 20:14, 18; 1 Sam 17:32–33; 28:1; 2 Sam 10:17; 11:17; 21:15; 1 Kgs 12:24; 2 Kgs 14:15; 19:9; and Dan 11:11 (in Theodotion also Dan 10:20). 59. The idea of conflict with (µετά) is found in the New Testament, but with other verbs: e.g., ἀδελφὸς µετὰ ἀδελφοῦ κρίνεται (1 Cor 6:6), ἐγένετο οὖν ζήτησις ἐκ τῶν µαθητῶν Ἰωάννου µετὰ Ἰουδαίου (John 3:25), and περὶ τούτου ζητεῖτε µετ᾽ ἀλλήλων (John 16:19). 60. For other examples, see LSJ, s.v. “µετά,” and BDAG, s.v. “µετά.” 61. Historia Alexandri Magni, Recensio γ (lib. 1) 46.80 and Recensio K 286.12. See Mitsakis, “Διήγησις περὶ τοῦ Ἀλεξάνδρου καὶ τῶν µεγάλων

58

Biblical and Ancient Greek Linguistics 6

Cedrenus who writes concerning the battles of the Saracenes against the Christians: καὶ κατὰ ἔαρ ὁµοίως ἐπολέµουν µετὰ τῶν Χριστιανῶν ἐπὶ ἑπτὰ ἔτη (“and likewise, for seven years they fought against the Christians around spring,” Compendium Historiarum 1.765.5). Henry Thayer’s conclusion is also my conclusion: by the time Revelation was written, to combine a conflicting µετά with πολεµεῖν was “a usage foreign to the native Greeks.”62 Apparently, the collocation πολεµεῖν µετά, found seven times in Revelation (2:16; 12:7, 17; 13:4, 7; 17:14; 19:19), is a Semitic transfer, which was first committed by way of translation in the Septuagint.63 Then, John’s familiarity with the LXX Greek definitely facilitated the import of this transfer in Revelation. 5.3 The Directional ἐν Lücke argues that the expression εἰσῆλθεν ἐν αὐτοῖς (Rev 11:11) is a Hebraism which translates the ‫ בּוֹא ְבּ‬formula.64 NA28 follows the agreement between !115 and A, but there are several manuscripts and versions which replace ἐν with εἰς (!47, ‫א‬, 69, 424, 2845, 2494, GOC, BYZ, RPT, and MGK). The question is whether the directional ἐν is peculiar or regular in Greek. The overlap between the two prepositions used with verbs of motion and rest was perceived in ancient Greek as in the New Testament.65 I will bring forth a few examples from various

πολέµων,” 286. 62. Thayer, Greek-English Lexicon, s.v. “µετά.” 63. As interesting as it may seem, the scribes tried to modify only the combination in Rev 12:7 into κατὰ τοῦ δράκοντος (Textus Receptus, STE, RPT, SCR, and MGK), leaving the remaining six as we have them. 64. Lücke, Offenbarung des Johannes, 2:459. 65. “But in the New Testament, as in the older Greek, the real idea of each of the eight cases is manifest, though the process of blending has made further progress as is seen in the practical equivalence of εἰς and accusative and ἐν (the locative) with verbs of rest and motion. The practical absence of cases in the Hebrew would accentuate this tendency to some extent” (Robertson, Short Grammar, 89). See also Wallace, Greek Grammar, 372; Moule, Idiom Book, 75–76.

MOȚ Semitic Influence

59

periods that will confirm this overlap in Greek and invalidate Lücke’s Hebraic explanation. Dio Cassius wrote εἰσῆλθεν ἐν χιτωνίσκῳ and Aesop wrote εἰσῆλθεν ἐν τῷ πλοίῳ αὐτῶν.66 In like manner, Acta Pauli 44.4 contains εἰσῆλθεν ἐν τῇ πέτρᾳ ζῶσα and Acta Thomae 16.16 reveals how the apostle Thomas εἰσῆλθεν ἐν τῇ Ἰνδίᾳ. Finally, a Greek chronicler from Antioch named Ioannis Malalas (491– 578) uses the directional ἐν idiom at length. In Chronographia 36.6 it is found, εἰσῆλθεν ἐν αὐτῷ [ἱερὸν Ποσειδῶνος], 93.9 reads εἰσῆλθεν ἐν τῇ Τροίᾳ ὁ αὐτὸς Πάρις.67 These examples cover around 1200 years of Greek.68 It seems obvious then that the directional sense of ἐν appears legitimate in all periods of the Greek language and no Hebrew causation ought to be suspected.69 Now if ἐν + dative is at times equivalent to εἰς + accusative,70 there is no wonder why “all prepositions in late medieval Greek govern the accusative.”71 The driving principle of this shift towards the prominence of the accusative case is already at work in Revelation. 5.4 The Peculiar Use of ἐκ Aune conjectures that the use of ἐκ in τοὺς νικῶντας ἐκ τοῦ θηρίου (15:2) is Latin.72 This proposal is older as it was already known to Ebrard: “Νικῶντας ἐκ τοῦ θηρίου is a formerly non-existent construction, is hardly a Latinism (for victoriam ferre ex aliquo),

66. Aesop, Fabulae 21.2.5; Dio Cassius, Historiae Romanae 59.25.8.2– 3. 67. Malalas uses this idiom many more times. Other instances which testify of the same thing are 140.21; 184.19; 211.17; 222.13; 224.8; 264.7; etc. 68. A very informed and careful analysis is offered in BDF, 117. 69. Bortone, Greek Prepositions, 209. 70. Murray Harris mentions that ἐν and εἰς share some common features ever since Classical Greek, though infrequently. During Hellenistic Greek, this phenomenon became more obvious, being present in the whole New Testament. Harris, Prepositions and Theology, 84–85. 71. Riekert, “Reconsidering Prepositions and Case Assignment,” 364. Based on his analysis of Rev 4–5, Riekert argues that occasionally ἐπί + accusative had the function of ἐπί + genitive or dative so he draws the conclusion that the distinction “was on the wane” (366). 72. Aune, “Latinism in Revelation,” 691–92.

60

Biblical and Ancient Greek Linguistics 6

[but] more likely an intended Hebraism, a pregnant construction, ‘who the conquerors were away from the beast’.”73 Like others,74 Ebrard opts for a Hebraic explanation and suggests that the use of ἐκ here infers that the conquerors escaped from the beast. At the same time, Winer, Buttmann, and James Moffatt are in favor of a Latin explanation.75 However, most interpreters agree that here we have a pregnant construction,76 the only difference between them coming from what they supply for what is missing. Some scholars argue for an ablatival (separative) ἐκ. In this case, the ones who conquered secured their victory by separating themselves from the enemy (the beast and its image).77 Others emphasize the act of deliverance “from the beast and its image.”78 Still others emphasize the conflict, out of which the conquerors came.79 To these, I would suggest a fourth option, and that is to take the preposition ἐκ as combining both the source and the partitive aspects.80 The result would be that the victors conquered some of those pertaining to the beast, which are described in the phrase ὅλη ἡ γῆ ὀπίσω τοῦ θηρίου (Rev 13:3). It seems difficult to argue that the source of the prepositional peculiarity in τοὺς νικῶντας ἐκ τοῦ θηρίου (15:2) is either Hebrew or Latin, due particularly to a lack of evidence. It rather looks like a Greek pregnant construction. In this case, one should

73. Ebrard, Die Offenbarung Johannes, 422. Newport interprets the phrase as a Hebraism. Newport, “The Use of EK in Revelation,” 226–27. 74. Dougherty, “The Syntax of the Apocalypse,” 358. 75. Buttmann, A Grammar of the New Testament Greek, 147; Winer, Idiom of the New Testament, 367. 76. E.g., Charles, Commentary, 2:28. 77. Turner, Syntax, 260; Beale, The Book of Revelation, 790; Blass, Grammar, 126; BDF, 114; Robertson, Grammar of the Greek New Testament, 598. Swete states, “[t]he construction is a pregnant one, ‘by virtue of their victory they escaped out of the hand of the enemy’” (Apocalypse of St. John, 191). 78. Moffatt, Revelation, 443; Thayer, Greek-English Lexicon, s.v. “ἐκ”; Robertson, Grammar of the Greek New Testament, 598; Burton, The Greek Testament with English Notes, 552. 79. Beckwith, The Apocalypse of John, 674; Allo, L’Apocalypse, 231. 80. Luraghi, Prepositions and Cases, 97–99.

MOȚ Semitic Influence

61

evaluate the context in order to find out what is emphasized: the spiritual resistance of the saints, the deliverance of God, or the battle, and then suggest the appropriate sense of ἐκ. A second peculiar use of ἐκ in Revelation, which was signaled in secondary literature, is the collocation µετανοεῖν ἐκ, employed by John five times (2:21, 22; 9:20, 21; 16:11). Charles, Aune, and Newport suggest a Hebraic explanation, the collocation seemingly reflecting (primarily in Symmachus, but less in LXX) a Hebrew idiom such as ‫שׁוּב ִמן‬.81 But there is no need for this conjecture. First, the idiom µετανοεῖν ἐκ is not absent from Greek literature, but it is found three times in Hermas, a Greek document written probably by a Latin author, with little or no Hebrew to have influenced him.82 Second, the Hellenistic overlap between ἐκ and ἀπό makes µετανοεῖν ἐκ to be practically equivalent to µετανοεῖν ἀπό, which is found in Jer 8:6, John Chrysostom, Justin Martyr, and other Greek writers. In all references above, the idiom µετανοεῖν ἐκ denotes separation.83 6. Synthesis, Evaluation, and Implications The study of the four classes of prepositions in the book of Revelation brought to light divergent conclusions. Thus, the preposition ἀπό followed by the nominative is probably an instance of a nominative in apposition to a genitive, with the genitive missing. It is an intralingual error caused by an infelicitous ellipsis. The idiom “to fight against [µετά]” is a Septuagintalism, found nowhere else in contemporary or older Greek. This is an interlingual error caused by the Greek translation of the Hebrew OT. The directional sense of ἐν is purely Greek, but falsely viewed a Hebraism. The phrase “conquering away from [ἐκ]” sounds Latin and makes sense in Greek only as a contructio praegnans, to which absent details

81. Charles, Commentary, 1:71; Aune, Revelation 1–5, clxxx; Newport, “The Use of EK in Revelation,” 225–26. 82. See µετενόησαν ἐξ in Hermas 72.6.4 and 74.5.2, and µετενό[ησαν] ἐκ in 100.2.3. 83. Dougherty, “The Syntax of the Apocalypse,” 357.

62

Biblical and Ancient Greek Linguistics 6

need to be supplied. Lastly, µετανοεῖν ἐκ is found in documents of non-Semitic authors, which invalidates the Hebraic explanation. This brings out several facts to consider when Semitic influence is measured in the New Testament Greek syntax. First, if a particular use of a preposition is possible in both Greek and Hebrew, there is no reason to suspect a transfer from Hebrew. Second, before a prepositional peculiarity in Greek is judged as Semitic, it must be checked whether writers from non-Semitic backgrounds commit the same irregularity. Third, it must be also considered whether a prepositional usage that was peculiar maybe at the time of writing became regular later on in the development of the Greek language. Lastly, Semitic influence in Greek should be tested through the Second Language Acquisition approach, which, though it usually applies in empirical environments, deserves a place in the methodological spectrum of New Testament studies. Bibliography Allo, Ernest Bernard. Saint Jean L’Apocalypse. 2nd ed. Paris: Librairie Victor Lecoffre, 1921. Aune, David E. “A Latinism in Revelation 15:2.” JBL 110.4 (1991) 691–92. ———. Revelation 1–5. WBC 52A. Dallas: Word, 2002. Barr, David L. “The Apocalypse of John.” In The Blackwell Companion to the New Testament, edited by David E. Aune, 632–51. Chichester: WileyBlackwell, 2010. Beale, Gregory K. The Book of Revelation: A Commentary on the Greek Text. Grand Rapids: Eerdmans, 1999. Beckwith, Isbon T. The Apocalypse of John: Studies in Introduction with a Critical and Exegetical Commentary. New York: Macmillan, 1919. Black, David Alan. Learn to Read New Testament Greek. Nashville: B&H, 2009. Blass, Friedrich. Grammar of New Testament Greek. Translated by Henry St. John Thackeray. London: Macmillan, 1898.

MOȚ Semitic Influence

63

Bortone, Pietro. Greek Prepositions from Antiquity to the Present. Oxford: Oxford University Press, 2010. Bousset, Wilhelm. Die Offenbarung Johannis. Göttingen: Vandenhoeck und Ruprecht, 1906. Brown, H. Douglas. Teaching by Principles: An Interactive Approach to Language Pedagogy. 3rd ed. New York: Pierson Longman, 2007. Burton, Edward. The Greek Testament With English Notes. 3rd ed. Oxford: John Henry Parker, 1848. Büschel, F. “eimí [to exist], ho ṓn [‘I am’].” In Theological Dictionary of the New Testament, edited by Gerhard Kittel and Gerhard Friedrich, translated by Geoffrey William Bromiley, 206–207. Grand Rapids: Eerdmans, 1995. Buttmann, Alexander. A Grammar of the New Testament Greek. Andover, MA: Waren F. Draper, 1891. Charles, Robert Henry. A Critical and Exegetical Commentary on the Revelation of St John. Edinburgh: T&T Clark, 1920. Cook, Vivian. Second Language Learning and Language Teaching. 4th ed. London: Hodder Education, 2008. Cowden, T. Laughlin. “The Solecisms of the Apocalypse.” PhD diss., Princeton University, 1902. Davidson, Samuel. An Introduction to the Study of the New Testament: Critical, Exegetical, and Theological. 3rd ed. London: Kegan Paul, 1894. Dougherty, Edward. “The Syntax of the Apocalypse.” PhD diss., The Catholic University of America, 1990. Ebrard, Johann Heinrich August. Die Offenbarung Johannes. Königsberg, Prussia: Unzer, 1853. Ellis, Rod. The Study of Second Language Acquisition. 2nd ed. Oxford: Oxford University Press, 2012. Ewald, Georgio Henrico Augusto. Commentarius in Apocalypsin Johannis: Exegticus et Criticus. Leipzig: Librariae Hahnianae, 1828. Friberg, Timothy, Barbara Friberg, and Neva F. Miller. Analytical Lexicon of the Greek New Testament. Baker’s Greek New Testament Library. Grand Rapids: Baker, 2000. Fromkin, Victoria, Robert Rodman, and Nina Hyams. An Introduction to Language. 7th ed. Boston: Thompson, 2003.

64

Biblical and Ancient Greek Linguistics 6

Gass, Susan M., and Larry Selinker. Second Language Acquisition. 3rd ed. London: Routledge, 2008. Gehman, Henry S. “The Hebraic Character of the Septuagint Greek,” VT 1 (1951) 81–90. Guillemard, William Henry. Hebraisms in the Greek Testament. London: George Bell and Sons, 1879. Harris, Murray J. Prepositions and Theology in the Greek New Testament. Grand Rapids: Zondervan, 2012. Harrison, Gessner. A Treatise on the Greek Prepositions and on the Cases of Nouns with Which These Are Used. Philadelphia: J.B. Lippincott, 1858. Janse, Mark. “The Greek of the New Testament.” In A History of Ancient Greek: From the Beginnings to Late Antiquity, edited by A.-F. Christidis, 646–53. Cambridge: Cambridge University Press, 2010. Lee, Sang-Il. Jesus and Gospel Traditions in Bilingual Context: A Study in the Interdirectionality of Language. Berlin: de Gruyter, 2012. Léonas, Alexis. Recherches sur le langage de la Septante. Fribourg: Academic Press Fribourg, 2005. Lightbown, Patsy M., and Nina Spada. How Languages Are Learned. 2nd ed. Oxford: Oxford University Press, 2011. Lücke, Friedrich. Versuch einer vollständigen Einleitung in die Offenbarung des Johannes oder Allgemeine Untersuchungen. Bonn: Eduard Weber, 1852. Luraghi, Silvia. On the Meaning of Prepositions and Cases: The Expression of Semantic Roles in Ancient Greek. Studies in Language Companion Series 67. Amsterdam: Benjamins, 2003. Lust, Johan, Erik Eynikel, and Katrin Hauspie. A Greek-English Lexicon of the Septuagint. Rev. ed. Stuttgart: United Bible Societies, 2003. Mitsakis, K. “Διήγησις περὶ τοῦ Ἀλεξάνδρου καὶ τῶν µεγάλων πολέµων.” Byzantinisch-Neugriechische Jahrbücher 20 (1970) 263–90. Moffatt, James. The Revelation of St. John, the Divine. The Expositor’s Greek Commentary 5. London: Hodder and Stoughton, 1897. Moț, Laurențiu Florentin. Morphological and Syntactical Irregularities in the Book of Revelation. Linguistic Biblical Studies 11. Leiden: Brill, 2015. Moule, C.F.D. An Idiom Book of New Testament Greek. 2nd ed. Cambridge: Cambridge University Press, 1990.

MOȚ Semitic Influence

65

Moulton, James Hope. “Grammatical Notes from the Papyri.” The Classical Review 18.3 (1904) 151–155. Moulton, J.H., and W.F. Howard. A Grammar of New Testament Greek. Vol. 2, Accidence and Word-Formation. 3rd ed. Edinburgh: T&T Clark, 2006. Mussies, Gerard. The Morphology of the Koine Greek as Used in the Apocalypse of St. John: A Study in Bilingualism. NovTSup 27. Leiden: Brill, 1971. Newman, Barclay Moon. A Concise Greek-English Dictionary of the New Testament. Stuttgart: United Bible Societies, 1993. Newport, Kenneth G.C. “Semitic Influence in Revelation: Some Further Evidence.” AUSS 25.3 (1987) 249–56. ———. “The Use of EK in Revelation: Evidence of Semitic Influence.” AUSS 24.3 (1986) 223–30. Omaggio, Alice C. Teaching Language in Context: Proficiency-Oriented Instruction. Boston: Heinle & Heinle, 1986. Parker, Frank, and Kathryn Riley. Linguistics for Non-Linguists: A Primer with Exercises. 4th ed. Boston: Allyn & Bacon, 2005. Porter, Stanley E. Idioms of the Greek New Testament. Biblical Languages: Greek 2. Sheffield: JSOT, 1994. ———. “The Language of the Apocalypse in Recent Discussion.” NTS 35 (1989) 582–603. Rhenferdius, Jacobus., ed. Disertationum Philologicum-Theologicarum de Stylo Novi T. Syntagma. Leovardie: Heronis Nautae, 1701. Riekert, S.J.P.K. “Reconsidering Prepositions and Case Assignment in the Text of Revelation 4 and 5.” HTS Teologiese Studies 60 (2004) 349–67. Robertson, A.T. A Grammar of the Greek New Testament in the Light of Historical Research. 3rd ed. London: Hodder and Stoughton, 1919. ———. A Short Grammar of the Greek New Testament, for Students Familiar with the Elements of Greek. London: Hodder and Stoughton, 1908. Sollamo, Raija. “Some ‘Improper’ Prepositions, such as ἐνώπιον, ἐναντίον, ἔναντι., etc., in the Septuagint and Early Koine Greek.” VT 25.4 (1975) 773–82. Spada, Nina, and Patsy M. Lightbown. “Second Language Acquisition.” In An Introduction to Applied Linguistics, edited by Norbert Schmitt, 104–23. 2nd ed. London: Hodder and Stoughton, 2010.

66

Biblical and Ancient Greek Linguistics 6

Swanson, James. Dictionary of Biblical Languages with Semantic Domains: Greek (New Testament). Oak Harbor, WA: Logos Research System, 1997. Swete, Henry Barclay. The Apocalypse of St. John. 2nd ed. New York: Macmillan, 1907. Thackeray, Henry John. A Grammar of the Old Testament Greek According to the Septuagint. Vol. I: Introduction, Orthography, and Accidence. Cambridge: Cambridge University Press, 1909. Thompson, Steven. The Apocalypse and Semitic Syntax. Cambridge: Cambridge University Press, 1985. Turner, Nigel. A Grammar of New Testament Greek. Vol. 3, Syntax. 3rd ed. Edinburgh: T&T Clark, 1963. Turner, Nigel. A Grammar of New Testament Greek. Vol 4, Style. Edinburgh: T&T Clark, 1976. Vallae, Lavrentii. Viri Tam Graecae Quàm Latinae Linguae Doctissimi, in Nouum Testamentu[m] Annotationes. Basil: Cratander, 1526. Wallace, Daniel B. Greek Grammar Beyond the Basics. Grand Rapids: Zondervan, 1996. Webb, Joseph M., and Robert Kysar. Greek for Preachers. Atlanta: Chalice, 2002. Winer, Georg Benedikt. Grammar of the Idiom of the New Testament. 7th ed. Translated by Gottlieb Lünnemann. Andover, MA: Warren F. Draper, 1886. ———. “De Solecismis qui in Apocalypsi Joannea Inesse Dicuntur.” In Exegetische Studien, 144–58. Leipzig: C.H.F. Hartmann, 1827. Young, Richard A. Intermediate New Testament Greek: A Linguistic and Exegetical Approach. Nashville: Broadman & Holman, 1994.

[BAGL 6 (2017) 67–98]

IDENTIFYING BARRIERS TO UNDERSTANDING: USING HILL’S MATRIX TO EXAMINE CONTEXTUAL MISMATCH IN ACTS 12:15 Jacob Bullock Pacific Institute of Languages, Arts, and Translation Ukarumpa, Eastern Highlands Province, Papua New Guinea Abstract: Relevance Theory offers historical-grammatical interpretation a model of human communication that aids in clarifying the reason modern audiences inappropriately apply their own context to a biblical text. Hill’s matrix, drawing on the model proposed by Relevance Theory, is a tool allowing expositors to explore the inappropriate context readers apply to the biblical text. Hill’s matrix can aid interpreters in the discernment of assumptions as appropriate or inappropriate to apply to a text in a search for authorial meaning. Applying Hill’s matrix to Acts 12:15 an exegete can identify both inappropriate assumptions modern American readers bring to the text as well as those contextual assumptions needed to find authorial meaning which are missing from modern readers’ context. (Article) Keywords: Acts 12, historical-grammatical, New Testament backgrounds, interpretation, Relevance Theory, angel, communication theory.

1. Introduction Teachers of biblical interpretation often use such slogans as “A text without a context becomes a pretext for a prooftext.”1 In his widely used work, The Hermeneutical Spiral, Grant Osborne portrays the centrality of the study of context for hermeneutics: “I tell my classes that if anyone is half asleep and does not hear a

1.

Carson, Showing the Spirit, 51.

68

Biblical and Ancient Greek Linguistics 6

question that I ask, there is a 50 percent chance of being correct if he or she answers ‘context.’ The term itself covers a vast array of influences on a text.”2 Modern historical-grammatical interpretation has focused on the historical, cultural, and linguistic context of texts.3 Practitioners accept that an understanding of context is a vital foundation for understanding authorial meaning.4 However, because many who practice this method of interpretation are unaware of the advances modern linguistics has made in understanding the dynamics of human communication, practitioners have not systematically examined a portion of the original context of the biblical text. Although hermeneutic efforts up until the present time have touched many of modern communication theory’s facets, many interpreters and expositors of the biblical text have not made use of all of the tools and theoretical understanding it offers. Relevance Theory offers practitioners of historical-grammatical interpretation new insights. The interpretative matrix proposed by Harriet Hill is one method for reaching such insights. This matrix uses Relevance Theory as the foundation for a diagnostic structure designed to identify mismatches between the perceived shared contextual assumptions of the secondary audience of modern readers and the original authors of the biblical text, along with their intended primary audience. The secondary audience I have chosen for comparison with the primary audience is that of modern lay readers in twenty-first century American culture.5 Following a short introduction to Relevance Theory and Hill’s matrix, I will use the matrix to compare the assumptions of twenty-first century American readers approaching the biblical

2. Osborne, The Hermeneutical Spiral, 39. 3. Mare, “Guiding Principles,” 19–23. 4. Fantin, “Society and Culture,” 8. 5. While modern American culture is very heterogeneous, I have chosen not to specialize the nature of the secondary audience as to any particular cultural subgroup within North America. Contextual assumptions of different subgroups can vary widely. However, as most modern English Bible translations targeted to Americans do not attempt to specialize their translation to a subgroup, I have not attempted to specialize my study to a subgroup.

BULLOCK Identifying Barriers

69

text with those of the primary audience. For this exercise, I will use the passage of Peter’s reception by the disciples after he was released from prison by the angel in Acts 12:15. I will use Hill’s matrix to diagnose mismatches in the cultural assumptions between the primary audience and the secondary modern American audience. After I present the findings of my diagnosis, I will offer suggestions for the correction of mismatches in assumptions. These corrections will include both assumptions the secondary audience falsely believes it shares with the primary audience and assumptions unknown to the secondary audience whose introduction might help them find relevance in the text. I will address these corrections based on the strength with which they are held by the secondary audience and how strongly these mismatches need to be addressed by expositors helping the secondary audience interpret the biblical text. 2. An Overview of Relevance Theory The historical-grammatical model for biblical interpretation and the work of Bible expositors can be enriched by applying the insights of Relevance Theory. As noted above, much of the emphasis of the historical-grammatical model of interpretation is understanding meaning within the contextual backing of the content of Scripture. As a method, it continues to be enriched by advances in the field of linguistics. In the last few decades, biblical interpretation has benefitted from increased understanding and application of linguistic concepts such as discourse analysis. While historical, grammatical, and discourse considerations are all indispensable tools in the study of the historical and linguistic context of the biblical text, Relevance Theory offers interpreters a tool with which to examine an element of context not yet systematically studied by many Bible interpreters. Relevance Theory reveals the context of the shared cognitive environment of the speaker or writer with their original audience. A shared cognitive environment is context that the speaker assumes he or she shares with his or her audience, including language, values, assumptions, experiences, knowledge, and

70

Biblical and Ancient Greek Linguistics 6

culture. While translators and interpreters have at times intuitively addressed this shared context between speakers and their audiences, the entirety of this context has not been a focus of deliberate consideration at the same level as strictly cultural or even grammatical or semantic analysis.6 For all interpreters of a text, their understanding of human communication affects how they go about the task of interpreting that text. The ideas biblical interpreters have about how speakers communicate meaning to their audience has an impact on how they search for meaning in the Bible.7 An interpreter’s understanding of how humans communicate is the “communication model” they use when interpreting and translating a text. The communication model used for translating many early twentiethcentury English Bible translations, such as the RSV, and much of the practice of the historical-grammatical method of interpretation in the modern era was the “sender-messagereceiver model” or “code model.”8 Described by Shannon and Weaver in 1949,9 and later reframed and expanded by Nida,10 this model states that messages including Scripture are “encoded” in language by the sender and then “decoded” from language by the recipient to find meaning.11 This theory implies that communicators encode the entirety of communication in the “linguistic content” of the message and that a proper decoding results in proper meaning. For Scripture translation, this entails decoding of the message from the original languages using grammar, syntax, semantics, and historical use of language, and then proper encoding by the translator into the language of the recipients of the translation. While often modified by historical, literary, and cultural study, the “code model” has also served as the assumed model of communication underlying much of historical-grammatical hermeneutics in the twentieth century.

6. 7. 8. 9. 10. 11.

Hill, The Bible at Cultural Crossroads, 3. Brown, Scripture as Communication, 13–16. Smith, “Bible Translation and Relevance Theory,” 13. Shannon and Weaver, The Mathematical Theory of Communication. Nida, The Theory and Practice of Translation, 22. Hill, The Bible at Cultural Crossroads, 13.

BULLOCK Identifying Barriers

71

However, because simply re-encoding a message into a receptor language often does not preserve pragmatic meaning,12 Bible translators became increasingly dissatisfied with the “code model” and by the 1960s developed a modification that focused more on meaning than merely linguistic content.13 While still based on the “sender-message-receiver” theory of communication, this “dynamic equivalence model” recognized that the linguistic content did not supply all the contextual information translators knew was necessary to find meaning in a text.14 Interest in supplying implied information in translations sprang up among translators.15 The motivation for this change in practice was to remove obstacles to the receptor language communities’ understanding, acceptance, and ultimate use of Scripture.16 The “dynamic equivalence model” itself was later revised into the “functional equivalence model.”17 While these meaning-based translation approaches compensated for a secondary audience’s lack of contextual information, translators and interpreters lacked a strong, guiding linguistic understanding of how meaning is communicated, outside of the purely linguistic content of a message. A breakthrough came with the work of Sperber and Wilson. In Relevance: Communication and Cognition, they challenged the idea that meaning is encoded entirely in the linguistic content of the message itself and suggested that, instead, communication is rooted in an “appeal to shared context (cognitive environment)” between the sender and receiver of a message.18

12. “Pragmatic meaning” is the meaning of a message apart from merely the sum total of its linguistic content. In other words, “pragmatic meaning” is the total meaning purposed by a speech act and not only what is being said with words. I will discuss this at length below. 13. Beekman and Callow, Translating the Word of God, 22. 14. Nida and Reyburn, Toward a Science of Translating. 15. For a full discussion see Beekman and Callow, Translating the Word of God, 19–32. 16. Larson, Meaning-Based Translation, 24–25. 17. Kerr, “Dynamic Equivalence,” 5. 18. Hill, The Bible at Cultural Crossroads, 3.

72

Biblical and Ancient Greek Linguistics 6

This shared context is called the “shared assumptions” of the speaker and the audience.19 The implications of this research revolutionized the understanding of human communication. This theory of “relevance” states that when they communicate, people use the linguistic content of the message, the words, “to appeal to shared context” with their hearers. Recipients of a message decode the linguistic content of an utterance, but ultimately find meaning (relevance) in the message by searching the common context of knowledge, values, and experiences they believe they share with the communicator and then use the two together to “infer” meaning from the words.20 Hearers go to the trouble of decoding messages and searching for relevance because there is a reward (cognitive effect) in finding meaning. However, the search for meaning has a real “cost” involved for an individual’s mind.21 People want to find meaning, but due to the processing costs involved, they will only search as far as needed to do so.22 If a recipient searches for meaning without success, they will only continue searching up to a certain point before they give up on finding it in the message. The cost has become too high for them. For example, many who begin a Bible reading plan end up giving up in the middle of the Mosaic Law or the Major Prophets. It is fatiguing to seek meaning in poetry or commands without the necessary shared context with the author in which to find meaning (relevance). The processing cost becomes too high, and the individual may stop the reading plan or skip ahead to a part of the Old Testament in which they can more easily find meaning. However, when the reward (cognitive effect) of finding relevance seems to be high, even if the processing is costly, it can inspire ardent persistence to find meaning. Seminary students reading New Testament documents in the original Greek labor to expand their shared context in terms of the usage of vocabulary and grammar, as well as background information to

19. 20. 21. 22.

Sperber and Wilson, Relevance, 2. Sperber, “Understanding Verbal Understanding,” 192. Sperber and Wilson, Relevance, 132. Sperber, “Understanding Verbal Understanding,” 189–91.

BULLOCK Identifying Barriers

73

the text, to aid their search for meaning in the Greek text. Because most do not know the Greek language well, and the cultural concepts and idioms involved are unfamiliar, the effort to find meaning is very costly. Though the processing cost is high, the individual may consider the cognitive effect worth the effort it takes to find meaning. When a student’s time is more limited and their understanding of Greek language, culture, and idiom (the context they have worked hard to gain) is degraded by disuse, they may have an increasing need for secondary tools, such as parsing guides, to find meaning in the text. The processing cost may become once again too high, and they will give up the search for relevance. They may return, once again, to relying on a translated English text, rich in low-cost meaning. When processing a message, people assume that it has meaning (relevance). When receiving a message, receivers apply contextual assumptions to the message until they find meaning. Even if the search for meaning was successful, it may be unsatisfying if gaps in contextual understanding lead to a reliance on guesswork. We may harbor doubts that the meaning we have found is what the speaker intended.23 We show this tendency when we accept an interpretation of Scripture we do not find entirely convincing because the search for more satisfying, convincing meaning seems fruitless or not worth the processing cost. Moreover, if the meaning we have found from our search for relevance is different from the meaning intended by the speaker, communication has ultimately been a failure. The purpose and meaning of the sender, their “communicative intent,” has failed. For example, some modern American readers find relevance in Jer 10:3–424 by seeing it as a reference to the modern tradition of a Christmas tree.25 However, this is an 23. Sperber, “Understanding Verbal Understanding,” 194. 24. “For the practices of the peoples are worthless; they cut a tree out of the forest, and a craftsman shapes it with his chisel. They adorn it with silver and gold; they fasten it with hammer and nails so it will not totter” (Jer 10:3–4 NIV). 25. This interpretation, held by some American Protestants, was used by American political news commentator, Joy Reid, to critique the American politician, Sarah Palin’s possession of a Christmas tree. Reid, “The Ed Show.”

74

Biblical and Ancient Greek Linguistics 6

anachronism; the original audience would never have found relevance in this meaning, as Christmas trees were an unknown concept and not part of the cognitive environment they shared with the speaker. They would have understood this as a clear reference to the practice of idolatry.26 In looking for relevance, hearers first access more frequently used parts of their context to find meaning.27 The original communicator was signaling a familiar concept to the original audience: idols made from wood. Some modern American readers, in their search for relevance, access a different cognitive environment, one in which idols exist as a fuzzy concept. The modern American readers’ search for relevance initially results in a Christmas tree with such a dissatisfying cognitive effect that many will immediately cast off this notion and search further for meaning despite the cost. Another example of inappropriately applied context is the reaction of many church members of African polygamous societies to 1 Tim 3:2, Paul’s instruction that an elder must be the “husband of one wife.”28 In polygamous societies, people often immediately apply their context of polygamy to this passage, seeing Paul’s instructions as a prohibition of polygamy. This results in high cognitive effect and, because they immediately see application of this meaning, they search no further. Further search for meaning would not seem worth the cost.29 Since polygamy was not practiced in Hellenistic society and is strictly forbidden in surviving marriage contracts from the era, it is more likely these instructions were a directive to sexual purity in light of practices such as temple prostitution and other sexually immoral behavior practiced in the wider culture.30 In this example, the communicative intent of the speaker, Paul, has failed in regards to the secondary audience. Speaker meaning has

26. 27. 28. 29. 30.

Thompson, The Book of Jeremiah. Sperber and Wilson, Relevance, 77. Hill, The Bible at Cultural Crossroads, 20. Sperber and Wilson, Relevance, 142. Knight, The Pastoral Epistles, 158–59.

BULLOCK Identifying Barriers

75

been thwarted due to the application of inappropriate contextual assumptions.31 While secondary audiences often apply wrong assumptions about context, wrong assumptions rarely lead to wrong meaning in primary communication, where speakers are able to provide immediate feedback to correct false assumptions.32 For example, if I had arranged with my wife to get milk from the store on my way home and when I come in the door she says to me “Did you get it at the store?” I know that she is referring to the shared context of our earlier communication. However, if one day, unbeknownst to me, my neighbor wins the lottery and I come home and my wife says, “Did you hear what happened today?” my mind may search in vain for relevance in her statement. I may only search for meaning in the context of the trivial matters of the day and find no satisfying meaning. Having failed to reach a satisfying conclusion to my search, I may quickly give up the search and say, “No, I didn’t. Please tell me.” Alternatively, trying a new context, I may falsely assume she means the news report I heard on the radio about a fire and say to her, “Yes, I heard it was a great tragedy.” However, my response may puzzle her since the context of her message was that our next-door neighbor has won the lottery. If communication clearly fails, then in primary communication the sender of the message may supply immediate clarification of the message.33 The sender may supply the missing context. When the missing context (a neighbor’s good fortune) is supplied, the incorrect context of a disaster in another city is shown to be false, and communication succeeds. Because successful communication of meaning is a result of shared contextual assumptions, there are fewer occurrences of communication failure resulting from falsely assumed shared context with people we know well; our cognitive environments

31. Keane states that in the realm of religious communication because they often do not share the same context as the primary audience hearers are particularly apt to apply inappropriate context to a message. See Keane, “Religious Language,” 47. 32. Hill, The Bible at Cultural Crossroads, 38. 33. Hill, The Bible at Cultural Crossroads, 24.

76

Biblical and Ancient Greek Linguistics 6

have more in common. The fewer assumptions the hearer shares with the speaker, the more likely communication is to fail. Falsely assumed shared cognitive environment, therefore, is often the basis for intercultural misunderstandings. While individuals hold a variety of assumptions, not all of the assumptions have the same qualities. Assumptions can be divided between assumptions of what is true and assumptions of what is not true.34 A man may assume the truth that all the members of his family are humans. This is an example of an assumption of truth. He may also assume the falsity of his ability to walk on water. This is an assumption of something that is not true. Both of these assumptions are part of his cognitive environment. In addition to assumptions of truth and falsity, individuals also have differing degrees of strength with which they hold each assumption. Not all assumptions are held to the same degree of strength.35 Each of the assumptions the man holds in the previous example, those of his family being human and him being unable to walk on water, are strongly-held assumptions. It would be extremely difficult for him to receive a message or have an experience that would alter those assumptions. However, for the same man, the assumption that his children will be there when he gets home from work may be a more weakly-held assumption. It may easily be changed by a message from his wife saying his children will visit their grandparents for the weekend. He may not need much convincing to change his assumption. He may readily alter it, based on his perception of the trustworthiness of his wife. If he has a strong assumption that his wife is trustworthy in what she communicates to him, that assumption combined with the linguistic content of the message and other assumptions changes the weakly-held assumption that he will see his children upon returning from work. His wife does not have to keep presenting evidence to prove the truth of her message. His assumption is easily changed. However, if his wife phoned him at work and explained that she has discovered that their daughter is, in fact, a

34. Hill, The Bible at Cultural Crossroads, 3. 35. Wilson and Sperber, Meaning and Relevance, 209.

BULLOCK Identifying Barriers

77

cat, it will take some convincing to triumph over the stronglyheld assumption that his daughter is not. The implication of this is that strongly-held assumptions must be challenged strongly if they are to change. Weakly-held assumptions may be easily changed.36 3. How Relevance Theory Aids the Goals of HistoricalGrammatical Interpretation For purposes of Scripture interpretation, the significance of Relevance Theory has been underappreciated. While historicalgrammatical interpreters have long recognized the need for context to understand authorial meaning, context has often been limited to grammatical, literary, semantic, and cultural aspects of the communication.37 According to Sperber, the wider background of culture and language use is only part of the shared context of the sender and recipient of a message.38 The cognitive environment a sender assumes they share with their audience comprises more than culture or grammar; it includes all shared values, experience, and knowledge.39 This includes any personal history a speaker might have with their audience. For example, interpreters have long pondered the significance of the alternation of the verbs ἀγαπάω and φιλέω in the exchange between Peter and Jesus in John 21. While we can analyze the passage linguistically, culturally, semantically, and literarily, the cognitive environment that Peter and Jesus assume they share with each other contains far more. Its entirety is ultimately inaccessible to interpreters. Translators cannot fully access the shared personal history between Jesus and Peter. Because it is inaccessible, secondary readers may falsely assume they share more of the author and primary audience’s mutual cognitive environment than they do, and thereby find relevance in the exchange of Peter and Jesus that the Evangelist never intended.

36. 37. 38. 39.

Hill, The Bible at Cultural Crossroads, 5. Osborne, The Hermeneutical Spiral, 37–56. Sperber, “Understanding Verbal Understanding,” 192. Sperber and Wilson, Relevance, 2.

78

Biblical and Ancient Greek Linguistics 6

This is the same as the failure of the polygamous cultures in interpreting 1 Tim 3:2 and some American audiences in interpreting Jer 10:3–5. Each secondary audience has found meaning through the application of contextual assumptions, but the meaning in each case has not been the meaning the author intended their audience to find. Rather than being discouraged by the inaccessible aspects of the cognitive environment shared by speaker and primary audience, we should find motivation to systematically identify and correct our secondary-audience tendency to apply inappropriate contextual assumptions to the biblical account, assumptions that lead us to satisfying yet erroneous meaning in the text. However, we cannot address contextual assumptions haphazardly and risk missing errors that need correction. Practitioners of the historical-grammatical hermeneutic must have a tool for identifying and correcting mismatches between the cognitive environments of original speakers and the modern communities for whom they interpret. While it is ultimately an impossible task in the modern community to fully emulate the cognitive environment of the speaker and the primary audience, every effort interpreters make to correct error increases the accuracy of their community’s ultimate understanding. 4. Hill’s Matrix: Applying Relevance Theory to Biblical Interpretation Harriet Hill supplies us with a much-needed tool in her practical work, The Bible at Cultural Crossroads, which is an application of the implications of Relevance Theory to the task of biblical interpretation and translation. In it, she provides a theoretical matrix with which a secondary audience can weigh their cognitive environment against what can be known of the context and cognitive environment of the original audience.40 This matrix helps an interpreter to see which aspects of the environments of the primary and secondary audiences are mutual, which aspects

40. Hill, The Bible at Cultural Crossroads, 193.

BULLOCK Identifying Barriers

79

the secondary audience falsely assumes to be mutual, and which are shared by the secondary audience without their knowledge. Hill’s matrix clarifies for the secondary audience those aspects of its cognitive environment that are inappropriate to apply to the biblical text and which aspects of the primaryaudience’s cognitive environment they lack and therefore need to consider. In other words, it shows an interpreter which aspects of their cognitive environment are in line with that of the primary audience, and which need to be challenged as inappropriate to apply to the text. In addition, it reveals elements the interpreter lacks from their own cognitive environment, which must be incorporated into their own interpretation of the text. For those preparing to teach the biblical text, the matrix reveals the assumptions their community brings to the text, which may need to be challenged. It also reveals assumptions that need to be introduced by the expositor, so that a correct interpretation can be made. In primary communication, the sender of the message can take an active part in correcting failed communication. This is not so with a secondary audience.41 The secondary audience, or those who interpret for them, must do the work of expanding their cognitive environment to match that of the speaker and primary audience as closely as possible. The first step an interpreter must take in the process of expanding a secondary audience’s cognitive environment is that of correcting the misassumptions a secondary audience brings to a text. The second step an interpreter must take is to recognize those aspects of the two cognitive environments that are shared between the two and therefore need no correction. There are two types of shared assumptions: those the secondary audience knows are shared, and those that they do not know are shared. After having identified the shared elements of cognitive environment, an interpreter must identify those elements of cognitive environment not shared between the secondary and primary audiences.42 These unshared assumptions

41. Hill, The Bible at Cultural Crossroads, 38. 42. Hill, The Bible at Cultural Crossroads, 37.

80

Biblical and Ancient Greek Linguistics 6

also divide into two types, those that the secondary audience falsely assumes are shared and those held by the primary audience of which the secondary audience is unaware.

Actually Shared

Hearer Thinks It Is Shared

Hearer Does Not Think It Is Shared

Quadrant 1: Intended Context (Assumed Shared, Is Shared)

Quadrant 2: Unrecognized Context (Do Not Know, Is Shared)

Not Quadrant 3: Actually Unintended Context Shared (Assumed Shared, But Is Not)

Quadrant 4: Missing Context (Do Not Know, Is Not Shared)

Table 1: Hill’s Mutual Cognitive Environment Matrix43

In Table 1, Quadrant 1 represents those contextually shared assumptions the secondary audience correctly assumes that they share with the primary audience.44 Much of the context of communication is centered on normal human experience: the need to eat, sleep, have shelter, etc. These normal human experiences make up a large portion of shared context between primary and secondary audiences. Additionally, when interpreting the biblical text, shared cognitive environment may be the result of the influence of Scripture on the culture, the church, or shared cultural ideals. For example, American culture has been highly influenced by a history of exposure to Scripture; therefore, concepts such as monotheism are common to the American mind. Quadrant 2 represents the unrecognized shared elements of cognitive environment that the secondary audience shares with the primary audience.45 As a Bible teacher in Papua New Guinea, I know that many of the cultures I work with have practices for the redemption of land, payment of debts, or levirate marriage carried out by next of kin that are similar to the ancient Hebrew

43. Hill, The Bible at Cultural Crossroads, 29. 44. Hill, The Bible at Cultural Crossroads, 39. 45. Hill, The Bible at Cultural Crossroads, 28.

BULLOCK Identifying Barriers

81

practices seen in the Old Testament. These similar practices have similar motivation in both cultures, to keep property and children within a line of patrilineal descent. However, until the similarities in these practices are explicitly pointed out to them, many of my students do not realize that they, as a secondary audience, share these assumptions and cultural motivations with the primary audience. They are elements of shared cognitive environment of which they are unaware. Quadrant 3 shows assumptions that the secondary audience has about the context of the message that were not assumptions held by the primary audience.46 They are “unintended” assumptions which the speaker never intended be applied to their message to find meaning. These are the contextual assumptions that the secondary audience thinks they share with the primary audience but that in reality are not shared.47 In the earlier example, the assumption by polygamous cultures that Paul is speaking into a polygamous culture in 1 Tim 3:2 would be placed in this quadrant. Because the audience assumed this familiar element of their cognitive environment was shared, they found relevance in applying it. Often a secondary audience finds relevance through these false assumptions and so they are not challenged as false. Even though communication of the message has ultimately failed, the secondary audience is unaware of the failure because they have some cognitive benefit from their search for relevance. They see no need to challenge their views. The biblical interpreter must take great care to correctly identify the falsely assumed shared context of Quadrant 3. This will become clear in the application of this diagnostic matrix to Acts 12. Quadrant 4 contains those contextual assumptions which are not shared between the primary and secondary audience and of which the secondary audience is unaware.48 Modern readers of the text are unaware of these assumptions. These must be filled

46. Hill, The Bible at Cultural Crossroads, 27. 47. Hill, The Bible at Cultural Crossroads, 41. 48. Hill, The Bible at Cultural Crossroads, 45.

82

Biblical and Ancient Greek Linguistics 6

in for the secondary audience before they can find speakerintended meaning. Without these contextual assumptions, relevance cannot be found and the audience may give up the search for meaning in a message or may settle for incomplete meaning. The previous example about the fatigue experienced by many modern readers as they read the Mosaic Law illustrates that when modern readers lack the contextual assumptions necessary to find relevance in a text, the result is increased processing cost. When modern readers proceed without the necessary assumptions for understanding, they may ultimately give up on finding relevance. Those seeking to read the biblical text while lacking the contextual background to understand the significance of the author’s message often face a fruitless, fatiguing search for meaning. 5. Practical Application of Hill’s Matrix for Interpreters and Expositors of the Biblical Text Applying Hill’s matrix can identify shared and unshared elements of cognitive environment between the primary audiences of Acts 12:15 and the secondary audience of twentyfirst century readers of the biblical text in the United States. The goal of this exercise is to show the effectiveness of Hill’s matrix in the application of Relevance Theory to identify and classify assumptions held by American Bible readers approaching the biblical text. This will show that Hill’s matrix can be practically applied by interpreters when preparing for exposition of the biblical text. 5.1 Acts 12:15 as a Test Case for Analysis The events of Acts 12:15 can illustrate the usefulness of Relevance Theory in general and Hill’s matrix specifically for accurate historical-grammatical interpretation. Acts 12:11–15 in the NIV states, Then Peter came to himself and said, “Now I know without a doubt that the Lord has sent his angel and rescued me from Herod’s clutches and from everything the Jewish people were hoping would

BULLOCK Identifying Barriers

83

happen.” When this had dawned on him, he went to the house of Mary the mother of John, also called Mark, where many people had gathered and were praying. Peter knocked at the outer entrance, and a servant named Rhoda came to answer the door. When she recognized Peter’s voice, she was so overjoyed she ran back without opening it and exclaimed, “Peter is at the door!” “You’re out of your mind,” they told her. When she kept insisting that it was so, they said, “It must be his angel.”

For the purpose of this analysis, I am operating on the interpretation that when the occupants of the house responded to Rhoda, they were indicating to her that it could not be Peter but instead must be his “guardian angel,” a concept established as being part of first-century Jewish thought from both literary and cultural influences in the intertestamental period.49 This interpretation states that the believers in the house praying for Peter believed that the voice at the door was, in fact, Peter’s guardian angel, who had assumed the timbre of his voice to speak with them. An early example of this type of angelic impersonation within second temple Jewish folk religion is the portrayal of a kinsman of Tobias by Raphael as recorded in Tobit 5:4. Later rabbinic literature, such as the Chronicles of Moses and Genesis Rabba, also contain examples of this belief of the angelic impersonation of humans.50 While both of these examples are not synchronous with the first century, there are documents of a more contemporary nature that illustrate the concept of angelic impersonation as being present in folk beliefs about angels among Jews in the first century.51

49. There are at least two divergent views from the interpretation of the phrase ἄγγελος αὐτοῦ that I have chosen to espouse for the scope of this paper. One of these is that the occupants believe ἄγγελος αὐτοῦ is a reference to the departed spirit of Peter after his execution. Polhill expounds this view. See Polhill, Acts, 282. The other is the view that this phrase refers to Peter’s human messenger sent from him in prison to the church. In his exposition, John Gill called this view, not widely accepted in the modern era, “untenable.” Gill, Exposition of the Old and New Testaments, 891. 50. Witherington, The Acts of the Apostles, 387. 51. Rapske compares the assumptions of the disciples with the apocryphal work Acts of Andrew where “the LORD” appearing as Andrew accompanies Iphidamia to the prison where Thomas is kept. See Rapske, The

84

Biblical and Ancient Greek Linguistics 6

Moulton expounded this view in the modern era.52 Charles supported and widely popularized Moulton’s view.53 There is general scholarly acceptance of this view and it has received support among such notable scholars as Bruce.54 With this hypothesis as a backdrop, I will use Hill’s matrix to evaluate twenty-first century American cultural assumptions about angels and identify points of commonality and contrast in the two cognitive environments. I will then analyze the findings, place them within the four quadrants of Hill’s matrix and proceed to discuss the findings of the diagnostic. 5.2 Assumptions about Angels in the Twenty-First Century American Cognitive Environment Angels are a popular concept in modern America. Often, Americans have highly-developed assumptions about angelic beings, based on the cultural and media influences surrounding them. While concepts of the angelic in popular American thought are often reflective of biblical influence upon the culture, information garnered only from the Bible is insufficient to cover the range of ideas presented in American popular media. As biblical literacy decreases, people have an abundance of books, comics, music, motion pictures, and internet sites that comprise their concept of the angelic. The challenge we face as modern American interpreters of the biblical text is that angels figure largely in the daily life of American culture. In America, there are widely-accepted folk religious beliefs about angels shaped by a meshing of religious ideas from popular media,55 New Age beliefs, and Christian

Book of Acts, 415. 52. Moulton, “IT IS HIS ANGEL,” 516. 53. Ferguson, “Angels of the Churches in Revelation 1–3.” 54. Bruce, The Acts of the Apostles, 286. 55. Folk Religion as defined by Yoder as “the totality of all those views and practices of religion that exist among the people apart from and alongside the strictly theological and liturgical forms of the official religion” (“Toward a Definition of Folk Religion,” 14).

BULLOCK Identifying Barriers

85

influences.56 In modern America, Ombres reflects that it is likely easier for people to believe in angels than to believe in Jesus.57 Wiens observes that, Belief in angels has truly become popular in our Western society at large. But while the popular interest and belief in angels grows, it is doubtful that real understanding of angels has increased. In fact, it may be that all of the popular interest in the subject may actually leave people more confused than ever on the meaning of this word.58

In twenty-first century America, angels are often portrayed or discussed in film, television, popular literature, and daily conversation.59 They are widely portrayed as the result of a departed human soul who, because of their good deeds on earth, has become an angel in heaven.60 Angels are pictured as gendered and it is common in fictional media to see romantic interaction between humans and angels.61 Angels in popular American folk belief take many forms, including “warriors, rebels, intermediaries, comforters, protectors, and guides.”62 But with the individualism inherent in American culture, the focus is often on the role of angels within the lives of individual humans.63 They are portrayed as not only providing protection, but also doing battle on behalf of individuals.64 Draper and Baker in Angelic Belief as American Folk Religion observe that angels have a near-universal, cross-religion appeal because of their benevolence and affirmation of the worth of the individual.65 Though not American, Pope Francis, being the leader of the Roman Catholic Church in America, recently urged all American Catholics to commune and receive guidance from their “guardian angel” and to show respect for their

56. 57. 58. 59. 60. 61. 62. 63. 64. 65.

Draper and Baker, “Angelic Belief,” 624–25. Ombres, “God, Angels and Us,” 51. Wiens, “Angels,” 230. Rovano, “Angel as a Fantasy Figure,” 58. Newton, “Angels,” 6. Rovano, “Angel as a Fantasy Figure,” 58. Draper and Baker, “Angelic Belief,” 637. Rovano, “Angel as a Fantasy Figure,” 58. Rovano, “Angel as a Fantasy Figure,” 59. Draper and Baker, “Angelic Belief,” 624.

86

Biblical and Ancient Greek Linguistics 6

presence.66 In this same broadcast, the religious faithful were warned that this personal angel is witness to their secret sins. Though American culture has a diminishing association with the biblical text, it seems to be maintaining and expanding its folk theology of the angelic. Angels are popular, and beliefs about their nature and role are diverse and strongly-held assumptions. While American assumptions about angels continue to change and grow, we can synthesize them into the following overarching concepts: angels are seen in American culture as gendered, sexual beings, who, while ultimately given tasks to fulfill, can operate out of freewill and outright rebellion. They interact on a personal level and engage in friendship and romance with individual humans. In addition, they have personal assignments as guides, comforters, protectors/defenders, message bearers, and record keepers of human sin. Angels are also popularly viewed as the posthumous state of moral humans. Because angels are such a common cultural image, and beliefs about them are often rooted in spiritual conviction, assumptions surrounding them are often strongly-held. For the purpose of analysis, I will focus on the American assumption of angels as self-willed comforters, guides, protectors, guardians, and warriors. I will also examine and contrast the American assumption that angels are the posthumous state of moral humans. 5.3 Assumptions about Angels in the Cognitive Environment of Second Temple Jews in Palestine The popularization of angels is not a modern phenomenon. A similar popularizing of angels happened in Second Temple Judaism.67 Folk belief in angels in Second Temple Judaism was also highly-developed. Interest in angels revolved around their roles of message bearers, warriors, protectors, and the agents of God’s rule. Much intertestamental literature deals with the angelic and much of the early church’s belief in angels was

66. Pope Francis, “Respect and Listen to Your Guardian Angel,” [n.d.]. 67. Macumber, “Angelic Intermediaries,” 5.

BULLOCK Identifying Barriers

87

shaped by Jewish beliefs strengthened by these literary works.68 An example of the image of angels presented by intertestamental literature is the compelling figure of Raphael in Tobit 5:4 and 8:2–3. Raphael’s role as guardian and guide to Tobias exemplifies the role of angels as guides and protectors of the righteous. This concept figures strongly in intertestamental writings. Moulton’s view is that this preoccupation with the angelic in the intertestamental period had a profound impact on popular beliefs about angels in first-century Palestine. He goes on to theorize that many of these beliefs about the angelic were in some way influenced by contact with the belief systems of the Persians during the exile.69 Whatever the cause, there is ample evidence to suggest that the cultural assumptions of Palestinian Jews in the first century were not entirely based on canonical Old Testament accounts, but instead included popular culturallyheld elements not derived from the accounts of Scripture. We must not make the mistake of thinking the Jewish cultural writings of the this era were ever elevated to the status of the accepted Hebrew canon, but we must realize that, as an influence, they were essential to the daily thought and practice of the Jews in Palestine. As first-century Jewish society was diverse and had myriad literary and philosophical influences outside of the biblical writings, the task of establishing the assumptions of the original audience is complex.70 Despite this, much can be learned from the study of intertestamental and first-century writings. These literary sources, such as the apocryphal and pseudepigraphal works, portray angels as rulers and champions of nations.71 Often the role and actions of angels in the text underlines Israel’s privileged status. Many writings picture Israel’s patron angel Michael triumphing over the rebellious angels of the other nations.72 However, angels do not only function as guardians at

68. 69. 70. 71. 72.

deSilva, “Apocrypha and Pseudepigrapha,” 64. Moulton, “IT IS HIS ANGEL,” 520–21. deSilva, “Apocrypha and Pseudepigrapha,” 63–64. Stuckenbruck, “Angels of the Nations,” 30. Davidson, Angels at Qumran, 75.

88

Biblical and Ancient Greek Linguistics 6

the national level; there are also examples of the personal role of angels, including Raphael’s depiction in Tobit 12:11–20. Delineation between the angelic and the demonic became more defined during intertestamental times.73 By the first century, there was a strong sense of moral dualism between fallen and unfallen angels. Fallen angels are depicted as those in rebellion to God; unfallen angels are depicted as the warriors of light defending the righteous.74 Angels, in the intertestamental period, were seen as more than just guardians. The image of angels as bearers of messages and bringers of revelation from God was very defined and accepted by the first century.75 Angels were anticipated as the bearers of messages from God to humans.76 The book of Jubilees, a work of intertestamental Jewish literature, portrays angels as the bringers of the law to Moses on behalf of God.77 In addition to their previously mentioned roles, intertestamental literature depicts angels as agents of divine power, ruling the nations at the command of Yahweh.78 Angels served as regents of the nations in the spiritual realm. This idea may have sprung partly from ancient Canaanite influences and partly from the biblical account, in the book of Daniel, of Michael’s struggle with the “Prince of Persia”79 Second Temple Judaism had a highly developed belief in angels as the agents of God’s work. Angels are pictured as giving the law, doing the work of creation at the direction of God and as God’s intermediaries in the bringing of judgement on the

73. Xeravits, Dualism in Qumran, 5. 74. Davidson, Angels at Qumran, 302. 75. Najman, “Angels at Sinai,” 315–16. 76. We see the image of angels as messengers reflected in Paul’s warning in Gal 1:8 that even an angel bringing another gospel should not sway the Galatians. In addition, angels function widely as messengers in the Apocalypse. 77. Najman, “Angels at Sinai,” 320. 78. Davidson, Angels at Qumran, 266. 79. Heiser, The Divine Council, 14.

BULLOCK Identifying Barriers

89

nations.80 The Jewish writer Philo portrays angels as the ultimate agents in the creation of Adam at the direction of God.81 We can summarize folk belief in angels in Second Temple Judaism as follows. Angels were seen as agents of God’s will; they were spiritual beings whose roles included ruling, protecting, message bearing, and championing the cause of the righteous. Angels existed in one of two moral states: either completely obedient to the will of God or in complete rebellion against God. They function as the agents of God in bringing his will upon the earth. 5.4 Comparing Cultural Assumptions: Filling in Hill’s Matrix Placing the two sets of cultural assumptions presented in the matrix proposed by Harriet Hill, I will address each of the assumptions in the context of where it falls within the matrix and offer insights into how some of the assumptions need to be addressed. This process will help the secondary audience expand their cognitive environment and thereby find the intended meaning of the author/speaker in their search for relevance. 5.4.1 Quadrant one: Shared contextual assumptions. Although there is significant divergence in cultural assumptions between first-century Jews in Palestine and twenty-first century American Bible readers, there are many assumptions about angelic beings which are shared by the two cultures. The assumption that angels can function as the bearers of messages between God and humans is one of these. Another is the belief that angels can act as comforters and protectors. Because angels’ function as warriors is seen in both Hollywood movies and intertestamental documents from Qumran as well as the Old Testament canon, this role is accessible in both sets of assumptions. Finally, in both popular American anecdote and Old Testament and intertestamental writings, angels are mistaken for human beings. Much of the similarity in these assumptions is likely due to the profound effect the canon of Scripture has had on modern

80. Najman, “Angels at Sinai,” 320. 81. Grabbe, An Introduction to Second Temple Judaism, 35.

90

Biblical and Ancient Greek Linguistics 6

American culture. These cultural assumptions are strongly-held and strongly-shared between the two cognitive environments and therefore they need no challenge or adjustment in the course of modern interpretation. While the outworking of these assumptions may not be identical in each context, there is enough parallel that it is unlikely an expositor will need to prove any of these assumptions to their community. An expositor does not need to expand a community’s cognitive environment but only affirm these elements of shared context. 5.4.2 Quadrant two: Shared but unrealized contextual assumptions. I will now consider assumptions that fall under Quadrant 2, those assumptions that are shared but are not realized by the secondary audience to be shared assumptions. An example of these is the impact of folk theology on angelic belief in both the primary and secondary audiences. As I demonstrated above, there existed a highly developed Jewish folk theology of angels within the first century. While the lay practitioners of the Jewish faith may have wholeheartedly adopted the ideas of cultural influence, we need not assume that the theologians of the era did not have a clear delineation of which ideas about angels were theologically orthodox and which were not. Many commentators, when expounding on this passage, feel the need to fit the statements of the people in the house to Rhoda of “it is his angel” into the framework of normative early church doctrine. However, I suggest that those in the house speaking to Rhoda were reacting out of their unchallenged folk theology of angels. It is unnecessary to suppose that the author of Acts is quoting their words as anything but an example of their unbelief that their prayers would be answered and Peter would be released. When a secondary audience understands that this account need not be prescriptive in terms of theology, it frees them to see that those in the house were awash in cultural ideas and folk belief about the nature and practice of angels, just as modern American readers are. Understanding that believers in the first century were influenced in their thinking by more than the canonical Hebrew Bible frees the interpreter to consider the theologically unorthodox elements in the primary audience’s

BULLOCK Identifying Barriers

91

assumptions, that they too were influenced by a culturally-driven folk theology. Even though the many apocryphal and pseudepigraphal books of the intertestamental era may not have been seen as authoritative, they were still the popular media of their era and were important as such.82 We know that even New Testament writers themselves felt they contained important cultural ideas from which they could draw parallels.83 The benefit for the secondary audience in understanding this shared folk theology context is that the secondary audience no longer needs to account for the disciples’ statements in Acts 12:15 in a systematic New Testament theology. Expositors who clearly state that the passage is based in folk theology can overturn the assumptions of any readers who see the Bible, and this passage in particular, as fundamentally prescriptive. As the secondary audience likely holds an assumption of the prescriptive nature of the statements in this passage, its nonprescriptive nature needs to be strongly and directly stated to the secondary audience to challenge and replace their previous assumption of its prescriptive nature. 5.4.3 Quadrant three: Contextual assumptions falsely assumed shared by the secondary audience. When the two sets of cultural assumption are compared, the most vivid divergences fall into Quadrant 3. The most potentially problematic of these assumptions is the modern folk belief of angels as the future spiritual state of people who lived a morally upright existence. This belief, which is prevalent in American folk belief and media, is not one paralleled in the assumptions of middle first-century Jews in Palestine.84 While angels were believed to occasionally appear as human beings, in no record do we see the idea that

82. deSilva, “Apocrypha and Pseudepigrapha,” 63–64. 83. One of the most notable examples of this is Jude’s use of accounts from two pseudepigraphal works as sources for illustration in his epistle. 84. Carrell, Jesus and the Angels, 80. Carrell does state that a similar belief in posthumous humans becoming angels later developed in Judaism due to the influence of surrounding cultures. However, he argues that the belief did not develop until the beginning of the second century at the earliest.

92

Biblical and Ancient Greek Linguistics 6

human beings could be transformed into one of these spiritual beings. This familiar modern American cultural assumption is easily accessed and will likely influence the secondary audience’s interpretation of Acts 12:15. Because the image of angels as the postmortem existence of righteous people is culturally very strong, it is easily assumed to be shared context between the original audience and the secondary audience. Interpreters and expositors of this passage need to give careful attention to this mismatch in cultural assumptions between the primary and secondary audience. Unchallenged, this assumption is likely to result in the American readers finding relevance, and ultimately meaning, that was never the communicative intent of those speaking to Rhoda. Many, applying this assumption, may think that those speaking to Rhoda believed that Peter, a righteous man, having been executed, had returned in angelic form to communicate with them. This interpretation follows a common typology of many modern American anecdotes about interactions people have had with angels.85 Because the folk belief of the righteous returning in angelic form is a strongly-held assumption, the challenge to it as a strongly-held assumption will need to be strong and direct. An expositor will need to directly challenge this idea as not part of the original audience’s assumptions and outside the realm of correct understanding. Another belief the secondary audience may think they share with the primary is that of angels acting of their own accord and even in self-sacrificial ways. In the intertestamental period, there was a clear delineation in Jewish thinking between fallen and unfallen angels.86 With a clear dualistic understanding of the moral nature of angels, unfallen angels never exhibit independent will in carrying out the commands of God. In intertestamental literature, angels who act of their own will are those who are acting in rebellion against the will of God.87 While this modern assumption may not skew the modern American

85. Wiens, “Angels,” 228. 86. Reimer, “Rescuing the Fallen Angels,” 234. 87. Davidson, Angels at Qumran, 48.

BULLOCK Identifying Barriers

93

reader’s interpretation of Acts 12:15, it is part of the modern folk theology of the angelic and should be part of any systematic correction of assumptions that the secondary and primary audiences do not share. 5.4.4 Quadrant four: Contextual assumptions not shared by the secondary audience and of which the secondary audience is unaware. The fourth quadrant contains assumptions which are held by the primary audience of which the secondary audience in unaware. One element of cognitive environment of which the secondary audience is unaware is the plethora of intertestamental literature and rabbinic teaching that influenced Jewish folk theology of the angelic in first-century Palestine. While it is not guaranteed that the secondary audience will ever expose themselves to these documents, it is enough to inform them that these writings exist and are one of the main influences of these folk beliefs. Because the secondary audience is unaware of these documents and their influences on primary-audience assumptions, they will need to be introduced, yet because they are not aware of them, no misconceptions need be challenged. Therefore, briefly mentioning their existence and roles as primary influences is sufficient for the secondary audience to understand these assumptions and expand their cognitive environment. Introducing these first-century assumptions will have the added effect of strengthening the modern-day audience’s assumption that the primary audience dealt with cultural beliefs about angels that were not necessarily rooted in the Old Testament literature. The results are summarized in Table 2.

94

Biblical and Ancient Greek Linguistics 6 Hearer Thinks It Is Shared

Hearer Does Not Think It Is Shared

Quadrant 1: Intended Context (Assumed Shared, Is Shared)

Quadrant 2: Unrecognized Context (Do Not Know, Is Shared)

Actually • Angels functioning as Shared comforters, message bearers, warriors, protectors, and agents of divine will. • Angels can be mistaken for humans Quadrant 3: Unintended Context (Assumed Shared, But Is Not)

• Jews in Palestine had a “folk theology” of angels apart from orthodox theology

Quadrant 4: Missing Context (Do Not Know, Is Not Shared)

Not Actually • Angels are the posthumous state • Jewish concept of guardian Shared angel rooted in intertestamental of moral humans • Unfallen angels can act of their and rabbinic literature own will or self-sacrificially Table 2. Hill’s Matrix and Acts 12:15

5.5 Results: Incidences of Mismatch in Cognitive Environments Hill’s matrix identifies mismatches in the cognitive environments of the primary and secondary audiences of Acts 12:15, which, if left uncorrected, are likely to lead to nonspeaker-intended relevance. Analysis reveals that the unshared assumption most needing to be countered for the modern American reader is the folk belief that moral humans may become angels posthumously and return to earth on tasks. Without direct contradiction, this strong modern assumption is likely to color the secondary audience’s understanding of this passage in profound ways. Both the existence of Jewish folk theology of angels and the existence of the non-canonical intertestamental literature need to be introduced to the secondary audience to expand their cognitive environment to coincide more fully with that of the first-century audience. Since these contextual elements are either unknown or unrealized, their introduction is likely to be enough for the secondary audience to adopt them. They are not

BULLOCK Identifying Barriers

95

challenging a preconceived assumption, but rather forming a new assumption in the audience’s mind. 6. What Relevance Theory and Hill’s Matrix Offer HistoricalGrammatical Interpretation Relevance Theory offers biblical interpretation a model of communication through which the factors leading to contextual mismatches between primary and secondary audiences can be better understood. While Relevance Theory is complex in its explanation of human communication, when applied through Hill’s Matrix, it becomes a tool that those without a deep understanding of linguistics can use to diagnose false assumptions of their communities and themselves about the biblical text. Hill’s method effectively provides a framework for classification of assumptions the secondary audience either falsely applies or fails to apply to the text. Uncorrected, these mismatched assumptions prevent proper understanding of author-intended meaning. Hill’s matrix has great potential as a practical tool for use by scholars and lay readers alike in preparing for exposition of the biblical text. Bibliography Beekman, John, and John Callow. Translating the Word of God: With Scripture and Topical Indexes. Grand Rapids: Zondervan, 1974. Brown, Jeannine K. Scripture as Communication: Introducing Biblical Hermeneutics. Grand Rapids: Baker Academic, 2007. Bruce, F.F. The Acts of the Apostles: The Greek Text with Introduction and Commentary. 3rd rev. ed. Grand Rapids: Eerdmans, 1990. Carrell, Peter R. Jesus and the Angels: Angelology and the Christology of the Apocalypse of John. Cambridge: Cambridge University Press, 1997. Carson, D.A. Showing the Spirit: A Theological Exposition of 1 Corinthians 12–14. Grand Rapids: Baker, 1987. Davidson, Maxwell. Angels at Qumran: A Comparative Study of 1 Enoch 1–36; 72–108 and Sectarian Writings from Qumran. London: A&C Black, 1992.

96

Biblical and Ancient Greek Linguistics 6

deSilva, David A. “Apocrypha and Pseudepigrapha.” In Dictionary of New Testament Background, edited by Craig A. Evans and Stanley E. Porter, 58–64. Downers Grove, IL: IVP, 2000. Draper, Scott, and Joseph O. Baker. “Angelic Belief as American Folk Religion.” Sociological Forum 26 (2011) 623–43. Fantin, Joseph D. “Society and Culture: Aspects of the First-Century World for a More Contextually Driven Exegesis.” BAGL 4 (2015) 7–29. Ferguson, Everett. “Angels of the Churches in Revelation 1–3: Status Quaestionis and Another Proposal.” BBR 21 (2011) 371–86. Gill, John. John Gill’s Exposition of the Old and New Testaments. Atlanta: Baptist Standard Bearer, 1810. Grabbe, Lester L. An Introduction to Second Temple Judaism: History and Religion of the Jews in the Time of Nehemiah, the Maccabees, Hillel, and Jesus. London: A&C Black, 2010. Heiser, Michael S. “The Divine Council in Late Canonical and Non-Canonical Second Temple Jewish Literature.” PhD diss., University of Wisconsin– Madison, 2004. Hill, Harriet S. The Bible at Cultural Crossroads: From Translation to Communication. Manchester: St. Jerome, 2006. Keane, Webb. “Religious Language.” Annual Review of Anthropology 26 (1997) 47–71. Kerr, Glenn J. “Dynamic Equivalence and Its Daughters: Placing Bible Translation Theories in Their Historical Context.” Journal of Translation 7 (2011) 1–19. Knight, George W. The Pastoral Epistles: A Commentary on the Greek Text. NIGTC. Grand Rapids: Eerdmans, 1992. Larson, Mildred L. Meaning-Based Translation: A Guide to Cross-Language Equivalence. 2nd ed. Lanham, MD: University Press of America, 1998. Macumber, Heather. “Angelic Intermediaries: The Development of a Revelatory Tradition.” PhD diss., University of St. Michael’s College, 2012. Mare, W. Harold. “Guiding Principles for Historical Grammatical Exegesis.” Grace Journal 14.3 (1973) 14–25. Moulton, James Hope. “‘IT IS HIS ANGEL.’” JTS 3 (1902) 514–27. Najman, Hindy. “Angels at Sinai: Exegesis, Theology and Interpretive Authority.” DSD 7 (2000) 313–33.

BULLOCK Identifying Barriers

97

Newton, Robert. “Angels.” The North American Review 282.3/4 (1997) 4–6. Nida, Eugene A. The Theory and Practice of Translation. Leiden: Brill, 1969. Nida, Eugene A., and William David Reyburn. Toward a Science of Translating. London: Tavistock, 1964. Ombres, Robert. “God, Angels and Us.” New Blackfriars 86 (2005) 48–61. Osborne, Grant R. The Hermeneutical Spiral: A Comprehensive Introduction to Biblical Interpretation. 2nd ed. Downers Grove, IL: IVP, 2006. Polhill, John B. Acts: An Exegetical and Theological Exposition of Holy Scripture. Nashville: Holman Reference, 1992. Rapske, Brian. The Book of Acts and Paul in Roman Custody. The Book of Acts in its First Century Setting 3. Grand Rapids: Eerdmans, 1994. Reid, Joy. “The Ed Show for Thursday, December 26th, 2013.” No pages. Online: http://web.archive.org/web/20140318100440/http://www.nbcnews .com/id/53993759/ns/msnbc-the_ed_show/#.Vz6VHUeNgkI. Reimer, Andy M. “Rescuing the Fallen Angels: The Case of the Disappearing Angels at Qumran.” DSD 7 (2000) 334–53. Rovano, Marcelaine Wininger. “The Angel as a Fantasy Figure in Classic and Contemporary Film.” Journal of the Fantastic in the Arts 5.3 (1993) 58– 74. Shannon, Claude E., and Warren Weaver. The Mathematical Theory of Communication. Chicago: University of Illinois Press, 1949. Smith, Kevin Gary. “Bible Translation and Relevance Theory: The Translation of Titus.” D.Litt. diss., University of Stellenbosch, 2000. Sperber, Dan. “Understanding Verbal Understanding.” In What Is Intelligence?, edited by J. Khalfa, 179–98. Cambridge: Cambridge University Press, 1994. Sperber, Dan, and Deirdre Wilson. Relevance: Communication and Cognition. 2nd ed. Oxford: Blackwell, 1995. Stuckenbruck, L.T. “Angels of the Nations.” In Dictionary of New Testament Background, edited by Craig A. Evans and Stanley E. Porter, 29–31. Downers Grove, IL: IVP, 2000. Thompson, J.A. The Book of Jeremiah. NICOT. Grand Rapids: Eerdmans, 1980. Wiens, H. “Angels.” BT 51 (2000) 224–32.

98

Biblical and Ancient Greek Linguistics 6

Wilson, Deirdre, and Dan Sperber. Meaning and Relevance. Cambridge: Cambridge University Press, 2012. Witherington, Ben. The Acts of the Apostles: A Socio-Rhetorical Commentary. Grand Rapids: Eerdmans, 1998. Xeravits, Géza G. Dualism in Qumran. New York: T&T Clark, 2010. Yoder, Don. “Toward a Definition of Folk Religion.” Western Folklore 33.1 (1974) 2–15.

[BAGL 6 (2017) 99–126]

MONOSEMY IN BIBLICAL STUDIES: A CRITICAL ANALYSIS OF RECENT WORK Ryder A. Wishart McMaster Divinity College, Hamilton, ON, Canada Abstract: This review article critically engages two recent monographs that utilize Charles Ruhl’s theory of monosemy to analyze the New Testament. After outlining Ruhl’s theory, I discuss how Gregory Fewster attempts to model monosemy within the linguistic framework of Systemic Functional Linguistics, and how Benjamin Lappenga does so within the framework of Relevance Theory. Each makes important contributions, but I argue that neither has significantly improved on Ruhl’s original model and that some of the modifications of Ruhl’s theory end up being unhelpful or unclear. Nevertheless, both authors have persuasively exhibited the usefulness of a monosemic approach to studying biblical words and texts. (Review Article) Keywords: Monosemy, Gregory P. Fewster, Benjamin J. Lappenga, Charles Ruhl, Lexical Semantics.

1. Introduction Even though the dominant trend in the lexical and grammatical traditions within biblical studies has been towards polysemy, there are dissenting voices in biblical studies.1 Stanley Porter, Gregory Fewster, and Benjamin Lappenga argue that observed variation is not inherent in the semantics of words themselves but rather a function of the context within which words occur. In other words, they agree that variation arises from contextual modulation, and that words themselves have a much more

1.

For discussion and references, see Porter, Linguistic Analysis, 51–53.

100

Biblical and Ancient Greek Linguistics 6

unified core of semantic information. In this sense, these three proponents of monosemy describe their position as a “monosemic bias.”2 In this paper, I will first introduce Ruhlian monosemy, and then assess and critique both Fewster and Lappenga’s monograph-length studies incorporating monosemy, concluding that their work indicates some intriguing possibilities for future development, despite drawbacks apparent in their appropriation of Ruhl. 2. Words, Polysemy, and Monosemy in Biblical Studies Words have always been a topic of much discussion for biblical scholars. On the one hand, lexical semantics has proven to be a fruitful area of research when engaged within biblical studies. For example, words can be analyzed in order to refine our assumptions regarding the meaning of biblical texts. On the other hand, lexical semantics has also proven to be a false foundation for many studies. Words can be invested with too much meaning, causing the proliferation of baseless claims that often take the form of posturing about the “true” meaning of certain words.3 One of the assumptions that has arguably contributed to this convoluted state of affairs is the semantic role of polysemy. If by polysemy one simply means to refer to the variation in meaning that can be observed in actual utterances, there are few who would disagree with the claim that polysemy is “the natural condition of words.”4 However, if by polysemy one means to refer to a feature inherent in the words themselves—evident in any dictionary that lists multiple distinct senses for a lexeme— then there are, as mentioned, dissenting voices.5 The basis of the monosemic approach is the work of Charles Ruhl. On his view, words do not have multiple distinct senses—utterances do. Instead, a word can be understood as a monosemous signifier,

2. Porter, Linguistic Analysis, 53; Fewster, Creation Language, 36; Lappenga, Paul’s Language, 27, 29. 3. Barr, Semantics, 107–60. 4. Geeraerts, Theories of Lexical Semantics, 42. 5. See note 1.

WISHART Monosemy in Biblical Studies

101

contributing a single, monosemous semantic meaning to each and every utterance in which it occurs. Instances of words that appear to have incommensurably different meanings, then, are better understood as figurative extensions or homonyms (i.e. two words each with single meanings), rather than as one word with multiple meanings. At bottom, this disagreement is a matter of orientation toward the data. Scholars who value systematicity and tight semantic descriptions will likely gravitate toward a monosemous perspective, whereas those who value the diversity and dynamism of polysemy are probably not going to find Ruhl convincing. Monosemy is a minimalist orientation;6 polysemy is a maximalist one.7 While neither side would claim thoroughgoing polysemy or monosemy within a lexicon—that is, some words will have one meaning and some words will have more or will be considered homonyms, regardless of the position—each side can be understood as a disposition that tends in either one direction or the other. A reengagement of Ruhl’s original monograph, I argue, is important for future incorporation of monosemy into biblical studies, and to this end I will offer an overview of Ruhlian monosemy and an example of its application before my critical analysis of Fewster and Lappenga’s recent monographs. 3. Ruhlian Monosemy Charles Ruhl’s work on lexical monosemy comprises the first attempt to theorize a programmatic explanation of lexical monosemy.8 According to Ruhl’s theory of monosemy, the

6. Ruhl, Monosemy, ix. 7. Geeraerts, Theories of Lexical Semantics, 182. 8. See the following two key works for a representative account of his work: Ruhl, On Monosemy; Ruhl, “Data, Comprehensiveness, Monosemy.” The minimalist priority represented by monosemy has an unclear origin. While Ruhl was the first to publish a significant monograph with the word monosemy in the title, he was not the first to approach linguistic analysis as a minimalist. Two others are worth mentioning here. First, William Diver’s approach— Columbia School linguistics—attempts to discover a consistent or invariant

102

Biblical and Ancient Greek Linguistics 6

lexicogrammatical meaning, or meaning potential of a sign, is a generalization of its meaning in all its contexts. I take “meaning in context” to be a sign’s apparent contribution to the contextual messages to which it contributes. If one generalizes what is common to all contextual messages that include a sign, so Ruhl articulates, one has identified the lexicogrammatical meaning of the sign, because the sign is the common denominator between all of its uses in a corpus. Ruhl describes this insight as the comprehensiveness principle: The comprehensiveness principle: The measure of a word’s semantic contribution is not accuracy (in a single context) but comprehensiveness (in all contexts).9

Providing a method to actualize this comprehensive approach, Ruhl’s hypothesis in On Monosemy is twofold: (1) “A word has a single meaning,” and (2) “If a word has more than one meaning, its meanings are related by general rules.”10 As he explains, “An meaning that accurately describes each linguistic sign in a language. Diver began to work out his theoretical approach in the 1960s, and thus the school of thought he originated constitutes a precursor to Ruhl’s work. Ruhl at one point mentions the meaning–message distinction in reference to several Columbia School proponents. See Ruhl, On Monosemy, 33. For an outline of Diver’s theory, with a description of his originality, see Huffman, “Linguistics of William Diver.” A second minimalist linguistic analysis has already been mentioned, namely, Stanley Porter’s Verbal Aspect in the Greek of the New Testament, with Reference to Tense and Mood. Porter’s approach seeks to explain the formal features of the Greek verbal system. For Porter (Verbal Aspect, 7), “formal” features are morphologically-based. Porter (Verbal Aspect, 13, 75) thus identifies morphological categories, and attempts to explain as much data as possible with little to no exceptions by postulating a single, consistent meaning for each grammatical form. Verbal Aspect was published in the same year as Ruhl’s On Monosemy, and thus these works both represent important though apparently independent advances in minimalist linguistics. Furthermore, while Ruhl addresses lexical monosemy, Porter’s work focuses on grammatical monosemy. Though Diver had originally analyzed part of the Greek verbal system, Porter’s analysis has—unlike Diver’s—stood the test of time. See discussion of Diver’s analysis in Reid, “Quantitative Analysis.” Ultimately, Ruhl’s work on monosemy, particularly his accounts of comprehensiveness and abstraction, was not unprecedented. 9. Ruhl, “Data, Comprehensiveness, Monosemy,” 172. 10. Ruhl, On Monosemy, 4.

WISHART Monosemy in Biblical Studies

103

initial presumption of monosemy does not question the existence of multiplicity; rather, it implies that current analyses find too much multiplicity too easily, and so provides a means for testing each particular claim.”11 This initial presumption Ruhl terms a “monosemic bias.” While the term bias seems to imply that monosemy is a theoretical position, a predisposition toward the data—and there is some truth to this—what Ruhl actually advocates is a method of identifying a word’s semantics. “This Monosemic Bias,” he explains, “implies a priority of research: a full detailed exploration of a word’s variant range before considering its possible paraphrase relationships with other lexical items.”12 The bias he speaks of, then, is actually a method of analysis: “Assume that any meaning that is not present in all contexts of a word is not part of the word’s inherent meaning; if this fails, assume distinct meanings are figuratively related.”13 That is, Ruhl argues that lexical analysis should proceed from the assumption of monosemy, attempting to explain observed variations in meaning by positing pragmatic mechanisms at work in actual utterances.14 An example will be useful. 3.1 Pragmatic Factorization of Ἀρχή The noun ἀρχή is assigned seven senses in BDAG:15 (1) The commencement of something as an action, process, or state of

11. Ruhl, On Monosemy, 5. 12. Ruhl, On Monosemy, 4. 13. Ruhl, On Monosemy, 234. 14. For Ruhl, the terms pragmatic and semantic are moving targets, though they always identify some kind of boundary between what is contextual and what is decontextualized to some degree. At times, he seems to be saying that semantics is concerned with invariant meaning at some rank (word, word group, sentence, paragraph, or discourse), whereas pragmatics concerns meaning beyond the rank being analyzed. At other times Ruhl describes semantics as intralinguistic meaning and pragmatics specifically as extralinguistic meaning. See Ruhl, On Monosemy, 17. However, see discussion below of Lappenga’s assessment of Ruhl’s use of these terms. 15. Ruhl frequently uses the treatments of dictionaries as an entry point when seeking a common denominator.

104

Biblical and Ancient Greek Linguistics 6

being, beginning (2) One with whom a process begins, beginning (3) The first cause, the beginning (4) A point at which two surfaces or lines meet, corner (5) A basis for further understanding, beginning (6) An authority figure who initiates activity or process, ruler, authority (7) The sphere of one’s official activity, rule, office

Each of these senses can be tentatively related to a general sense of INITIATION,16 whether that be temporal as in (1), spatial as in (4), or mental as in (5). These uses show that ἀρχή does not carry within itself the more concrete distinctions of spatial–nonspatial, temporal–atemporal, mental–physical; rather, ἀρχή can relate to any of these, but only by virtue of co-text and context. Moreover, the senses offered in BDAG exhibit several notable uses of metonymy. Senses (2) through (3) can be understood as metonymic specification, more concretely specifying the beginning as the entity who causes the beginning. The metonymically-shifted meaning of the causative entity can be further metonymized to refer to the role associated with the entity—the entity’s office or rule. According to the nuances reflected in senses (2) and (3), furthermore, this beginning is not marked as either the beginning or simply a beginning. In the spatial beginning of (4) we see a similar imprecision: the corner appears to the observer to be the place where something begins; it is a beginning, while not necessarily constituting the only or first. Thus, these observations can be restated as pragmatic mechanisms: (a) spatiality condition: ἀρχή does not semantically distinguish between spatial and non-spatial meaning. Whether the term refers to a spatial meaning is determined pragmatically. (b) temporality condition: context is also required to distinguish between temporal and atemporal meaning. The temporality of the sense is

16. The use of small caps is an orthographic convention that I have adopted from Columbia School work. My intention is to signal a definition or meaning—one that in some sense parallels the sparsity or indeterminacy of the term being defined—while avoiding the misunderstanding that a meaning like INITIATION is a gloss or translation equivalent for the term being defined. For a Columbia School example, see Huffman, Categories of Grammar, 31.

WISHART Monosemy in Biblical Studies

105

determined pragmatically. c) physicality condition: ἀρχή must be contextually modulated to distinguish between mental or physical senses. Mental or physical meaning is determined pragmatically. d) ἀρχή can be metonymically extended to refer to the causative entity that initiates a process, or further generalized to the causative entity’s role. These figurative extensions are determined pragmatically.

What I am doing is simply noting the variation in meaning that is evidenced among the various senses and describing those variations as pragmatic effects. The rationale behind this move is simple: if ἀρχή can potentially mean any of these various senses, it does not convey any one of them specifically; rather, the variation is best explained with reference to pragmatic conditioning—that is, the effect of co-text and context—or to figurative extensions of more concrete senses. Important questions at this point include where the pragmatic conditions come from and how many one is allowed to posit to explain the data. Ruhl offers an explanation: By now, some readers may have the uneasy feeling that pragmatic rules are beginning to proliferate without restraint. A good theorist is likely to wonder where it will end. But linguists who expect to find a limited number of pragmatic rules, or rules typically with only a few options, are mistaking the task; they are trying to make pragmatic rules into semantic rules. Listing pragmatic rules may be an infinite task: all knowledge of the world can be included. In dealing with language, we are used to expecting only a few possibilities; but pragmatic rules can be much more various, since our full knowledge is much more various. This difference between semantic and pragmatic (between what is relatively closed and what is relatively open) is a key part of this book’s argument. A pragmatic rule is justified if it accounts for data, and as fully as possible.17

The first step in Ruhl’s method is to outline the variation—even the polysemy—of a word as it is actually used in utterances: what kind of variation in meaning can be observed? Here I have assumed that BDAG offers a relatively thorough account of the variation ἀρχή exhibits, though it is likely that BDAG has

17. Ruhl, On Monosemy, 36.

106

Biblical and Ancient Greek Linguistics 6

underestimated the amount and scope of this variation.18 The boundaries between senses are subjectively assigned, but the semantics of the word, according to Ruhl’s minimalist notion of semantics, are more objective—though not totally objective, of course—by virtue of the fact that one must identify those pragmatic effects that are not common to “all” contexts of a word. One cannot be objective when identifying the contextually modulated message of an utterance, because there are too many variables to consider, and the reader brings numerous assumptions to the text. Only after this diversity in meaning is noted can a line be drawn between a word’s semantics and its pragmatic modulations. As Ruhl explains, “the boundary of semantic and pragmatic cannot be drawn generally in advance, but must be discovered, word by word, phrase by phrase, even sentence and discourse by sentence and discourse. No reasonable theory can evade or postpone this necessity.”19 As for ἀρχή, perhaps INITIATION does capture the semantics of the word; perhaps it does not. What it does attempt is the identification of a unifying factor that draws together all of the various senses of ἀρχή without overspecifying or resorting to etymology or “original meaning.” I leave it to the reader to evaluate the preliminary semantic definition offered in this brief example. To summarize: Ruhl ascertains a lexeme’s semantics by pragmatic factorization.20 Pragmatic factorization explains variation, rather than simply recording it. By positing pragmatic mechanisms, Ruhl attempts to account for how a stable semantic core of meaning is modulated by context.

18. As Ruhl (On Monosemy, 173) says, “I am arguing that we cannot discover the sense(s) of a word without fully gauging its applications. Dictionary definitions, especially of common words, highlight a few applications, which implicitly deny a unified sense, and thus underestimate the full range of applications.” 19. Ruhl, On Monosemy, 71. 20. Ruhl, On Monosemy, 71. I have introduced the description “pragmatic factorization” for clarity. Ruhl calls this method a monosemic bias.

WISHART Monosemy in Biblical Studies

107

4. Fewster’s Creation Language in Romans 8 (2013) Gregory Fewster’s work, exploring the ramifications of analyzing Paul’s use of “creation language,” especially the noun κτίσις in Romans 8, constitutes the first major monograph dedicated to exploring the impact of lexical monosemy in New Testament studies. Fewster’s work deserves careful attention, as lexical analysis is an area of biblical studies that has exhibited widespread and systemic problems.21 Here I will offer an overview of Fewster’s Systemic Functional Linguistics (SFL)based model of lexical analysis, which he calls “corpus-driven systemic-functional monosemy.”22 Second, I will take issue with two aspects of Fewster’s study: his dismissal of Ruhl’s model of monosemy as simply a “cognitive” approach, and his method of analyzing metaphorical or figurative word meanings, which is less adequate than the method proposed by Ruhl. 4.1 Summary of Fewster’s Corpus-Driven Systemic-Functional Monosemy Fewster’s goal is to “describe and defend . . . a robust lexical semantic methodology.”23 He accomplishes this task by undertaking a study that (1) covers a general theory of words, (2) analyzes a particular word by means of corpus data, and (3) introduces a nuanced account of a “lexicogrammatical metaphor theory.”24 Each of these steps will be summarized in order. 4.1.1 Systemic functional monosemy. Of primary importance to Fewster’s modelling of Paul’s language is the systemic nature of language.25 SFL, according to Fewster, views language as a “social semiotic,” realizing social realities by means of language functions that are constrained by a language-specific system of

21. Fewster, Creation Language, 13–17; Lappenga, Paul’s Language, 10–13. 22. Fewster, Creation Language, 73. 23. Fewster, Creation Language, 17. 24. Fewster, Creation Language, 83. 25. Another important feature is the functional nature of language, but I will deal with this below; here I will focus on the systemic aspect.

108

Biblical and Ancient Greek Linguistics 6

choices.26 In other words, the book of Romans is a social interaction between Paul and his readers, and this interaction is realized or constituted by the functional language Paul uses in the letter. The meaningfulness of the functions Paul performs— the meaning of what he says—depends on what he could have said instead. The meaning of one language choice depends on the other choices that were available to the speaker. Therefore, the meaning of Paul’s creation language in Romans 8, on Fewster’s SFL-based model, should be analyzed in comparison to the systemically-constrained choices available to Paul, and the mutual realization that takes place between the context of situation and the language involved.27 4.1.2 Corpus-driven analysis. When it comes to the data Fewster chooses to incorporate into his analysis, he uses a corpus-driven approach. He explains, “Sound conclusions require, first and foremost, a reasonable and balanced environment for observation.”28 This environment is provided by a corpus of texts. That is, instead of focusing solely on a small sample text, such as Romans, he compiles a corpus of representative documents.29 He goes on to clarify, “A corpus is used to provide statistically relevant data based on patterns in language instances that can be generalized to the language as a whole, from which specific texts might be compared.”30 Fewster chooses to draw generalizations from a large sample of language in order to better

26. Fewster, Creation Language, 39. 27. It should be pointed out that this view of language is not fundamentally at odds with other approaches to language, such as Relevance Theory (which is the framework of choice for Lappenga’s analysis; see below). Contra Clark (Relevance Theory, 359), the difference between these approaches is simply the locus of analysis. Where SFL considers the bidirectional impact of language on context and context on language, Relevance Theory chooses instead to focus on the (more or less) unidirectional impact of language upon the reader’s mental state. 28. Fewster, Creation Language, 53. 29. See “Appendix One: Outline of Specialized Corpus,” in Fewster, Creation Language, 175–76. 30. Fewster, Creation Language, 54.

WISHART Monosemy in Biblical Studies

109

understand a smaller sample, and this approach lends cogency to his analysis. 4.1.3 Lexicogrammatical metaphor. A third aspect of Fewster’s approach is a lexicogrammatical view of words as regards metaphor. That is, whereas traditional grammar represents lexis (word-choice) and grammar as entirely distinct areas of language meaning, Fewster chooses to view lexis as a grammatical choice—that is, lexicogrammar unites the two.31 Fewster should be commended for attempting to account for metaphorical word usage using a lexicogrammatical view of language. The idea of lexicogrammar is powerful, even though much work remains to be done. For example, we have lexicons of Hellenistic Greek, and we have grammar books—some of these are better than others—but we do not have lexicogrammars of Hellenistic Greek. Indeed, it is difficult to imagine what a lexicogrammar would actually look like. Nevertheless, such questions need to be explored, as they will open up new avenues of grammatical analysis that cannot be addressed according to the traditional lexis–grammar distinction. Fewster’s attempt to outline a theory of lexicogrammatical metaphor, I would argue, is on precisely the right track. Fewster also examines metaphorical usage from the perspective of monosemy. In his model, metaphor involves an atypical usage of a lexeme. However, this atypical usage not only includes the usual tropes like anger is heat, or argument is war. These kinds of atypical usages constitute lexical metaphors because they involve lexical meanings. Lexicogrammatical metaphor is an atypical usage of a lexeme in which grammatical meanings are involved.32 An example that comes up in Fewster’s study is κτίσις, which typically refers to a created thing, an entity, but which can also be used to refer to the process of creating. This usage, Fewster argues, demonstrates grammatical metaphor because the process κτίζω is lexicalized as a noun, κτίσις. This

31. Fewster, Creation Language, 44–45. Cf. pp. 82–93. 32. Halliday and Matthiessen, Halliday’s Introduction, 698–731; Thompson, Introducing Functional Grammar, 233–54.

110

Biblical and Ancient Greek Linguistics 6

kind of grammatical metaphor is called nominalization. Lexicogrammatical metaphor, then, attempts to take account of the fact that metaphors can be lexical or grammatical, or even both at the same time. While I affirm the effort to view lexis and grammar as a continuity, I will outline some issues with Fewster’s approach to lexicogrammatical metaphor in the next section. 4.2 Critical Assessment of Fewster’s Creation Language in Romans 8 There are two general issues I want to raise with respect to Creation Language. First, Fewster misrepresents key aspects of Ruhlian monosemy. While this misrepresentation is probably unintentional, a clearer account of Ruhl’s method would have helped Fewster avoid needless, and perhaps deleterious, duplication of some of Ruhl’s methodological steps. Second, Fewster’s proposed method of analyzing metaphorical or figurative word meaning is weaker than Ruhl’s proposed method. In some ways, Creation Language prematurely dismisses Ruhl because of a perceived association with cognitive linguistics, and this premature dismissal causes Fewster to miss out on some of Ruhl’s insights regarding monosemic analysis. 4.2.1 Fewster’s construal of Ruhl. In Fewster’s monograph, he adopts Ruhl’s theory of monosemy, but with hesitation. He is critical of Ruhl’s theory on the grounds that it is, he claims, based in “cognitive linguistics.”33 Fewster explains, Ruhl’s version of ‘extreme monosemy’ does have some shortcomings in terms of its utility in the present context. Monosemy is fundamentally a cognitive linguistic theory and in that regard is primarily concerned with a lexeme’s semantics as it relates to the conceptual ordering of the mind, and thus shares many of the shortcomings of cognitive polysemy. The theory, therefore, may be unable to adequately address meaning in terms of social interaction.34

33. Fewster, Creation Language, 36–37. 34. Fewster, Creation Language, 39.

WISHART Monosemy in Biblical Studies

111

Several things can be said in response to this claim. First, Ruhl actually self-identifies as a tranformational-generative linguist. To pigeonhole his theory as a “cognitive linguistic theory” is simply unclear. There are aspects of Ruhlian monosemy that can be called cognitive in some sense. For example, generalized definitions, Ruhl argues, best reflect the way we remember words, and his notion of semantic fields, he claims, reflects the structure of the human mind in terms of the way we draw abstractions. However, it is unclear why this “cognitive” element should disqualify Ruhl’s arguments without further critique, especially as Ruhl’s monosemic bias does not require analysis of cognitive aspects of meaning. Moreover, whereas cognitive linguistics is generally maximalist in its approach,35 Ruhlian monosemy is a minimalist endeavour.36 Grouping Ruhl together with cognitivists thus obscures the distinctiveness of his position. Second, the world is not divided into cognitive and functional linguistics. Rather, cognitive linguistics, a movement that emerged in the 1980s, takes the functional nature of language to be fundamental.37 So while Fewster is fully justified in adopting SFL as a framework, I would argue that Ruhlian monosemy is compatible with functionalism, including both social and cognitive frameworks. This is an important point to make, because if monosemy is going to have a broad impact on biblical studies, it will need to be, to some degree, framework independent. SFL asks distinctive questions about language instances, but it is not inherently better than cognitive approaches; it is just better at answering the distinctive questions that it asks. For example, SFL asks why particular language is used, and finds its answer in the particular context of use.38 Moreover,

35. Geeraerts, Theories of Lexical Semantics, 183. 36. Ruhl, On Monosemy, xi. 37. Geeraerts, Theories of Lexical Semantics, 267. 38. Fewster (Creation Language, 39) explains, “Context informs language use, while language use in turn forms and re-forms the social context. Systemic [functional] linguistics, therefore, posits a direct link between a given context and the use of language within that context.”

112

Biblical and Ancient Greek Linguistics 6

because language is an irreducibly social phenomenon, a particular context of language use is necessarily describable as a social context—and SFL proponents would argue that this is actually the most appropriate way to view the context. Yet I would agree with Fewster that it is possible to adopt a social, functional framework, to ask social and functional questions, and nevertheless to utilize key features of Ruhl’s approach to word meaning. This claim is evident when we consider the building blocks of Fewster’s theory and method. Fewster’s account is theoretically rigorous. He introduces, independently of Ruhl, a corpus-driven approach, a systemic view of language, and a lexicogrammatical view of language, and he explains the advantages of SFL over a “cognitive” approach by citing its systemic view of language choice and meaning, its functional approach to analysis, and the fact that lexis and grammar are treated as ends on a continuum. Yet each of these theoretical and methodological building blocks is already demonstrably present in Ruhl’s monograph. Ruhl does view language as a system, he adopts a mediating approach between formalist and functionalist approaches,39 he bases his observations on the examination of naturally-occurring utterances, and he explicitly treats lexis and grammar as parts of a broader continuum from abstract to concrete.40 Moreover, all of these points are in fact theoretically indispensable for Ruhl’s method of ascertaining what words mean. So while Fewster is fully justified in looking to other sources as he examines each of these issues, the premature nature of his dismissal of Ruhl’s monosemy is evidenced by his inclusion of additional pieces of theoretical infrastructure as though Ruhl’s theory does not already provide them. Third, Fewster’s method of analyzing lexemes, though developed in detail, lacks the coherence that makes Ruhl’s method so attractive in the first place. Ruhl’s hypothesis is, very simply, that contextual information can be filtered out of a semantic definition by positing pragmatic conditions that

39. Ruhl, On Monosemy, 200. 40. Ruhl, On Monosemy, 182–83.

WISHART Monosemy in Biblical Studies

113

account for the meaning variation that is evident. Fewster may be attempting this kind of pragmatic explanation, but it is unclear how his analysis takes place. He simply presents his findings along with a number of corpus examples and some discussion. Noting these points, I find it perplexing when Fewster states, “In light of its [Ruhl’s theory’s] (occasionally unhelpful) cognitive beginnings, I have shifted the notion of the monosemic bias into a systemic functional framework. In this light, abstracted semantic values are understood as meaning potential that is realized in the lexicogrammar of discourse.”41 It is unclear by what procedure Fewster actually arrives at his “abstracted semantic values.” 4.2.2 Fewster’s lexicogrammatical metaphor. In addition to these concerns about Fewster’s representation of Ruhl, I have concerns about the way that Fewster identifies a metaphorical usage as an atypical or “dynamic” usage of a lexeme that diverges from the “typical” or “congruent” usage.42 This way of understanding the distinction between literal and metaphorical is underdeveloped at best. Contrary to Fewster’s claims, it is unlikely that the literal meaning of a word can be reliably determined on the basis of frequency or “typicality.”43 A metaphorical extension of a word is

41. Fewster, Creation Language, 167. 42. Fewster, Creation Language, 88. 43. I am interpreting Fewster’s comments on pp. 88–89, where he says that corpus analysis reveals the most typical patterns of usage, to entail that typicality is a matter of numerical frequency, as this seems the most likely way to read his explanation. Fewster may have meant something different by statements like “typicality in a corpus” and “the corpus is a helpful tool for measuring congruence”—perhaps something more akin to Halliday’s view of congruent usage as “the most straightforward coding of the meanings selected” (Halliday and Matthiessen, Halliday’s Introduction, 731). However, neither Fewster’s nor Halliday and Matthiessen’s actual method of identifying what is congruent is obvious in their discussions. Fewster sometimes describes congruent and typical usages as the same thing, but he also notes that metaphorical uses may be typical. Regardless, my argument in this section is that Fewster would have been better served theoretically by engaging with the approach to metaphor analysis already provided by Ruhl.

114

Biblical and Ancient Greek Linguistics 6

not just an atypical usage, but a non-literal or figuratively extended use of a word. I suggest that Ruhl’s understanding of metaphor and his method of distinguishing literal from figurative usages are more intuitive and reliable than Fewster’s, because of Ruhl’s programmatic definition of abstraction as superordinacy. Viewing a vocabulary as a cline of increasing concretion, Ruhl is able to make sense of figurative usage in a way Fewster cannot. This is best illustrated by way of example. In On Monosemy, Ruhl analyzes the noun ice. What he finds is that there are at least two meanings of ice that are very difficult to relate to each other. On the one hand, ice can be defined as “water frozen solid.”44 But, notes Ruhl, this literal meaning does not encompass figurative, emotion-related uses of ice (i.e. implying fear or horror). So on the one hand, as “water frozen solid,” ice is a very concrete word—that is, it is easily referable to an extralinguistic reality, and on the other hand, as a sense of fear or horror, ice is a more abstract word. According to Ruhl, this is an example of metaphoric extension because the more abstract sense can be understood as a figurative extension of the concrete sense. In fact, if ice did not literally mean “water frozen solid,” it is unclear how ice as a sense of dread would make sense at all—a sense of dread is like the feeling of ice. The direction of figurative extension here is critical; frozen water is not called ice because it is like a sense of dread, but a sense of dread can be called ice because it (apparently) reminds us of the feeling of water frozen solid. Fewster claims that the more frequent meaning is the literal, and the atypical meaning is the figurative; by contrast, Ruhl demonstrates that the literal meaning is rather the more concrete of the two meanings, the one on which the figurative meaning relies for meaningfulness and comparison. “Water frozen solid” is the ground; fear or horror is the figure. Thus, Fewster’s own model of metaphor, while attempting to incorporate the lexicogrammatical perspective Ruhl outlines, nevertheless neglects the programmatic definition

44. Ruhl, On Monosemy, 192.

WISHART Monosemy in Biblical Studies

115

of abstraction that makes monosemic metaphor analysis coherent. Moving through the final chapters of Fewster’s monograph, one is left somewhat confused as to the necessity of the monosemous lexical analysis offered earlier on. This confusion arises in part from Fewster’s inclusion of an abstracted, monosemic gloss alongside of an account of its metaphorical extension. In the monograph, κτίσις is accorded the gloss “created thing.”45 However, Fewster also contends that κτίσις participates in a semantic domain/chain that interacts with other semantic domains/chains in the construction of a large-scale metaphor about the redemption of humanity—the created things.46 If κτίσις is a monosemous word that can refer literally to the created human body—as Fewster contends—its use in Romans 8 is literal, not metaphorical.47 On Fewster’s own analysis, then, I would argue that a metaphorical extension of κτίσις does not actually occur in Romans 8. Because Fewster combines lexical and grammatical metaphor into one theoretical conception, he attempts to explain a large scale conceptual metaphor (involving multiple lexemes) as no different than a grammatical metaphor, and ends up describing a non-figurative use of κτίσις as a metaphor when in fact the usage of κτίσις in question fits with the monosemous definition he himself assigns to it. If κτίσις is a monosemic word that is used to refer to created things, including the human body, then its use in referring to the human body is not a figurative extension of its ground usage. Κτίσις in Romans 8 does exhibit grammatical metaphor (if the

45. Fewster, Creation Language, 146. 46. Fewster, Creation Language, 123–24. 47. Fewster (Creation Language, 148) makes it clear that he has in mind both lexical and grammatical metaphor, saying, “Nominalization of this sort is lexicogrammatical metaphor at its clearest; there has been a shift in the grammar of the passage, yet there are lexical implications as well. The choice of the nominal κτίσεως in this clause [i.e. in Rom 8.19] is metaphorical in the classical sense—one thing is being referred to in terms of another.” Perhaps the relationship between κτίσις and σῶµα could better be construed as hyponymous rather than metaphorical. That is, a σῶµα is a kind of κτίσις.

116

Biblical and Ancient Greek Linguistics 6

noun κτίσις is used to describe a process or act, although this usage, too, could be included in a monosemic definition); however, κτίσις does not exhibit lexical metaphor. Fewster’s inclusion of grammatical metaphor is important in moving lexical analysis forward for biblical studies, but his theory of lexicogrammatical metaphor—at least in Creation Language— ends up potentially misconstruing both lexical and grammatical metaphor as they pertain to κτίσις. In summary, Fewster’s Creation Language is an intriguing attempt at incorporating lexical monosemy into biblical studies, tested in regards to κτίσις in Romans 8. Fewster successfully demonstrates both the appeal of the theory and its utility for biblical studies. His analysis is cast in the framework of SFL, yet readers from various perspectives will benefit from his study. While I have outlined several areas that require further development or clarity, Creation Language is an important trailblazing work that is sure to play a critical role in the conversation about lexical semantics in biblical studies. 5. Lappenga’s Paul’s Language of Ζῆλος (2015) The second monograph I wish to discuss here is Benjamin Lappenga’s Paul’s Language of Ζῆλος: Monosemy and the Rhetoric of Identity and Practice. Discussion of this study will be brief for two reasons: first, Lappenga does not attempt to outline an extensive linguistic method as does Fewster;48 second, Lappenga’s study employs an onomasiological rather than a semasiological approach.49 A semasiological approach begins with (typically) a linguistic sign (or signs) and proceeds to ascertain information about that sign’s meaning within a speech community. An onomasiological approach, by contrast, begins with a meaning that occurs and then explores its communication by means of various linguistic signs.50 In order to explore how

48. This is not meant to imply a deficiency in Lappenga’s study, only that Fewster dedicates much more space to outlining his linguistic methodology. 49. Lappenga, Paul’s Language, 66. 50. As Geeraerts (Theories of Lexical Semantics, 23) puts it,

WISHART Monosemy in Biblical Studies

117

Paul talks about Christian social identity and practice, Lappenga examines the concept of zeal by means of the ζηλ- word group. Because Lappenga’s monograph is not strictly dedicated to linguistic analysis, his account of monosemy is briefer. I will primarily assess Lappenga’s comments and critiques with respect to both Ruhl and Fewster’s approaches to monosemy, with a brief discussion of Lappenga’s alternative method. While I am largely critical of Lappenga’s engagement with Fewster and Ruhl, Lappenga’s approach to monosemy, which he casts within the framework of Relevance Theory, presents an intriguing approach to using monosemy for qualitative lexical analysis. 5.1 Lappenga’s Account of Ruhl Lappenga’s proposal for monosemy, while innovative and promising, is unfortunately situated within what I take to be unjustified and highly critical discourse. He identifies three problems with Ruhl’s On Monosemy: (1) the “need for more terminological clarity regarding semantics and pragmatics,” (2) “overconfidence about our ability to determine what is in fact a ‘convention,’” and (3) Ruhl’s claims about general, abstracted word meaning.51 Each of these points deserves a response. First, Ruhl, contrary to what Lappenga claims, does offer a nuanced account of the semantics–pragmatics interface. Although this particular issue is mentioned throughout Ruhl’s book, Section 7.2, “Semantic–Pragmatic,” is particularly notable.52 Second, I would question Lappenga’s assertion that Ruhl exhibits overconfidence. Lappenga, for his part, takes a highly individualized view of lexical meaning, given that everyone must have an ultimately unique mental lexicon.53 Ruhl’s idea of “semasiology starts from the expression and looks at its meanings, onomasiology starts from the meaning and looks at the different expressions.” 51. Lappenga, Paul’s Language, 29. 52. Ruhl’s alleged lack of clarity is attributed to a few stray comments at the beginning of On Monosemy, which Ruhl subsequently elaborates upon. 53. Lappenga, Paul’s Language, 44. However, Lappenga attempts to identify mental lexicon entries primarily in regard to the effect of written texts on the readers of that text.

118

Biblical and Ancient Greek Linguistics 6

generalized semantic meaning shared by users of a language, then, would seem to be in conflict with this position. However, one of the key themes of On Monosemy is the complex interaction between language as an idealized system of conventional meanings and language as a highly diverse set of events and instances. The following quotation hopefully answers Lappenga’s charge of “overconfidence about our ability to determine what is in fact a ‘convention.’” Ruhl explains that only some linguistic knowledge is conventional or shared among all users of a given language: some words are common knowledge, while others are not; and, even with a common-knowledge word, some of its contexts are common and some are not. That is, we have both unity among all speakers, and on a fairly specific level, and also diversity, by dialect, idiolect, register, or otherwise. I am claiming that some knowledge of a language is shared by all speakers of that language. More generally, some highly abstract knowledge will be common to speakers of all languages.54

Rather than exhibiting overconfidence, Ruhl simply attempts to make generalizations of the data. Third, Lappenga notes that “Ruhl’s proposal for monosemy moves in the right direction but falls short because of its reliance on a lowest common denominator approach to ‘general meaning.’”55 What Lappenga specifically objects to is the apparent uselessness of a meaning that is abstracted from all contextual usages. He claims, “Even if certain patterns can be detected among the uses of a word, how can such a vague ‘general meaning’ be useful, particularly to the study of ancient texts such as those found in the NT?”56 Similarly, he refers to this perspective on semantic meaning as “fruitless abstraction.”57 This point is the most important, as it presents a potentially serious objection to the use of Ruhlian monosemy in biblical studies. After all, if it is useless, then it is unlikely to appeal to

54. 55. 56. 57.

Ruhl, On Monosemy, 144–45. Lappenga, Paul’s Language, 54. Lappenga, Paul’s Language, 30. Lappenga, Paul’s Language, 31.

WISHART Monosemy in Biblical Studies

119

biblical scholars who are exploring the potential usefulness of linguistics within the field of biblical scholarship. Because this third critique of Ruhl dovetails so much with Lappenga’s critique of Fewster, however, I will proceed to this latter critique and mention Ruhl where relevant. 5.2 Lappenga’s Account of Fewster (and Ruhl) According to Lappenga, Relevance Theory is a framework for explaining how words come to have particular meaning in texts, not an explicit method for identifying those meanings.58 For a method of definition, therefore, he turns to Ruhl’s monosemy. Based on his understanding of Ruhl, Lappenga does not attempt to define zeal, and actually critiques Fewster for defining κτίσις. Ruhlian monosemy, according to Lappenga, is a theory of the radical underspecification of words; words cannot be adequately defined, hence they should not be defined.59 Unfortunately, this is a misunderstanding of Ruhl. Using a paraphrase to describe a lexeme is not a problem, because Ruhl’s theory does not explicitly reject paraphrases; he argues only that many, not all, words are incapable of being adequately paraphrased.60 In fact, Ruhl explicitly claims the opposite of what Lappenga asserts. As noted above, Ruhl defines ice as “water frozen solid.”61 Lappenga offers the argument that only procedural (i.e. grammatical) but not conceptual words can be paraphrased. Yet Ruhl paraphrases ice because it is a conceptual

58. Lappenga, Paul’s Language, 52. According to Relevance Theory, the most relevant mental items are activated for the lexeme’s mental grab-bag. 59. Lappenga, Paul’s Language, 34–35. 60. Lappenga (Paul’s Language, 55) notes that, in contrast to Ruhl’s work, Fretheim’s “more helpful” study of monosemy also uses paraphrases, but explains that this is not as problematic as Ruhl’s use because Fretheim offers definitions for procedural words, rather than conceptual words. Cf. Fretheim, “In Defence of Monosemy.” Fretheim analyzes four terms, and the last (Norwegian: med en gang/med det samme) is actually, according to Fretheim, a conceptual lexeme, with an encoded “conceptual meaning” shared, moreover, by the two “synonymous lexical items” (“In Defence of Monosemy,” 107). 61. Ruhl, On Monosemy, 200.

120

Biblical and Ancient Greek Linguistics 6

rather than grammatical word.62 Contrary to Lappenga’s construal of Ruhl, Ruhl actually says, “A literal meaning [i.e. definition] need not be solely semantic, if at all.”63 In other words, a definition should usually include some pragmatic information; the more concrete the word, the more pragmatic information is necessary in order to produce an adequate definition. Though I agree with Lappenga that the way Fewster arrives at his definition of κτίσις is unclear, it is a misrepresentation when Lappenga labels Ruhl’s notion of general meaning a “fruitless abstraction” that Fewster “does not move past.”64 Lappenga also mistakenly critiques Porter and Fewster for merely “paying lip service to Ruhl,”65 claiming, “My approach maintains ties with Ruhl; Fewster’s does not.”66 I find this to be a puzzling critique in three ways. First, both parties are manifestly advocating the use of a monosemic bias within biblical studies, so why not highlight the common ground? Second, neither party even refers to, much less interacts critically with or “maintains ties with” Ruhl’s theory as it is developed beyond the first few chapters of On Monosemy.67 Third, there is no reason beyond

62. Ruhl, On Monosemy, 177. 63. Ruhl, On Monosemy, 192–93. 64. Lappenga, Paul’s Language, 31–33. In particular, Lappenga (Paul’s Language, 35–36) claims, “Like Porter, Fewster proposes a monosemic meaning that, although paying lip service to Ruhl, fundamentally diverges from Ruhl’s radical underspecification of semantic content in favor of a readily definable monosemic value . . . Yet Fewster has neither endorsed Ruhlian monosemy nor offered an alternative semantics-based argument for the monosemic bias. Thus it is difficult to see how either Porter’s or Fewster’s ‘monosemic account’ has moved us any further than other proposals for ‘general meaning.’” Lappenga’s critique specifically concerns the use of a “readily definable monosemic value,” as opposed to a radically underspecified—or unspecifiable—semantic value. This critique would hold more water if Ruhl did not explicitly offer paraphrases/definitions for other words such as ice. 65. Lappenga, Paul’s Language, 35. 66. Lappenga, Paul’s Language, 36. 67. As far as I can tell, neither Fewster nor Lappenga actually makes reference to Ruhl beyond the halfway point of his monograph. Porter

WISHART Monosemy in Biblical Studies

121

utility or coherence to demand strict adherence to one theory. Lappenga should know this, as he proceeds to castigate Ruhl’s notion of general meaning while still attempting to utilize a monosemic bias. As long as coherence is achieved, theorists are free to be creative. Lappenga points out that “Fewster’s dismissal of cognitive approaches is far too sweeping, since language-in-use can in fact interact efficiently with a cognitive framework.”68 As noted above, I agree with Lappenga’s critique on this point. However, I find Lappenga’s dismissal of Fewster to be equally sweeping. “Unfortunately,” Lappenga disputes, “by dispensing with the way Ruhl understands monosemy actually to function, Fewster has forfeited any substantive theoretical basis for holding to a monosemic bias.”69 I cannot see why this critique does not cut both ways. It is unclear how Lappenga’s own particular reading of Ruhl justifies adopting Ruhl’s method, much as it is unclear how Fewster’s reading of Ruhl condemns Fewster’s own appropriation of monosemy. Luckily for both authors, this critique in fact cuts neither way. There is no fundamental affinity or dissonance between monosemy and either Relevance Theory or SFL, and whether one framework is closer to Ruhl’s than the other is entirely immaterial. Fewster’s definition of κτίσις is also the object of Lappenga’s criticism. The latter asks, If ‘something that has been brought into existence’ is present in all occurrences [of κτίσις], how does Fewster account for instances where the act of creation and not a thing is in view? . . . Fewster does not consider such questions before arriving at his preferred monosemic value for κτίσις.70

This critique is once again puzzling, since Fewster discusses grammatical metaphor at length and explicitly notes that κτίσις exhibits a common pattern of being a nominalization of the act

(Linguistic Analysis, 51–53) engages in a more thorough summary of Ruhl than either. 68. Lappenga, Paul’s Language, 37. 69. Lappenga, Paul’s Language, 36. 70. Lappenga, Paul’s Language, 36.

122

Biblical and Ancient Greek Linguistics 6

of creation. Thus, notes Fewster, “This nominalized construal functions, first and foremost, to concretize the reader’s experience of a creative act, that is to say, a creation always implies a creative act.”71 In summary, Lappenga’s critiques of both Fewster and Ruhl appear to be unwarranted. Thankfully, this does not invalidate Lappenga’s own novel approach to monosemy, which I will now briefly describe. 5.3 Lappegna’s Monosemy-Based, Relevance-Theory-Inspired Analysis Lappenga’s goal is to analyze the development of a concept within Paul’s letters. While ζῆλος is assigned multiple senses in the lexicon, Lappenga aims to identify what it is that semantically bridges Paul’s repeated uses of the ζηλ- stem. Paul’s repetition, he argues, is intended to shape the audience’s concept of ζῆλος, to modify “the cognitive environment of the hearer.”72 “My argument,” he explains, “is that multiple occurrences within a text or corpus do in fact shape one another, and that the appropriate way to account for this (especially for exegetical study) is to endorse a monosemic bias.”73 Lappenga relies on Sperber and Wilson’s Relevance Theory, with its attendant view of meaning: specifically, the meaning of an utterance is underdetermined by the words used to communicate it.74 An author chooses words that will convey the most meaning while requiring the least effort. Ultimately, this leads Lappenga to construe the ζηλ- word group as a “grab-bag” of ideas and concepts that are variously activated in context according to the audience’s assumptions about what is most relevant to Paul’s message. “To summarize my proposal,” he

71. Fewster, Creation Language, 112. 72. Lappenga, Paul’s Language, 41. 73. Lappenga, Paul’s Language, 54. According to Lappenga, Paul creates an ad hoc concept of zeal through repetition of the ζηλ- word group. It should be noted that Fewster (Creation Language, 116), too, includes a discussion of ad hoc domains. 74. Lappenga, Paul’s Language, 41.

WISHART Monosemy in Biblical Studies

123

says, “a definition of ζῆλος might contain some elements that can be communicated with words, but must be conceived of as a single grab-bag of mental items (memories, mental images, pieces of encyclopedic and/or anecdotal information).”75 Of course, words can be conceived of in many different ways. Linguistic analysis is not merely the presentation of objective facts about language; it necessarily involves a construal of language. Psychological approaches to meaning analysis are not inherently more suitable than social approaches any more than psychology represents the human experience “better” than sociology does. It is simply a matter of arriving at different answers to different questions. Lappenga’s Relevance-Theoryinspired approach to monosemy provides a useful tool for discovering what can be made of key concepts in the biblical text. Viewing multiple instances of a lexeme together helps us to avoid the mistake of assuming that we cannot infer meaning beyond what is explicitly lexicalized in a given utterance. “Most important,” explains Lappenga, “is that the writer will leave implicit everything the reader can be trusted to supply ‘with less effort than would be needed to process an explicit prompt.’”76 Thus the analysis of concepts, in light of Lappenga’s study, should not attempt merely to reproduce the explicatures of the biblical text; it should also seek to identify the most relevant implicatures, with the aim that we would read the entire text according to its own implicit and explicit aims, so far as those can be inferred. Though Lappenga diverges significantly from Ruhl, his analysis of the ζηλ- stem fits well within Ruhl’s notion of semantic fields. The stem, on this view, comprises a superordinate category with semantic substance that is subdivided more specifically by its hyponyms. Thus, Lappenga’s approach to lexical analysis provides a case study in analysis of lexical items using monosemy.

75. Lappenga, Paul’s Language, 56. 76. Lappenga, Paul’s Language, 48. Citing Green, “Relevance Theory and the Literary Text,” 215.

124

Biblical and Ancient Greek Linguistics 6 6. Conclusion

My primary critique of both Fewster and Lappenga is the limited sense in which they conceive of monosemy as a methodology. While Ruhl’s method involves positing pragmatic mechanisms to factor out contextual modulation in order to identify a word’s semantics, neither of the authors attempts an explicit analysis along these lines. When they mention a “monosemic bias” they seem to mean something like a willingness to assume that monosemy is more correct than polysemy. Thus Fewster ends up arriving at a definition based on a process that he does not make entirely explicit in his monograph (he outlines many observations, but does not explain exactly how he factors out contextual modulation), and Lappenga decides to avoid providing a definition altogether. Fewster has displayed the corpus-linguistic use of a monosemic bias within the framework of SFL, and Lappenga has offered an example of how relevance theory can be used in conjunction with monosemy to answer questions about concepts within the biblical text. Both of these authors have helped to set a trajectory for the kind of monosemy-based study that Porter has proposed,77 and it is to be hoped that others will see the potential as well. Future development should take cues from both these authors, and methodological questions should continue to be raised and answered. Though I have offered a number of critical remarks regarding both of the monographs mentioned in this review, my criticisms do not detract from the value of either the monographs themselves or the role that a monosemic bias plays in these monographs. It is always easier to critically evaluate than to do the hard work of theoretical trailblazing, and all the more so in biblical studies, where traditional methods are heavily entrenched. Both of these monographs have advanced the minimalist trajectory for analyzing the meaning of words, and for that the authors are to be strongly commended.

77. Porter, Linguistic Analysis, 59.

WISHART Monosemy in Biblical Studies

125

Bibliography Barr, James. The Semantics of Biblical Language. London: Oxford University Press, 1961. Clark, Billy. Relevance Theory. Cambridge Textbooks in Linguistics. New York: Cambridge University Press, 2013. Fewster, Gregory P. Creation Language in Romans 8: A Study in Monosemy. Linguistic Biblical Studies 8. Leiden: Brill, 2013. Fretheim, Thorstein. “In Defence of Monosemy.” In Pragmatics and the Flexibility of Word Meaning, edited by Enikö Németh T. and Károly Bibok, 79–115. Current Research in the Semantics/Pragmatics Interface. Amsterdam: Elsevier, 2001. Geeraerts, Dirk. Theories of Lexical Semantics. Oxford: Oxford University Press, 2010. Green, Keith. “Relevance Theory and the Literary Text: Some Problems and Perspectives.” Journal of Literary Semantics 22 (1993) 207–17. Halliday, M.A.K., and Christian M.I.M. Matthiessen. Halliday’s Introduction to Functional Grammar. 4th ed. London: Routledge, 2014. Huffman, Alan. The Categories of Grammar: French Lui and Le. Studies in Language Companion Series 30. Amsterdam: Benjamins, 1997. ———. “The Linguistics of William Diver and the Columbia School.” Word 52 (2001) 29–68. Lappenga, Benjamin J. Paul’s Language of Ζῆλος: Monosemy and the Rhetoric of Identity and Practice. BibInt 137. Leiden: Brill, 2015. Porter, Stanley E. Linguistic Analysis of the Greek New Testament: Studies in Tools, Methods, and Practice. Grand Rapids: Baker Academic, 2015. ———. Verbal Aspect in the Greek of the New Testament, with Reference to Tense and Mood. Studies in Biblical Greek 1. New York: Peter Lang, 1989. Reid, Wallis. “Quantitative Analysis in Columbia School Theory.” In Meaning as Explanation, edited by Ellen Contini-Morava and Barbara S. Goldberg. Trends in Linguistics: Studies and Monographs 84. Berlin: Mouton De Gruyter, 1995. Ruhl, Charles. “Data, Comprehensiveness, Monosemy.” In Signal, Meaning, and Message: Perspectives on Sign-Based Linguistics, edited by Wallis Reid et al., 171–89. Studies in Functional and Structural Linguistics 48. Amsterdam: Benjamins, 2002.

126

Biblical and Ancient Greek Linguistics 6

———. On Monosemy: A Study in Linguistic Semantics. SUNY Series in Linguistics. New York: SUNY Press, 1989. Thompson, Geoff. Introducing Functional Grammar. 3rd ed. New York: Routledge, 2014.

[BAGL 6 (2017) 127–36]

THE PERFECT TENSE-FORM, THE SON OF MAN, AND JOHN 3:13, ONCE MORE Stanley E. Porter and Andrew W. Pitts McMaster Divinity College, Hamilton, ON, Canada Arizona Christian University, Phoenix, AZ, USA Abstract: This article responds to the article by Madison Pierce and Benjamin Reynolds on the use of the perfect tense-form in John 3:13. While we commend their treatment of verbal aspect in their analysis, we offer several points of correction on several issues, including the semantics of the perfect tense-form, the use of the aorist participle, and the conditional clause. (Response) Keywords: Greek tense-form, perfect tense-form, stative aspect, aorist articular participle, conditional clause.

1. Introduction In a 2014 issue of NTS, Madison Pierce and Benjamin Reynolds put forward a useful study of verbal aspect theory and its potential implications (specifically related to issues of temporality) for ascent language in John 3:13. Pierce and Reynolds argue that the notoriously problematic passage regarding the ascent and descent of the Son of Man in John 3:13 can be solved through recent developments in Greek grammar, in particular verbal aspect theory and the timing of the ascent and descent.1 They accentuate the enigma of the verse, so often interpreted to indicate the previous ascent of the Son of Man.

1. Pierce and Reynolds, “Perfect Tense-Form.” See their article for representative references to scholars who hold to many of the traditional opinions to which they are responding.

128

Biblical and Ancient Greek Linguistics 6

Pierce and Reynolds rightly identify traditional understandings of the perfect tense-form (ἀναβέβηκεν) (as indicating a past action with continuing results) as a driving force behind these readings.2 This use of the perfect tense-form, combined with the subsequent aorist articular participle (καταβάς) (traditionally thought to indicate past action), results in the following translation: “No one has previously ascended into heaven, except the one who has descended from heaven, the Son of Man, has ascended into heaven,” with the last phrase implied by the verbless “exception” clause.3 As Pierce and Reynolds point out, numerous scholars endorse this interpretation of the grammar of John 3:13, especially of the use of the perfect tense-form. Some even pronounce it as the “literal” or “natural” interpretation of the verse.4 Consequently, Pierce and Reynolds list five different proposals in the history of scholarship that attempt to remedy the difficulty of this interpretation and its use in the Johannine context. We applaud this effort to apply linguistic criticism to questions of exegesis. We also appreciate their mentioning, even if tentatively, the stative semantic content for the perfect tense-

2. See, for example, Blass and Debrunner, Greek Grammar, 175–76. They have been preceded and followed by many others (see Pierce and Reynolds, “Perfect Tense-Form,” 153 n. 20), including some who argue from a linguistic perspective. These include, among others, Fanning, Verbal Aspect, 119–20 and Andrason and Locatell, “Perfect Wave.” This is not the place to criticize these views, except to say that they seem to assume the prior definitions and then seek to impose them upon Greek usage; utilize lexical aspect as the basis of their analysis; and impose unnecessary (and one might add, unsatisfactory) typological understandings upon the Greek verbal edifice. 3. This translation is a modified version of the one presented by Pierce and Reynolds, “Perfect Tense-Form,” 150. 4. Pierce and Reynolds (“Perfect Tense-Form,” 150–51) provide several examples of those who recognize the traditional interpretation of the grammar of the verse. Some scholars apparently do not recognize the linguistic models at play that contribute to their understandings, so they refer to their understandings in such terms as “taken literally” (Michaels, Gospel of John, 195) or “natural reading” (Ashton, “Johannine Son of Man,” esp. 513–14). It almost defies belief that such naïve statements can be made by scholars in the twenty-first century.

PORTER AND PITTS Perfect Tense-Form

129

form. However, they fail to understand the interrelationship of semantics and pragmatics in grammaticalizing temporality via contextual features. While the point of recent work on verbal aspect may seem simple at first (time is encoded through context not verbal morphology), calibrating verbal aspect with temporal deictic (contextual) features involves precision in grammatical interpretation of contextual features at several levels of language: the word level, the clause level, and the clause complex level (and perhaps higher levels as well). Although Pierce and Reynolds apparently grasp the basic point in verbal aspect theory that time is not located in the verb, they do not exhibit competence in their assessment of broader temporal deictic features and their interaction with the (aspectual) semantics of the verb. This creates problems throughout their analysis. 2. Word Level Analysis We do not intend here to be overly negative. Pierce and Reynolds are correct in their understanding at a number of places. We agree that the traditional interpretation of the perfect tense-form is to be questioned and that Greek tense-forms are not temporal in sense, including both the perfect- and aorist-tense forms. However, we are conflicted. While we welcome and celebrate the attempt of Pierce and Reynolds to utilize verbal aspect theory to solve interpretive problems, their analysis is contaminated by a number of inaccurate assumptions related to Greek grammar and linguistics that penetrate other elements of their interpretation. The first problem that this study faces involves clear identification of the semantics of the perfect tense-form and, with it, issues revolving around the relation of clausal semantics to discourse semantics.5 Verbs encode at the word level a

5. This is similar to but different from the distinction between semantics and pragmatics. We prefer to conceptualize the distinction as one between clausal and discourse semantics, in which there are various ranks within clausal semantics that help to modulate the meaning. This is not the place to go into further detail about this distinction. However, it makes for a more satisfactory

130

Biblical and Ancient Greek Linguistics 6

semantic aspectual value—i.e. verbs grammaticalize aspect as a semantic value that is realized at all levels of analysis. If the interpreter skews analysis at this semantic level then higher levels of analysis will also be undermined since the value of a form results from the modulation of its semantic value by higherlevel contextual features. Therefore, failure to properly identify semantic features at the word level will introduce problems with higher levels of analysis, including (but not limited to) attempts to identify the temporal features of a discourse and their relation to verbal aspects. The theory developed within Pierce and Reynolds’s article seeks to construct a semi-eclectic approach to the perfect tense, but in combining several theories, it creates a contradictory analysis at some points. They state that, “Though the aspectual value for the perfect is currently debated, the consensus among proponents of verbal aspect is that time value is not the primary feature of the verbal form.”6 But the claim that the perfect does not encode time as a primary feature of the form provides a strictly negative assessment. What does the perfect encode? Since the article seems to urge—more than anything else—a fundamental point about translation (i.e. we should translate οὐδεὶς ἀναβέβηκεν εἰς τὸν οὐρανὸν as “no one ascends to heaven”), one would think that the semantics of the perfect tense-form would factor in as an important component of the argument before translation is attempted. Second, whereas Pierce and Reynolds are ready to accept the non-temporal semantics of Greek tense-forms (i.e. time is a contextual not a formal feature), they do not clearly commit themselves to the aspectual meaning of the perfect tense-form.7

explanation than the one between semantics and pragmatics, in which there is not a formalized meaning relationship between the two levels and where theories of polysemy are usually assumed. On some of these issues, see Porter, “Systemic Functional Linguistics,” esp. 32–47. 6. Pierce and Reynolds, “Perfect Tense-Form,” 153. 7. They note two recent views. The first, that the perfect tense-form grammaticalizes stative aspect, is held by Porter (Verbal Aspect, 245–90; although they only cite Porter, Reed and O’Donnell, Fundamentals, 315, a beginning grammar!), McKay (“Use of the Ancient Greek Perfect” and “On the

PORTER AND PITTS Perfect Tense-Form

131

As a result, rather than adopting one of the five previous views for understanding John 3:13, they simply posit without substantive support the so-called “timeless perfect,” to indicate “a unique quality of the Son of Man.”8 So they miss entirely the significance of the aspect (i.e. the semantics) of the perfect tenseform. Porter’s proposal, upon which they apparently (at least partially) depend, argues that the perfect tense-form grammaticalizes stative aspect.9 That is, the verbal process is conceptualized by the author as a state of affairs of the subject, without specific temporal placement. (And even if the perfect tense-form is understood as indicating imperfective aspect with heightened proximity, as has been suggested, this still suggests problems for their understanding.) The clause is thus to be understood as indicating something like: “no one [is, was, has been, etc.] in an ascended-into-heaven state,” with temporal reference left open by the clause itself (though various translational equivalents may be possible). The result is that Pierce and Reynolds miss the semantic significance of the perfect tense-form by failing to determine its verbal aspect.

Perfect,” 289–329), and Louw (“Die Semantiese Waarde,” 23–32); and the other, that the perfect grammaticalizes imperfective aspect with heightened proximity, by Campbell (Verbal Aspect, 210–11). Porter’s critique of Campbell’s position, along with some of those above (note 2), is found in Porter, Linguistic Analysis, 195–215. Note the recent proposal of Crellin, “Semantics of the Perfect,” whose conclusions, as he admits at several places, once one sorts through the linguistic typologizing and unnecessary appeal to lexical aspect, are surprisingly similar to those of Porter and McKay (he does not apparently know the work of Louw, who is found nowhere in the entire volume, if the index is to be believed). 8. Pierce and Reynolds, “Perfect Tense-Form,” 154. 9. Porter argues for the stative aspect of the perfect tense-form on the basis of the tri-aspectual structure of the Greek verbal system and not by imposing either the traditional understanding or a binary aspectual system or various forms of linguistic typology on the Greek verbal system.

132

Biblical and Ancient Greek Linguistics 6 3. Clause Level Analysis

Pierce and Reynolds’s failure to properly assess semantic aspectual values in John 3:13 results in further mistakes within their clause level analysis of the syntactical (temporal) features at work in the passage. We agree that clause level considerations may play an important role in determining the contribution of temporal elements to the discourse, but not in the way that Pierce and Reynolds propose. In order to invoke a syntactic pattern where adverbial aorist participles preceding their main verb tend to indicate previous action, they postulate a hypothesized main verb that follows the substantival participle in a particular syntactical order: “the one who descended from heaven, the Son of Man [has ascended to heaven]”!10 We notice several problems here. First, their reconstruction emerges entirely from considerations based upon English translation rather than the semantics of Greek. Such translational considerations apparently lead them to posit an elided verb that follows the participle. This creates a second problem. Even if we grant an elision here, on what basis do Pierce and Reynolds make their claim about word order? If the finite verb has been elided, how do Pierce and Reynolds know that the elision occurred after rather than before the participle? To state the problem more directly: the posited word order is based upon English grammar rather than Greek word order, making their reconstruction impossible (besides the fact that they have misunderstood the participle—see below). Furthermore, the elision very well may be the verb “ascend,” as Pierce and Reynolds propose, but a verb of existence (“be”) seems also plausible, potentially allowing for the following translation: “except for the existence of the coming-down-fromheaven one, the Son of Man.” So postulating an elided verb here really does nothing to help their case, only introducing further ambiguity. Finally, whereas Pierce and Reynolds are correct that the aorist participle is aspectual and not temporal, with temporality indicated by context (mostly, but not entirely, at the discourse

10. Pierce and Reynolds, “Perfect Tense-Form,” 150.

PORTER AND PITTS Perfect Tense-Form

133

semantic level), they believe that the substantival use of the aorist participle functions similarly to the adverbial use of the participle, in examining temporal patterns of syntax. Even if we grant their claim regrading word order, this still does not address the fact that they have applied principles governing adverbial participial structures to substantival participles. However, these syntactic descriptions are not transferable.11 In other words, Pierce and Reynolds incorrectly apply a syntactical generalization regarding the adverbial participle to a non-adverbial participle. This leaves their explanation of the verse with little to commend it. 4. Clause Complex Level Analysis The conditional clause complex similarly requires reinterpretation. So-called conditional structures are constructed around a complex of clauses involving a protasis clause and an apodosis clause (thus it is a set of clauses that creates a supposition-consequence relation). Pierce and Reynolds are correct that the εἰ µή phrase indicates an exception (or “if not”), not simply disjunction (“but”), and that factors at the clause complex level may contribute to the indication of time in a discourse. However, they go too far in positing the existence of the finite verb and more particularly its placement in the protasis clause. This is unnecessary in any case, as the participle (in this case the aorist) does not indicate time but verbal aspect. The verbal aspect of the aorist tense-form is perfective aspect. There is no syntactic factor that indicates the temporal placement of the process encoded. Thus, this phrase cannot be assumed to indicate a past act of descent, but can be interpreted semantically (though again, a range of translation equivalents may be possible) as indicating something like: “except (or if not) the coming-downfrom-heaven one, referring to the Son of Man.” The article is not

11. Pierce and Reynolds apparently get confused at least in part because they use Porter, Reed, and O’Donnell, Fundamentals, 110, even though the paragraph they are referring to clearly refers to the “adverbial participle.” See instead Porter, Verbal Aspect, 380–81; Porter, Idioms, 188.

134

Biblical and Ancient Greek Linguistics 6

a definitizer, but indicates the extent of the wordgroup and nature of the grammatical relation in which the participle is the headterm of the nominal construction, elaborated by the appositional phrase ὁ υἱὸς τοῦ ἀνθρώπου. Pierce and Reynolds conveniently use their newly grammatically informed interpretation of the verse to substantiate an orthodox solution to the crux interpretum in John 3:13: Jesus’ descent, indicated by the aorist participle, precedes the timeless description of him as the one who ascends, presumably an ascension that transpires after his descent and preserves the traditional understanding of the Gospel account. So, for Pierce and Reynolds, Jesus descends at the beginning and ascends at the end of his time on earth. As we have seen above, the grammar of the passage by no means requires this interpretation (especially when so many grammatical misunderstandings are involved). In order to understand this verse, we must move beyond considerations of verbal aspects at the word level. We must also assess the role of the conditional clausal structure in which they are utilized.12 Conditional clauses formulate logical and other relations between supposition and consequence statements. The specific logical relations between the protasis and the apodosis vary depending upon context, but in each case the protasis provides the logical supposition from which consequences are drawn. The conditional clause of John 3:13 has an inverted ordering that places the apodosis before the protasis and hence thematizes the consequence statement over the supposed one. 5. Conclusions In light of all three levels of contextual analysis considered above, we propose that the semantics of the entire conditional structure of John 3:13 can be rendered as indicating something like: “except for there being one who is a coming-down-fromheaven one, the Son of Man, there is no one in an ascended-intoheaven state”—though, again, a range of translational renderings 12. See Porter, Verbal Aspect, 291–320; Porter, Idioms, 254–67.

PORTER AND PITTS Perfect Tense-Form

135

of the semantics may be possible. In other words, the descending one is the logically posited supposition for the consequent implication regarding ascension. Except for such a one as the Son of Man (note that the conditional conjunction can be rendered as “if there is not one who . . . ”), there is no one who is in an ascended state. Our conclusion may seem close to (even if more nuanced and semantically viable than) the conclusions generated by Pierce and Reynolds’s study. And indeed, it is. However, our point is one about methodology, not results. If flawed methodology is employed then there is no way to assess the probability of the resulting interpretation(s). So while Pierce and Reynolds may end up at the right place (or, at least, close to it), they do so for reasons unconnected to proper assessment of the grammar of the passage which—they claim—is the major contribution of their article. Their conclusion is certainly not novel, so they showcase their method as the contributing component of their study. But this is precisely where their analysis is so flawed. Bibliography Andrason, Alexander, and Christian Locatell. “The Perfect Wave: A Cognitive Approach to the Greek Verbal System.” BAGL 5 (2016) 7–121. Ashton, John. “The Johannine Son of Man: A New Proposal.” NTS 57 (2011) 508–29. Blass, Friedrich, and Albert Debrunner. A Greek Grammar of the New Testament and Other Early Christian Literature. Translated by Robert W. Funk. Chicago: University of Chicago Press, 1961. Campbell, Constantine R. Verbal Aspect, the Indicative Mood, and Narrative: Soundings in the Greek of the New Testament. Studies in Biblical Greek 13. New York: Peter Lang, 2007. Crellin, Robert. “The Semantics of the Perfect in the Greek of the New Testament.” In The Greek Verb Revisited: A Fresh Approach for Biblical Exegesis, edited by Steven E. Runge and Christopher J. Fresch, 430–57. Bellingham, WA: Lexham, 2016. Fanning, Buist. Verbal Aspect in New Testament Greek. Oxford: Clarendon, 1990.

136

Biblical and Ancient Greek Linguistics 6

Louw, J.P. “Die Semantiese Waarde von die Perfectum in Hellenistiese Grieks.” Acta Classica 10 (1967) 23–32. McKay, K.L. “On the Perfect and Other Aspects in New Testament Greek.” NovT 23 (1981) 289–329. ———. “The Use of the Ancient Greek Perfect down to the End of the Second Century A.D.” Bulletin of the Institute of Classical Studies 12 (1965) 1– 21. Michaels, J. Ramsey. The Gospel of John. NICNT. Grand Rapids: Eerdmans, 2010. Pierce, Madison N., and Benjamin E. Reynolds. “The Perfect Tense-Form and the Son of Man in John 3:13: Developments in Greek Grammar as a Viable Solution to the Timing of the Ascent and Descent.” NTS 60 (2014) 149–55. Porter, Stanley E. Idioms of the Greek New Testament. 2nd ed. Biblical Languages: Greek 2. Sheffield: Sheffield Academic, 1994. ———. Linguistic Analysis of the Greek New Testament: Studies in Tools, Methods, and Practice. Grand Rapids: Baker, 2015. ———. “Systemic Functional Linguistics and the Greek Language: The Need for Further Modeling.” In Modeling Biblical Language: Selected Papers from the McMaster Divinity College Linguistics Circle, edited by Stanley E. Porter, Gregory P. Fewster, and Christopher D. Land, 9–47. Linguistic Biblical Studies 13. Leiden: Brill, 2016. ———. Verbal Aspect in the Greek of the New Testament, with Reference to Tense and Mood. Studies in Biblical Greek 1. New York: Peter Lang, 1989. Porter, Stanley E., Jeffrey T. Reed, and Matthew Brook O’Donnell. Fundamentals of New Testament Greek. Grand Rapids: Eerdmans, 2010.

ANCIENT SOURCES INDEX Old Testament Genesis 2:24 9:4 Exodus 3:14

46 45

21:15

57

1 Kings 8:24 12:24

45 57

2 Kings 14:5 14:15 19:9

56 57 57 61 78 73

53

Deuteronomy 1:29 8 Judges 5:20 11:20 20:14 20:18

57 57 57 57

Jeremiah 8:6 10:3–5 10:3–4

1 Samuel 17:32–33 28:1

57 57

Daniel 10:20 11:11

57 57

2 Samuel 10:17 11:17

57 57

Amos 2:11

45

Apocrypha Tobit 5:4 8:2–3

12:11–20 83, 87 87

88

138

Biblical and Ancient Greek Linguistics 6 New Testament

Matthew 1:16 2:9 5:10 5:24 5:32 5:34–36 5:43 10:32 10:41 11:11 11:22 11:23 11:24 13:48 15:22 15:23 15:27 19:5 19:9 19:19 21:2 22:39 23:16 23:18 23:20–22 23:25 23:35 24:18 24:38 26:24 26:58 26:64 27:43 28:1

30 30 30 35 37 46 37 46 45 25 31 31 31 36 45, 46 31 46 46 37 37 30 37 46 46 46 36 36 37 30 30 36 31 36 31

Mark 1:9 4:41 5:1 6:6 6:36 7:28 8:15 11:2 12:31 12:32 12:33 12:38 13:16 13:30 13:35 14:54 15:16

24, 45 24 37 30 30 46 46 30 37 31 37 46 37 31 31 36 36

Luke 1:44 1:58 4:18 9:12 9:62 10:27 11:1 11:39 11:40 11:51 12:8 13:33 17:31 19:4 19:30 20:46

45 46 30 30 37 37 31 36 36 36 46 36 37 35 30 46

Ancient Sources Index 24:49 John 1:16 3:13 3:25 4:5 4:31 6:1 15:11 16:19 17:11 18:6 20:14 20:30 21 Acts 2:39 4:29 4:31 5:38 12 12:11–15 12:15 13:42 16:36 17:30 19:9 20:32 21:5 24:25 26:11 27:22 Romans 1:5 1:18 4:17 5:11 5:12

139

23

7:22 8

29 127, 128, 131, 132, 134 57 37 36 37 24 57 30 37 37 23 77

8:19 8:22 13:9 16:25–26

36 107, 108, 115, 116 115 20, 30 37 31

1 Corinthians 6:6 6:18 12:20 14:5 15:2

57 30 36 30 30

2 Corinthians 2:17 4:1 7:5 10:16 11:23 11:28 12:2 12:19

30 29 36 37 28 37 30 30

Galatians 3:19 4:19 5:14

37 31 37

Ephesians 1:5–14 3:16 3:20

38 36 26, 31

Philippians 3:13

35

Colossians 1:23 2:18 3:1 4:5

20 46 26 36

45 36 29 36 81 82 67, 69, 82, 91–93 36 36 36 24 36 36 36 36 36 37 32 30 29 29

140

Biblical and Ancient Greek Linguistics 6

1 Thessalonians 2:11 31 3:10 26, 31 5:13 26 2 Thessalonians 1:9 23 1 Timothy 3:2 3:16

74, 78, 81 25

Hebrews 1:1 3:11 4:3

25 46 46

James 2:8

37

1 Peter 2:13

20

Jude 7

36

Revelation 1:4–6 1:4–5 1:4

55 55 52–54

2:16 2:21 2:22 3:12 4–5 4 4:2–3 4:6 5:11 6:8 7:11 9:20 9:21 10:6 11:7 11:11 12:6 12:7 12:17 13:3 13:4 13:7 14:20 15:2 16:11 17:14 19:19 21:2 21:10

Dead Sea Scrolls 1QS 8:14

56

46, 56–58 61 61 55 59 55 55 31 31 30 31 61 61 46 46 58 55 46, 57, 58 46, 57, 58 60 46, 57, 58 46, 57, 58 36 59, 60 61 46, 57, 58 46, 57, 58 55 55

Ancient Sources Index

141

New Testament Apocrypha Acts Paul 44.4

59

Acts Thom. 16.16

59

Greco-Roman Writings Aesop Fab. 21.2.5

59

Curtius Rufus Alexandri Magni Historia 46.80 57 286.12 57 Dio Cassius Hist. Rom. 59.25.8.2–3

59

John Malalas Chron. 36.6 59 93.9 59

140.21 184.19 211.17 222.13 224.8 264.7

59 59 59 59 59 59

Pausanias Descr. 4.27.8.4

57

Scholia in Aeschylum Th.635.6

57

Thucydides Hist. 1.59.2.4–5

57

Early Christian Writings Cedrenus Compendium Historiarum 1.765.5 58

Shepherd of Hermas 72.6.4 74.5.2 100.2.3

61 61 61

142

Biblical and Ancient Greek Linguistics 6 Papyri and Inscriptions

BGU III 910.2.11

54

P.Cair. 15.7

54

P.Oxy. III 527.r.2–4

54

P.Tebt.I 41:8–11

54

Second Philae Decree 197.9–14 54 197.17–19 54

MODERN AUTHORS INDEX Aarts, F. 19, 25, 40 Aarts, J. 19, 25, 40 Adams, F.A. 10–12, 15, 43 Allo, E.B. 49, 54, 60, 62 Andrason, A. 128, 135 Andrews, E. 42 Ashton, J. 128, 135 Aune, D.E. 49, 59, 60, 62 Baker, J.O. 85, 95, 96 Bakker, E.J. 16 Barr, D.L. 47, 62 Barr, J. 100, 125 Beale, G.K. 7, 16, 60, 62 Beckwith, I.T. 60, 62 Beekman, J. 71, 95 Bémová, A. 22, 42 Bennett, D.C. 34, 40 Benson, J.D. 26, 40 Bibok, K. 125 Black, D.A. 45, 62 Blass, F. 60, 62, 128, 135 Bloom, P. 34, 40 Bortone, P. 6, 9, 10, 14, 16, 34, 41, 59, 63 Bousset, W. 49, 63 Brendsel, D.J. 16 Bromiley, G.W. 63 Brown, H.D. 51, 63 Brown, J.K. 70, 95 Bruce, F.F. 84, 95 Burton, E. 60, 63 Büschel, F. 53, 63

Buttmann, A. 60, 63 Callow, J. 71, 95 Campbell, C.R. 131, 135 Carrell, P.R. 91, 95 Carson, D.A. 67, 95 Charles, R.H. 49, 53, 60, 61, 63 Christidis, A.-F. 64 Clark, B. 108, 125 Contini-Morava, E. 125 Cook, V. 50, 51, 63 Cowden, T.L. 49, 63 Crellin, R. 131, 135 Davidson, M. 87, 88, 92, 95 Davidson, S. 49, 63 Debrunner, A. 128, 135 Deissmann, G.A. 6 deSilva, D.A. 87, 91, 96 Dougherty, E. 49, 60, 61, 63 Draper, S. 85, 96 Draper, W.F. 63 Ebrard. J.H.A. 59, 60, 63 Eco, U. 41 Ellis, R. 47, 48, 50, 63 Evans, C.A. 96, 97 Ewald, G.H.A. 49, 53, 63 Eynikel, E. 64 Fanning, B. 128, 135 Fantin, J.D. 68, 96 Fawcett, R.P. 21, 25, 41

144

Biblical and Ancient Greek Linguistics 6

Ferguson, E. 84, 96 Fewster, G.P. 99–101, 107–17, 119–22, 124, 125, 136 Fresch, C.J. 135 Fretheim, T. 119, 125 Friberg, B. 56, 63 Friberg, T. 56, 63 Friedrich, G. 63 Fromkin, V. 51, 63 Funk, R.W. 135 Gass, S.M. 47, 50, 64 Geeraerts, D. 100, 101, 111, 116, 125 Gehman, H.S. 45, 64 George, C.H. 8, 9, 16 Gerth, B. 33, 41 Gill, J. 83, 96 Goldberg B.S. 125 Grabbe, L.L. 89, 96 Greaves, W.S. 26, 40 Green, K. 123, 125 Guillemard, W.H. 54, 56, 64 Hajcová, E. 22, 42 Halliday, M.A.K. 18–20, 23– 26, 37–41, 109, 113, 125 Harris, M.J. 10, 12, 14–16, 34, 41, 59, 64 Harrison, G. 45, 64 Hauspie, K. 64 Heiser, M.S. 88, 96 Herskovits, A. 34, 41 Hill, H.S. 67–71, 74–84, 89, 94–97 Hoffmann, S. 41 Horrocks, G.C. 9, 16 Howard, W.F. 46, 65 Huddleston, R. 18, 22, 28, 41 Hudson, R.A. 22, 41 Huffman, A. 102, 104, 125 Hyams, N. 63 Jakobson, R. 15 Janse, M. 49, 64

Jespersen, O. 16 Johnson, M. 34, 41 Jonge, C.C. de 6, 16 Junghanns, U. 16 Keane, W. 75, 96 Kerr, G.J. 71, 96 Khalfa, J. 97 Kittel, G. 63 Knight, G.W. 74, 96 Kühner, R. 33, 41 Kysar, R. 45, 66 Lakoff, G. 33, 41 Land, C.D. 17, 136 Lappenga, B.J. 99–101, 103, 107, 108, 116–25 Larson, M.L. 71, 96 Lee, S.-I. 45, 64 Léonas, A. 49, 64 Lightbown, P.M. 50, 51, 64, 65 Lindstromberg, S. 34, 41 Locatell, C. 128, 135 Louw, J.P. 32, 41, 56, 131, 136 Lücke, F. 49, 58, 64 Luraghi, S. 13, 14, 16, 34, 42, 60, 64 Lust, J. 56, 64 Macumber, H. 86, 96 Mare, W.H. 68, 96 Matthiessen, C.M.I.M. 109, 113, 125 McKay, K.L. 130, 131, 136 Metzger, B.M. 34, 42 Michaels, J.R. 128, 136 Miller, N.F. 63 Mitsakis, K. 64 Moffatt, J. 60, 64 Morley, G.D. 18–21, 26, 27, 32, 42 Moț , L.F. 47, 56, 64 Moule, C.F.D. 10, 16, 46, 58, 64 Moulton, J.H. 46, 54, 65, 84, 87, 96

Modern Authors Index Muir, J. 19, 42 Mussies, G. 56, 65 Najman, H. 88, 89, 96 Né meth T., E. 125 Newman, B.M. 56, 65 Newport, K.G.C. 49, 56, 60, 61, 65 Newton, R. 85, 97 Nida, E.A. 32, 41, 56, 70, 71, 97 O’Donnell, M.B. 130, 133, 136 Omaggio, A.C. 51, 65 Ombres, R. 85, 97 Ophuijsen, J.M. van 6, 16 Osborne, G.R. 67, 68, 77, 97 Parker, F. 51, 65 Peters, R.D. 35, 42 Pierce, M.N. 127–36 Polhill, J.B. 83, 97 Porter, S.E. 12, 13, 15, 16, 22, 33–36, 39, 42, 51, 53, 65, 96, 97, 99, 100, 102, 120, 124, 125, 127, 129–31, 133–36 Rapske, B. 83, 97 Reed, J.T. 130, 133, 136 Regier, T. 34, 42 Reid, J. 73, 97 Reid, W. 102, 125 Reimer, A.M. 92, 97 Reyburn, W.D. 71, 97 Reynolds, B.E. 127–36 Rhenferdius, J. 49, 65 Riekert, S.J.P.K. 59, 65 Riley, K. 51, 65 Robertson, A.T. 12, 16, 26, 33, 42, 49, 50, 53, 58, 60, 65 Rodman, R. 63 Ross, W.A. 16 Rovano, M.W. 85, 97 Ruhl, C. 99–103, 105–107, 110–14, 117–25 Runge, S.E. 135

145

Schmitt, N. 65 Selinker, L. 47, 50, 64 Sgall, P. 22, 42 Shannon, C.E. 70, 97 Smith, K.G. 70, 97 Smyth, H.W. 11, 16 Sollamo, R. 57, 65 Spada, N. 50, 51, 64 Sperber, D. 71–74, 76, 77, 97, 98 Stewart, J.S. 6, 16 Stuckenbruck, L.T. 87, 97 Swanson, J. 56, 66 Swete, H.B. 49, 60, 66 Szucsich, L. 16 Taylor, J.R. 9, 16, 34, 42 Thackeray, H.St.J. 46, 62, 66 Thayer, H. 58, 60 Thompson, G. 28, 42, 109, 126 Thompson, J.A. 74, 97 Thompson, S. 49, 66 Thrall, M.E. 36, 42 Tobin, Y. 42 Turner, N. 45, 46, 49, 60, 66 Vallae, L. 53, 66 Wallace, D.B. 58, 66 Watt, J.M. 27, 43 Weaver, W. 70, 97 Webb, J.M. 45, 66 Webster, J.J. 20, 41 Wiens, H. 85, 92, 97 Wilson, D. 71, 72, 74, 76, 77, 97, 98 Winer, G.B. 49, 50, 60, 66 Witherington, B. 83, 98 Wunderlich, D. 15 Xeravits, G.G. 88, 98 Yoder, D. 84, 98 Young, R.A. 55, 56, 66 Zimmermann, I. 15, 16, 43

BACK

McMaster Divinity College Press