THE DISCOURSE PRAGMATIC FUNCTION OF
’ΕΝ ΤΩΙ + INFINITIVE IN POST-CLASSICAL GREEK:
A WINDOW INTO TRANSLATION TECHNIQUE IN SEPTUAGINT GENESIS
by

MATTHEW GLASS

B.A., Christian Ethics, Union University, 2004
M.A., Biblical Studies, Reformed Theological Seminary, 2019

Thesis
Submitted In Partial Fulfillment of the Requirements for
The Degree of
Master of Theology
Septuagint Studies

FACULTY OF GRADUATE STUDIES
TRINITY WESTERN UNIVERSITY

May 2025
Ó Matthew Glass, 2025

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

ii

ABSTRACT
The syntactic construction ἐν τῷ + infinitive has a discourse pragmatic function that can be
clearly defined. Post-classical Greek compositional texts demonstrate that the Septuagint
Genesis translator used the construction consistently with conventional Greek usage. It
encoded presuppositional, identifiable topical information that cognitively anchored asserted
information. It refers anaphorically to a text-internal or text-external state of affairs that is
active in the discourse register while communicating the occasion in time for its predicator.
Due to its bi-directional function, it adds coherence. As one of several renderings in the
Septuagint for the Hebrew ‫ ְבּ‬+ infinitive construction and other Hebrew constructions, its
usage was not arbitrary but displayed an awareness of its function within the Post-classical
Greek linguistic system. Understanding its function within the clause and the broader
discourse, especially in contrast with other renderings for the Hebrew ‫ ְבּ‬+ infinitive, gives a
window into translation technique in Septuagint Genesis.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

iii

CONTENTS
ABSTRACT.............................................................................................................................. ii
TABLES AND FIGURES........................................................................................................ vi
SIGLA AND ABBREVIATIONS...........................................................................................vii
CHAPTER 1: INTRODUCTION.............................................................................................. 1
1.1. Introduction................................................................................................................... 1
1.2. Statement of Research Problem & Question................................................................ 2
1.3. Statement of Purpose.................................................................................................... 3
1.4. Statement of Hypothesis............................................................................................... 3
1.5. Statement & Explanation of Methodology................................................................... 3
1.5.1. Information Structure Analysis............................................................................ 4
1.5.2. Ancient Greek Information Structure.................................................................10
1.5.3. Corpus Linguistics............................................................................................. 12
1.6. Outline of the Study.................................................................................................... 14
CHAPTER 2: SYNTACTIC & LINGUISTIC ANALYSIS................................................... 17
2.1. Introduction................................................................................................................. 17
2.2. Grammatical & Syntactic Literature........................................................................... 19
2.2.1. Friedrich Blass, Albert Debrunner, and Robert Funk (1896-1961).................. 19
2.2.2. A. T. Robertson (1914)....................................................................................... 20
2.2.3. Maximilian Zerwick (1963)................................................................................21
2.2.4. Nigel Turner (1963)........................................................................................... 21
2.2.5. Ilmari Soisalon-Soininen (1965)........................................................................ 22
2.2.6. Basil Mandilaras (1973).................................................................................... 23
2.2.7. Stanley Porter (1989).........................................................................................24
2.2.8. Daniel Wallace (1996)....................................................................................... 25
2.2.9. Takamitsu Muraoka (2016)................................................................................26
2.3. Presuppositions about the Septuagint's Language...................................................... 25
2.3.1. Hebrew-Priority vs. Greek-Priority................................................................... 25
2.3.2. Conventional Post-classical Greek.................................................................... 27
2.3.3. Middle-level Register with Literary Elements................................................... 31
2.3.4. Translation with Hebraic Interference.............................................................. 34
2.4. Linguistic Analysis of ’Εν Τῷ + Infinitive................................................................. 38
2.4.1. Structural Semantic Analysis of ’Εν Τῷ + Infinitive......................................... 39
2.4.2. Information Structure Analysis of ’Εν Τῷ + Infinitive...................................... 44
CHAPTER 3: ANALYSIS OF POST-CLASSICAL COMPOSITIONAL TEXTS............... 48
3.1. Introduction................................................................................................................. 48
3.2. Post-classical Literary Texts....................................................................................... 50
3.2.1. Polybius, The Histories 1.23.8........................................................................... 51
3.2.2. Polybius, The Histories 1.51.9........................................................................... 52
3.2.3. Polybius, The Histories 1.62.4........................................................................... 53

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

iv

3.2.4. Polybius, The Histories 2.29.3........................................................................... 54
3.2.5. Polybius, The Histories 2.32.10......................................................................... 56
3.2.6. Polybius, The Histories 3.79.10......................................................................... 58
3.2.7. Polybius, The Histories 4.64.7........................................................................... 59
3.2.8. Polybius, The Histories 5.52.8........................................................................... 60
3.2.9. Polybius, The Histories 6.42.2........................................................................... 61
3.2.10. Polybius, The Histories 21.4.5......................................................................... 61
3.3. Post-classical Papyri................................................................................................... 62
3.3.1. PSI. 4.354 (254 BCE).........................................................................................63
3.3.2. P.Cair.Zen. 3.59379 (ca. 254-51 BCE)............................................................. 64
3.3.3. P. Koeln. 8.327 (250 - 150 BCE)....................................................................... 65
3.3.4. P.Tebt. 3.1.703 (ca. 210 BCE).......................................................................... 67
3.3.5. UPZ. 1.110 (21 September 164 BCE)................................................................ 68
3.3.6. P.Oxy. 4.743 (4 October 2 BCE)....................................................................... 69
3.4. Post-classical Inscriptions........................................................................................... 70
3.4.1. IvP II 252 (2nd-1st Century BCE)..................................................................... 70
3.4.2. I.Prose. 30 (96 BCE)..........................................................................................71
3.5. Summary..................................................................................................................... 73
CHAPTER 4: ANALYSIS OF SEPTUAGINT GENESIS TEXTS........................................ 74
4.1. Introduction................................................................................................................. 74
4.2. ’Εγένετο (ἔσται) + ’Εν Τῷ + Infinitive (Rendering ‫ ְבּ‬+ Infinitive).............................77
4.2.1. Genesis 4:8........................................................................................................ 78
4.2.2. Genesis 9:14...................................................................................................... 79
4.2.3. Genesis 11:2.......................................................................................................80
4.2.4. Genesis 19:29.....................................................................................................81
4.2.5. Genesis 35:17.....................................................................................................82
4.2.6. Genesis 35:18.....................................................................................................83
4.2.7. Genesis 38:28.....................................................................................................84
4.3. ’Εγένετο (ἔσται) + ’Εν Τῷ + Infinitive (Rendering Other Constructions) ................ 85
4.3.1. Genesis 24:52.....................................................................................................85
4.3.2. Genesis 42:35.....................................................................................................85
4.3.3. Genesis 44:31.....................................................................................................87
4.4. ’Εν Τῷ + Infinitive (Rendering ‫ ְבּ‬+ Infinitive)........................................................... 87
4.4.1. Genesis 19:33.....................................................................................................87
4.4.2. Genesis 19:35.....................................................................................................88
4.4.3. Genesis 28:6.......................................................................................................89
4.4.4. Genesis 32:20.....................................................................................................89
4.4.5. Genesis 32:26.....................................................................................................90
4.4.6. Genesis 34:22.....................................................................................................91
4.4.7. Genesis 35:1.......................................................................................................93
4.4.8. Genesis 35:7.......................................................................................................94
4.5. ’Εν Τῷ + Infinitive (Rendering Other Constructions)................................................ 95
4.5.1. Genesis 19:16.....................................................................................................95
4.5.2. Genesis 34:15.....................................................................................................96

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

v

4.6. Summary..................................................................................................................... 97
CHAPTER 5: ANALYSIS OF OTHER GREEK RENDERINGS......................................... 98
5.1. Introduction................................................................................................................. 98
5.2. Discourse Pragmatic Function of Other Renderings................................................ 102
5.2.1. Finite Verb....................................................................................................... 102
5.2.2. Temporal conjunction + Finite Verb............................................................... 102
5.2.3. Participle..........................................................................................................106
5.2.4. Παρὰ Τὸ + Infinitive........................................................................................ 109
5.2.5. ’Εν Τῷ + Noun.................................................................................................110
5.2.6. Adverb.............................................................................................................. 110
5.2.7. Noun................................................................................................................. 111
5.3. Summary................................................................................................................... 112
CHAPTER 6: CONCLUSION – A WINDOW INTO TRANSLATION TECHNIQUE......113
6.1. Summary of Findings................................................................................................113
6.2. Translation Issues......................................................................................................115
6.2.1. Issue of Register............................................................................................... 115
6.2.2. Issue of Frequency........................................................................................... 116
6.3. Areas for Further Research....................................................................................... 117
WORKS CITED.................................................................................................................... 119
APPENDIX 1: ’ΕΝ ΤΩΙ + INFINITIVE FREQUENCY IN POST-CLASSICAL TEXTS. 127

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

vi

TABLES AND FIGURES

Tables
Table 1.5.1: Key Terms in Information Structure.................................................................... 7
Table 3.1: ’Εν Τῷ + Infinitive in Post-Classical Literary Authors before CE.........................50
Table 4.1: ’Εν τῷ + Infinitive in Genesis.................................................................................77
Table 5.1: Frequency of Renderings for ‫ ְבּ‬+ infinitive in LXX Genesis............................... 100
Table 5.2.2: Temporal conjunction + Finite Verb Rendering Salient States of Affairs........ 104
Figures
Figure 2.4.1.1: Occasion in Time.............................................................................................41
Figure 2.4.1.2: Temporal Scope 1............................................................................................ 41
Figure 2.4.1.3: Temporal Cause 1............................................................................................ 42
Figure 2.4.1.4: Temporal Cause 2............................................................................................ 43
Figure 2.4.1.5: Temporal Scope 2............................................................................................ 43
Figure 2.4.1.6: Temporal Cause 3............................................................................................ 44

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

vii

SIGLA AND ABBREVIATIONS

Sigla
1x, 2x, etc.
+

referent appears one time (1x), two times (2x), etc., depending on the numeral
preceding x.
with; typically used for syntactic constructions like "ἐν τῷ + infinitive"
cause
Abbreviations

AG
BCE
BDF
BH
c.
CE
CG
EG
Gen
IS
LSB
LXX
MT
NETS
NT
PC
SS
TLG

Ancient Greek
before Common Era
Blass, Friedrich, Albert Debrunner, and Robert W. Funk. A Greek Grammar
of the New Testament and Other Early Christian Literature. Chicago: The
University of Chicago, 1961.
biblical Hebrew
circa
Common Era
Classical Greek
Epic Greek
LXX Genesis translator
information structure
Legacy Standard Bible
Septuagint
Masoretic Text
New English Translation of the Septuagint
New Testament
Post-classical
structural semantics
Thesaurus Linguae Graecae® Digital Library, ed. Maria C. Pantelia
(University of California) online at http://www.tlg.uci.edu

Papyri abbreviations are according to the designations found online at:
papyri.info/docs/checklist
Literary sources are cited based on the titles found online at:
loebclassics.com
Literary sources are abbreviated based on the titles found online at:
stephanus.tlg.uci.edu/

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

1

CHAPTER 1: INTRODUCTION
"Much nonsense has been written, without a sufficient knowledge of the Koine, about the
Hebraisms of the Greek Bible."1
–Raija Sollamo
1.1. Introduction
The Septuagint Pentateuch is a Greek translation from a Hebrew Vorlage that began in
the first half of the third century BCE. Due to its historical position as the largest extant
corpus of Greek prose from the Hellenistic era, it provides insight into the development of
PC Greek.2 As an exemplar of PC Greek, its text-linguistic character indicates that the LXX
translators were well-educated and could capably use conventional Greek.3 However,
influence from the source text also occurred, which is common in translated texts. Influence
from the Hebrew Vorlage is sometimes labelled "Hebraism" or "Semitism." This influence is
evinced by the usage of certain lexemes and syntactic constructions and is the result of
translation technique.4 This translation technique gave the LXX a style that distinguished it in
some ways from contemporary compositional texts. Due to serial fidelity (following Hebrew
word order) and isomorphic correspondence with the source text (using grammatical forms
similar to those in Hebrew), it has been observed that the LXX contains a higher frequency
of parataxis and nominal clauses and a lower frequency of participial phrases such as the
genitive absolute.5 Most syntactic patterns indicative of LXX Greek, though, are not unusual
and well-attested in Greek compositional corpora. Since the publication of documentary
sources such as papyri and inscriptions, enormous insight has been gained concerning PC
non-literary Greek. Scholars now recognize that Hebrew influence was less pronounced than
once thought because many features of biblical Greek that were assumed to be

1

Raija Sollamo, "Some 'Improper' Prepositions, such as ἐνώπιον, ἐναντίον, ἔναντι, etc., in the
Septuagint and Early Koine Greek," VT 25 (1975): 776.
2
James K. Aitken and Marieke Dhont, "The Septuagint with the History of Greek: An Introduction,"
Journal for the Study of Judaism 54 (2023): 444.
3
John A. L. Lee, The Greek of the Pentateuch: Grinfield Lectures on the Septuagint 2011-2012,
(Oxford: Oxford University, 2018), 123-72.
4
Ibid., 122.
5
Eberhard Bons, "The Septuagint and Greek Style," The T&T Clark Handbook of Septuagint
Research, eds. William A. Ross and W. Edward Glenny (London: T&T Clark, 2021), 94.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

2

unconventional have been found in these documentary sources.6 However, some have argued
that LXX Greek exhibits certain syntactic structures with greater frequency than is the case in
compositional texts.7 This type of influence, called positive interference, refers to
conventional syntax that was used with greater frequency because "it happens to resemble a
feature in the source language."8 When the LXX's syntax is considered as a whole, it is
generally recognized that the LXX contains middle-level PC Greek that evinces a distinctive
biblical style due to translation technique.9 Therefore, Septuagint Greek's diversity of style
warrants analysis in relation to both contemporaneous Greek and its Vorlage. This task
requires investigation of the Greek compositional corpora and the underlying Hebrew.
1.2. Statement of Research Problem & Question
The study of LXX syntax is still in early development.10 This thesis will examine one
syntactic construction that was used in the LXX over five hundred times: ἐν τῷ + infinitive.
The construction has often been characterized as a Hebraism.11 It was utilized in the
Pentateuch as a rendering for the Hebrew ‫ ְבּ‬+ infinitive construction approximately one
fourth of the time (less than half the time in the LXX as a whole). Additionally, it was on
occasion used as a rendering for other Hebrew syntactic constructions. Many scholars have
concluded that close formal correspondence with the Hebrew ‫ ְבּ‬+ infinitive construction and
6

Trevor Evans, "The Nature of Septuagint Greek: Language and Lexicography," in The Oxford
Handbook of the Septuagint, eds. Alison Salveson and Timothy Michael Law (Oxford: Oxford
University, 2021), 96.
7
Ibid., 94.
8
Aitken and Dhont, "The Septuagint within the History of Greek," 441.
9
Lee, The Greek of the Pentateuch, 269.
10
Trevor Evans, "The Grammarian Cannot Wait: Thackeray, Muraoka, and the Analysis of Septuagint
Syntax," Journal for the Study of Judaism 54 (2023): 562.
11
This was particularly true before the advent of computer technology to aid in the linguistic study of
AG corpora. See Friedrich Blass, Albert Debrunner, and Robert W. Funk, A Greek Grammar of the
New Testament and Other Early Christian Literature (Chicago: The University of Chicago, 1961),
208; A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research
(Logos Bible Software, 2006), 1069-72; Maximilian Zerwick, Biblical Greek: Illustrated by
Examples, trans. Joseph Smith S. J. (Rome: Editrice Pontificio Istituto Biblico, 1963), 133-35; Nigel
Turner, A Grammar of New Testament Greek: Syntax, vol. 3 (Edinburgh: T. & T. Clark, 1963), 14446; Ilmari Soisalon-Soininen, Die Infinitive in der Septuaginta (Helsinki: Suomalainen
Tiedeakatemia, 1965), 81-83; Basil G. Mandilaras, The Verb in the Greek Non-literary Papyri
(Athens: Ministry of Culture and Sciences, 1973), 340-41; T. Muraoka, A Syntax of Septuagint Greek
(Leuven: Peeters, 2016), 334-35.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

3

a temporal semantic relationship with its predicator12 (the usual way the Hebrew construction
was used) are evidence that it is a Hebraism in the LXX. The form, function, and semantics
in the LXX, they argue, can be explained as the result of a translation technique that
borrowed from the source text. Is this conclusion warranted, though, by a textual comparison
of PC compositional texts and the LXX?
1.3. Statement of Purpose
The goal of this thesis is to discern if the above conclusion about the use of ἐν τῷ +
infinitive is valid. To examine the validity of that conclusion, or lack thereof, it will be
necessary to compare the usage of the syntactic construction in PC compositional texts with
the LXX. Given the hundreds of attestations in compositional texts and especially in the
LXX, it will be necessary to narrow the scope of analysis. Therefore, the analysis will focus
on texts from Polybius and documentary sources (papyri and inscriptions) to determine the
default function of ἐν τῷ + infinitive in the PC Greek linguistic system. This analysis will
then be compared to texts from LXX Genesis to determine if the usage of the syntactic
construction in LXX Genesis was the result of borrowing from the Hebrew Vorlage, or if
there was/were some other reason(s) that can explain its usage in the LXX.
1.4. Statement of Hypothesis
This thesis will argue that Gen employed ἐν τῷ + infinitive in a manner consistent with
conventional Greek in compositional PC texts, demonstrating a clear understanding of its
discourse-pragmatic function and structural semantics.
1.5. Statement & Explanation of Methodology
This thesis will employ information structure analysis (i.e., discourse pragmatics)
within the framework of corpus linguistics. IS analysis is a functional approach to language
that seeks to describe the conditions under which specific grammatical structures are used.

12

Some scholars argue that the Greek corpora outside biblical Greek do not use the syntactic
construction to indicate a temporal semantic relationship with a predicator. The textual data in this
thesis will demonstrate that a temporal semantic relationship with a predicator is prototypical.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

4

This type of analysis is most effectively conducted within the context of a natural language
corpus. Unlike arbitrarily produced sentences, a natural language corpus provides valuable
insight into how native speakers use language. It serves as the best indicator of how a
language grammaticalizes meaning through linguistic forms and creates coherent, cohesive
communication by structuring information. Therefore, it is essential to outline these two
methodologies in greater detail as they will guide the analysis of the textual data in this
thesis.
1.5.1. Information Structure Analysis
Two approaches to linguistic analysis are relevant for this thesis. The first approach
analyzes language by examining its syntactic structure. This approach describes syntactic
structure not simply "for its own sake,"13 but also to determine the meaning in the text, i.e.,
structural semantics. The structural semantic approach identifies four categories to which all
syntactic forms belong: subject, predicator, complement, and adjunct.14 Within each of these
syntactic categories are various levels of complexity and formal structures. The structural
semantic approach, therefore, aims to describe the forms and structures of phrases, clauses,
and sentences and their semantic relationship. However, the structural semantic approach
does not describe why particular forms and structures are used in contrast to others within a
language, especially when two different formal structures are capable of encoding the same
meaning.
The second approach analyzes language to answer why a particular structure is used in
contrast to other choices within the linguistic system and how that structure functions in
communication. This is the functional approach to linguistic analysis, and IS analysis is one
such functional approach. However, IS analysis does not discard the structural semantic
approach, but is based upon it.15 Therefore, the description of linguistic forms and structures
13

Robert A. Dooley, “Functional Approaches to Grammar: Implications for SIL Training”
(manuscript). Cited in Stephen H. Levinsohn, Discourse Features of New Testament Greek: A
Coursebook on the Information Structure of the New Testament Greek, 2nd ed. (Dallas: SIL
International, 2000), vii. Dooley viewed the structural approach as describing structure essentially
"for its own sake."
14
S. E. Porter and M. B. O'Donnell, "Building and Examining Linguistic Phenomena in a Corpus of
Representative Papyri," in The Language of the Papyri, eds. T. V. Evans and D. D. Obbink (Oxford:
Oxford University, 2010), 299.
15
Levinsohn, Discourse Features of New Testament Greek, viii.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

5

provided by a structural semantic approach is used as the basis upon which to explain how
specific constituents are utilized within a sentence to contribute to the coherence of the
discourse. Additionally, IS analysis aims to explain the linguistic conditions that engender
the use of certain constituents as opposed to others within a linguistic system.
This thesis utilizes IS analysis as it is outlined in Knud Lambrecht's work on the topic.16
His work, though perhaps dated, was influential on Stephen Levinsohn and Steven Runge,
both of whom have written on discourse analysis in New Testament Greek.17 His work has
also been influential on other scholars who have written on IS, as is evident by the citations
in The Oxford Handbook of Information Structure.18 Lambrecht defines IS as "[t]hat
component of sentence grammar in which propositions as conceptual representations of
states of affairs are paired with lexicogrammatical structures in accordance with the mental
states of interlocutors who use and interpret these structures as units of information in given
discourse contexts."19 This definition is lexically dense because Lambrecht is attempting to
summarize his methodology succinctly. A clarification of this definition will be helpful.
Lambrecht argues that cross-linguistically, there are basic elements shared by all acts of
communication. At the very least, communication occurs between a speaker and a listener.
These interlocutors can communicate successfully when they understand each other's
utterances. Any new information added throughout their discourse must be added to already
given information for that information to be appropriately understood.20 Coherence results
when this is done successfully. Perhaps this principle can be illustrated by the
communication that occurred recently when someone gave driving directions. The person

16

Knud Lambrecht, Information Structure and Sentence Form: Topic, Focus, and the Mental
Representations of Discourse Referents (Cambridge: Cambridge University, 1994).
17
Levinsohn's work is cited above. See Steven E. Runge, Discourse Grammar of the Greek New
Testament: A Practical Introduction for Teaching and Exegesis (Bellingham, WA: Lexham Press,
2010). A comparison of the first and second editions of Levinsohn's Discourse Features demonstrates
this point. Runge also articulated Lambrecht's influence on his own work in a personal conversation.
18
Caroline Féry and Shinichiro Ishihara, eds., The Oxford Handbook of Information Structure
(Oxford: Oxford University, 2016).
19
Ibid., 5.
20
Although Lambrecht uses these terms, he also states that they have caused "great confusion"
(Information Structure and Sentence Form, 46). So, he uses other terms more frequently for the sake
of clarification and to better nuance his position. These terms will be outlined in the following
discussion.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

6

who gave directions needed to begin with a location that was known to the listener.
Otherwise, the listener would be lost. Only when a commonly known location was shared
between the interlocutors could the directions proceed successfully. The person who gave
directions stated:
(1) "From here, there's an intersection a hundred meters up ahead."
The person giving directions assumed the listener already knew the denotation of the
word "here," which did not need to be uttered. It would have been assumed by both
interlocutors that the listener would proceed from their current location. The word, then, is
presuppositional based upon their shared physical location in the "utterance-external world."
So, new information about the intersection was added to the given information. Once the new
information was communicated (i.e., activated within the discourse register), it became the
given information to which the next chunk of new information was anchored. The person
giving directions then said:
(2) "At the intersection, make a left turn."
Once again, the constituent, "At the intersection," contained given information and was the
basis for the addition of the new information, "make a left turn." Those same directions could
have been communicated without the given information, "At the intersection." However, the
addition of the given information enhanced the clarity and coherence of the utterance because
it allowed the new information to be cognitively anchored to something the listener already
knew.
Lambrecht argues that all communication proceeds in the same way as this brief
conversation containing driving directions. Lambrecht's methodology uses terms that better
nuance the ideas of "given information" and "new information." As the definition above
indicates, the encoding of IS is connected to what the speaker21 assumes to be the mental
state of the listener. This includes what the speaker assumes to be the ideational "common
ground"22 between the interlocutors, and therefore what the speaker assumes the listener will
cognitively identify in an utterance. Assumptions based on common ground are the reason
that communication is much easier in general with people whom we know well, rather than

21
22

Reference to a speaker can be substituted with the word "author," listener with the word "reader."
Lambrecht, Information Structure and Sentence Form, 59.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

7

with people with whom we are recently acquainted. Much of the conversation with new
acquaintances is an attempt to discern their mental state: what they know, what they have
experienced, where they have been, who they know, etc. Discerning these seemingly trivial
details of information aids communication because they allow interlocutors to make
appropriate assumptions and communicate coherently. Lambrecht nuances given
information–new information with three categories: presupposition–assertion (which concern
ideas in the text), identifiability–activation (which concern textual constituents), topic–focus
(which concern how textual constituents relate pragmatically).23
Table 1.5.1: Key Terms in IS
Given information
Presupposition (common ground)
Identifiability (active, accessible,25 inactive)
Topic
+ – Accentuation

New state of information24
Assertion (new ground)
Activation
Focus (can include a "thetic topic")
+ Accentuation

In these categories, the terms "presupposition," "identifiability," and "topic" – while not
synonymous – nuance the idea of given information in various ways. The ideas in an
utterance that a speaker assumes the listener will know or is willing to take for granted
constitute a presupposition;26 this information, as mentioned above, is what the speaker
assumes to be the common ground with the listener. In the two texts above concerning
driving directions, there are some shared ideas between the speaker and listener, although the
ideational content is not very profound. The person giving directions assumed (due to shared
physical space in the utterance-external world) that the person receiving directions knew
where they were located spatially and understood how to proceed from their current location

23

Ibid., 6.
There is a distinction between new information and a new state of information. New information
can be defined as things not known before that are now asserted and made known. A new state of
information, however, while including new information can also include things known before (i.e.,
given information) that are given a new relationship in a listener's mind. See Lambrecht (Information
Structure and Sentence Form, 210) for a more extensive discussion.
25
Information that is accessible can either be accessible textually, situationally, or inferentially (see
Lambrecht, Information Structure and Sentence Form, 109).
26
Lambrecht's definition of presupposition: "The set of propositions lexicogrammatically evoked in a
sentence which the speaker assumes the hearer already knows or is ready to take for granted at the
time the sentence is uttered" (Information Structure and Sentence Form, 52).
24

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

8

to another location. So, a cognitively identifiable constituent27 was used in the utterance. The
category identifiability, then, has to do with the grammaticalization of presuppositional
information. The speaker giving directions uttered the prepositional phrase "from here" based
upon the assumption that it would be identifiable to the listener. Information that possesses a
status as presuppositional and cognitively identifiable is used as the topic of the utterance.28
The topic is what the utterance is all about, and both speaker and listener must have a shared
assumption concerning the “aboutness” of an utterance if successful communication is to
occur. The topic of the first utterance in the driving directions would be something like,
"[Where to go] from here."
The corresponding categorical terms, "assertion," "activation," and "focus," likewise
nuance the idea of new information in various ways. An assertion contrasts with a
presupposition. In the driving directions, the first sentence asserted that "there's an
intersection a hundred meters up ahead." For the person who received the directions, this is
new information. The newly asserted information activates its ideational referent within the
driving directions discourse register.29 Therefore, any subsequent anaphoric reference to the
information will be identifiable to the listener. However, when it is in the process of
activation, the new information will not be identifiable; this, then, will be reflected in the IS.
Newly asserted information that is in the process of activation within the discourse register is
the focus of an utterance. Lambrecht describes different types of focus structures: predicatefocus, argument-focus, and sentence-focus.30 If there is no presuppositional topic, which is
often the case at the beginning of communication or during transitions to new topics, then a
topic must be asserted. When this is the case, the entire utterance is in focus (i.e., sentencefocus structure). These utterances are referred to as "thetic sentences," or "presentational
sentences."31 Topic and focus, then, are not always binary since a sentence's topic can also be
in focus.

27

In a written text, it would be appropriate to use the words "textual constituent."
The topic is not always presuppositional; see the following comment about "thetic sentences."
29
Lambrecht, Information Structure and Sentence Form, 74.
30
Ibid., 222.
31
N. A. Bailey, “Thetic Constructions in Koine Greek: with special attention to clauses with εἰμί 'be',
γίνομαι 'occur', ἔρχομαι 'come', ἰδού/ἴδε 'behold', and complement clauses of ὁράω 'see'” (PhD diss.,
Vrije Universiteit Amsterdam, 2009), 1.
28

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

9

There is a sociolinguistic dimension to the distinction between presupposition and
assertion, between identifiable information and information that is being activated. James
Paul Gee refers to this as a "social contract between the writer and reader."32 Speakers
structure their utterances in such a way that listeners are expected to treat certain elements as
common ground (regardless of whether the information was known by the listener). The
expectation on the part of the speaker is that the listener would "address any comments or
disagreements to the asserted information and simply take the assumed information for
granted."33 This basic reality about communication can become a tool used by speakers "to
manipulate listeners and readers into not bringing up (or not thinking too much about) what
they don’t want brought up or thought about."34 This is all accomplished by the way
information is structured.
The final term listed above in Figure 1.5.1 is accentuation. Accentuation must be viewed
as a pragmatic phenomenon that is distinct from focus. According to Lambrecht, "The focus
is that portion of a proposition which cannot be taken for granted at the time of speech."35 In
other words, focus concerns what the speaker wants the listener to know (the new state of
information). Focus is connected directly to what the speaker perceives to be the mental state
of the listener, and therefore what needs to be asserted and activated in the discourse.
Accentuation, on the other hand, has nothing to do with the speaker's perception of the
mental state of the listener, but everything to do with what the speaker wants the listener to
perceive as pragmatically salient.36 Therefore, accentuation certainly happens on
constituents in the focal domain, but can also happen with constituents in the topical
domain.37 This point will be important in chapter five when considering various renderings
that were used for the Hebrew ‫ ְבּ‬+ infinitive construction in LXX Genesis.
The mental state of interlocutors is shaped by two worlds. The first world is the textexternal world. According to Lambrecht, the text-external world "comprises (i) speech

32

James Paul Gee, Introducing Discourse Analysis: From Grammar to Society (London: Routledge,
2018), 65.
33
Ibid., 65.
34
Ibid., 65.
35
Lambrecht, Information Structure and Sentence Form, 207.
36
Ibid., 325. Pragmatic salience is also referred to as prominence.
37
Ibid., 325.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

10

participants, i.e., a speaker and one or several addressees, and (ii) a speech setting, i.e., the
place, time, and circumstances in which a speech event takes place."38 Returning to the texts
about driving directions: if the person who gave directions – instead of making a statement
such as "there's an intersection a hundred meters up ahead " – pointed to a landmark that
both interlocutors could see, "You see that up ahead," then reference would be made to the
text-external (utterance-external) world. Coherent communication would then proceed based
upon the presupposition they both possessed due to their physical location. The second world
that shapes the mental state of interlocutors is the text-internal world. This is the textlinguistic discourse world containing the ideas and textual constituents that constitute an
instance of communication.
1.5.2. Ancient Greek Information Structure
Cross-linguistically, there are six linguistic means through which IS can be
accomplished: intonation, lexemes, morphemes, particles, syntax, and word order.39 Each
language uses these means in different ways to encode IS. Some languages use only one of
these linguistic means (e.g., intonation in English), while others use a combination that
pragmatically interacts in language-specific ways.40 Helma Dik has argued that AG primarily
used word order to indicate IS.41 Her most recent monograph provides her updated schema
for IS in AG: [Setting] – Topic – Focus – Verb – [Remainder].42 Her basic premise is that the
setting and topic (both of which are given information) precede the focal constituent in the
sentence. The focal constituent is then followed by the verb and what she terms the
remainder constituents. Dejan Matić has shown that Dik’s schema for IS is only valid 49% of

38

Ibid., 36.
Nicolas Bertrand, "Information Structure and Greek," in Encyclopedia of Ancient Greek Language
and Linguistics, vol. 2, ed. Georgios K. Giannakis (Leiden: Brill, 2014), 241. Bertrand does not list
lexemes as a means of IS, but Lambrecht states that "the lexicon" can contribute (Information
Structure and Sentence Form, 31). In AG, articles and pronouns played a role. See Geoffrey
Horrocks, Greek: A History of the Language and Its Speakers, 2nd ed. (Chichester, UK: Wiley
Blackwell, 2010), 97, where he states that deictic pronouns were used to "control discourse structure."
40
Lambrecht, Information Structure and Sentence Form, 25, 31.
41
Helma J. M. Dik, “Word Order in Ancient Greek: A Pragmatic Account of Word Order Variation in
Herodotus” (PhD diss., University of Amsterdam, 1995). Also, Helma J. M. Dik, Word Order in
Greek Tragic Dialogue (Oxford: Oxford University, 2007).
42
Dik, Word Order in Greek Tragic Dialogue, 38.
39

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

11

the time in Xenophon's Anabasis.43 Furthermore, he states that the schema does not
accurately describe the other 51% of sentences due to "three major phenomena: (i) postverbal
elements that do not fit the label 'pragmatically unmarked' (henceforth postverbal foci); (ii)
more than just two elements before the verb (henceforth complexity of the left periphery);
and (iii) elements between the focus expression and the verb (henceforth focus intruders)."44
So, Matić's work on IS has built upon Dik's insights and has better nuanced her theory of AG
discourse pragmatics. He has attempted to account for the discrepancies in her word order
schema. Along with taking into consideration the role of syntax in IS, 45 he notes that
intonation could have possibly played a role in AG.46 However, scholars do not possess any
evidence for the intonation patterns used by speakers of AG since native speakers ceased to
exist millennia ago and they left no record of how they intonated their language to encode IS.
Dik’s and Matić's scholarship possesses many insights related to IS in AG compositional
texts. However, they provide little help for finding intentionality in the way the LXX
translators pragmatically structured information, since word order in the LXX is largely
dependent on the Hebrew source text. When the translators produced the Greek text, they had
little concern for capturing the interface between word order and IS that was natural and
intuitive to native Greek speakers. Scholars are only left with three linguistic means, then, for
discovering intentionality in the way LXX translators pragmatically structured information:
particles, syntax, and specific lexemes. In the usage of these three linguistic means, IS can
often be discerned where the translator possessed multiple translation equivalents from which
to choose to render the Hebrew Vorlage, especially when those renderings function
differently on the level of discourse.47 Therefore, the choice to use one specific rendering was
also a choice not to use other possible renderings. There has already been some helpful

43

Dejan Matić, "Topic, Focus, and Discourse Structure: Ancient Greek Word Order," Studies in
Language 27:3 (2003): 578.
44
Ibid., 578.
45
Ibid., 603.
46
Ibid., 586.
47
Marieke Dhont has argued that "a translator's decisions during the translation process are governed
by a multidimensional interplay of various factors that are determined by the translator's context"
("Septuagint Translation Technique and Jewish Hellenistic Exegesis," in T&T Clark Handbook of
Septuagint Research, eds. William A. Ross and W. Edward Glenny [London: T&T Clark, 2021], 24).
Pragmatic IS is one factor.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

12

scholarship done on how LXX translators used particles on the level of discourse.48 There
has also been some work done on how certain lexemes contribute to IS.49 Similarly, this
thesis will attempt to show IS using the linguistic means of syntax, particularly how a single
Greek syntactic construction encodes IS. This involves not quite the same approach to the
consideration of the role of syntax as that of Matić, who is more concerned with the
relationship between IS and the constituent ordering of subject, predicator, complement, and
adjunct.50 But it is consistent with Lambrecht’s concern to articulate the interface between
syntax and IS.51
There are a variety of syntactic constructions that could be used to communicate the same
propositional content. However, authors choose certain syntactic constructions due to the
concerns of pragmatic IS.52 Chapter five will discuss that the same propositional content
could be communicated using ἐν τῷ + infinitive, a participle, or a temporal conjunction +
finite verb. However, each syntactic construction serves a different pragmatic function. When
Gen chose one rendering and chose not to use the others, he was encoding a specific IS based
upon the discourse context and the perspective that he was intending to communicate.
1.5.3. Corpus Linguistics
Porter and O'Donnell define corpus linguistics as "the computer-aided empirical study of
naturally occurring language that has been collected into a representative sample, that is, the

48

See Christopher J. Fresch, Discourse Markers in Early Koine Greek: Cognitive-Functional Analysis
and LXX Translation Technique (Atlanta: SBL Press, 2023). Some other scholarship cited by Fresch
include Philippe Le Moigne, “Le livre d’Ésaïe dans la Septante: Ecdotique, stylistique, linguistique”
(PhD diss., L’École pratique des hautes études, 2001); Frank Polak, “Context Sensitive Translation
and Parataxis in Biblical Narrative,” in Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea
Scrolls in Honor of Emanuel Tov, ed. Shalom M. Paul, VTSup 94 (Leiden: Brill, 2003); Claude Cox,
“Tying It All Together: The Use of Particles in Old Greek Job,” BIOSCS 38 (2005): 41–54; Anneli
Aejmelaeus, Parataxis in the Septuagint: A Study of the Renderings of the Hebrew Coordinate
Clauses in the Greek Pentateuch, Dissertationes Humanarum Litterarum 31 (Helsinki: Suomalainen
Tiedeakatemia, 1982).
49
Stephen H. Levinsohn, "Towards a Unified Linguistic Description of οὗτος and ἐκεῖνος," in The
Linguist as Pedagogue: Trends in the Teaching and Linguistic Analysis of the Greek New Testament,
eds. Stanley E. Porter and Matthew Brook O’Donnell (Sheffield: Sheffield Phoenix Press, 2009),
206–219.
50
Matić, "Topic, Focus, and Discourse Structure: Ancient Greek Word Order," 629.
51
Lambrecht, Information Structure and Sentence Form, 1-2.
52
Ibid., 9.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

13

corpus."53 This definition indicates that, as a method, corpus linguistics compiles a
"representative sample" of naturally occurring language. As mentioned above, this naturally
occurring language is contrasted with texts that are arbitrarily produced, such as those that
were so often analyzed in generative grammar.54 This is particularly important for studying
IS because it is a grammatical component dependent on the discourse context in which
utterances are found. So, the goal in studying a language corpus is to better understand how
native speakers use language within a natural linguistic environment. Additionally, the
definition indicates that corpus linguistics is an empirical study. Corpus linguistics is,
therefore, inherently descriptive, and its conclusions about natural language are rooted in
observations made about the data in the corpus. Finally, Porter and O'Donnell's definition
indicates that corpus linguistics is computer-aided. To get an appropriate amount of data to
justify making general conclusions about natural language, a large quantity of data must be
analyzed. Only computer-aided technology allows linguists to appropriately analyze this
large quantity of data. Otherwise, they would have to manually analyze thousands and
thousands of documents just to find representative examples, which is what grammarians did
before the advent of computer-aided technology. This is why they made statements at times
that were based more on their intuition than on empirical data. They did not have the time to
manually analyze all the data from AG corpora, therefore, their conclusions were often based
on a few scattered texts and their intuitive understanding of the Greek language (which was
frequently right). However, one reason why ἐν τῷ + infinitive in the LXX has been wrongly
characterized as a Hebraism is perhaps due to the inability in previous generations to
thoroughly analyze the AG corpora. This inability is no longer the case with computer-aided
technology.
This thesis will examine texts in two PC Greek corpora. The first corpus consists of
literary texts. These texts have been accessed using the online databases Thesaurus Linguae
Graecae55 and the Loeb Classical Library.56 The statistics for the usage of the construction
53

S. E. Porter and M. B. O’Donnell, "Building and Examining Linguistic Phenomena in a Corpus of
Representative Papyri," in The Language of the Papyri, eds. T. V. Evans and D. D. Obbink (Oxford:
Oxford University, 2010), 289.
54
See Noam Chomsky, Syntactic Structures, 2nd ed. (Berlin: Mouton de Gruyter, 2002), 15.
55
https://stephanus.tlg.uci.edu
56
https://www.loebclassics.com

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

14

by an author or a text in the PC Greek literary corpus will be given in chapter three, where it
will be shown that the syntactic construction was used 397 times by 62 authors. Ideally,
every single usage of the syntactic construction should be examined to fully demonstrate how
the construction was used within the PC Greek linguistic system. However, there is not
enough space in this thesis to appropriately carry out that task. Therefore, it is necessary to
limit the scope of texts that will be discussed. Rather than discussing texts randomly, which
could perhaps be beneficial, one author has been chosen: Polybius. Although this might run
the risk of basing conclusions on a few scattered texts, Polybius's historical accounts
represent a genre similar to that of LXX Genesis and are somewhat indicative of PC Greek
prose.57 Additionally, the corpus of texts found in documentary sources – papyri and
inscriptions – will be discussed. These non-literary texts occupy a different language level
than that of Polybius and are more telling of how Greek was used in everyday life, both in
formal settings and in informal correspondence between individuals. Even though only
twenty texts will be discussed from two different compositional corpora, the hope is that this
will be a broad enough sample from a sociolinguistic perspective to elucidate the default
function of the construction in the PC linguistic system.
1.6. Outline of the Study
Chapter two of this thesis will deal with several issues concerning linguistic and syntactic
analysis. This chapter will be important because it will articulate key issues that are directly
related to the analysis of the texts that contain the construction. Chapter two will begin with a
summary of how grammarians have understood this construction (§2.2). This section will
function as a literature review. Next, the presuppositions that guide this thesis will be
outlined (§2.3). There will be three presuppositions that will be discussed. First, the language
of the LXX – including its syntax – is a specimen of conventional PC Greek. Second, the
LXX occupies a middle-level linguistic register and contains literary flourishes. Third, the
LXX is a translation with Hebraic interference. The translated texts in the LXX were
translated by competent translators who sought to faithfully render their Vorlage; therefore,
the Greek language in the LXX, although conventional in the Hellenistic era, also contains
57

Horrocks, Greek: A History of the Language and Its Speakers, 97.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

15

interference from the Vorlage. It is important to outline these presuppositions. One's
assumptions have a controlling influence on any scientific enquiry concerning how the data
are interpreted. For example, if one presupposes that "the syntax of the Septuagint is Hebrew
rather than Greek,"58 there will be a tendency to label syntactic constructions as Hebraisms.
However, if one maintains that the LXX is generally conventional Greek that possesses
Hebraic influence scattered throughout, then there will exist a more cautious approach in
labelling syntactic units as Hebraisms.59 Specific guidelines will be delineated that help
identify Hebraic influence on the LXX's syntax. Finally, chapter two will conclude with a
discussion of the construction’s role in the IS of AG (§2.4). It is intentional that this
discussion takes place before the discussion of PC Greek texts so that its conclusions might
be seen in the texts that follow.
The third chapter will begin the analysis of PC Greek corpora. As mentioned above, two
corpora in particular will be discussed: PC literary texts and PC documentary texts. To
demonstrate how the construction functioned in the literary texts, ten different texts (eleven
attestations) in Polybius will be described using pragmatic IS analysis. These texts contain
almost half of the attestations of the syntactic construction in Polybius's corpus. Additionally,
six texts (eight attestations) from non-literary papyri will be discussed as well as two
inscriptions. These are the only ten attestations of ἐν τῷ + infinitive that could be found in
documentary sources before the Common Era. Despite their small number, though, these
texts are very important for grasping the prototypical way in which the syntactic construction
was used by people who were native Greek speakers.
The fourth chapter will examine the twenty-three texts in LXX Genesis that use the
syntactic construction. Through a close examination of these texts from a discourse
pragmatic perspective, the thesis will demonstrate that ἐν τῷ + infinitive functioned in the
same way in LXX Genesis as it did in the compositional Greek texts in chapter three. The
texts will be analyzed in four sections. The first section will contain the seven texts where ἐν

58

Johan Lust, Erik Eynikel, and Katrin Hauspie, A Greek-English Lexicon of the Septuagint
(Stuttgart: Deutsche Bibelgesellschaft, 2003), viii-ix.
59
See William Ross, Post-classical Greek and Septuagint Lexicography (Atlanta: SBL Press, 2022),
40-62, for a helpful discussion of the "Hebrew-priority view" versus the "Greek-priority view" of the
language of the LXX.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

16

τῷ + infinitive was used as a translation equivalent for the Hebrew ‫ ְבּ‬+ infinitive
construction, and καὶ ἐγένετο or ἐγένετο δὲ or καὶ ἔσται was placed in the left periphery.60
The second division will contain eleven other texts where ἐν τῷ + infinitive was used as a
translation equivalent for the Hebrew ‫ ְבּ‬+ infinitive construction. The third division will
contain three texts where ἐν τῷ + infinitive was used as a translation equivalent for various
Hebrew constructions, and καὶ ἐγένετο or ἐγένετο δὲ or καὶ ἔσται was placed in the left
periphery. Finally, the fourth division will contain two texts where ἐν τῷ + infinitive was
used as a translation equivalent for various other Hebrew constructions.
The fifth chapter will look at the usage of ἐν τῷ + infinitive from another perspective in
LXX Genesis. This syntactic construction was one of eight renderings used to translate the
Hebrew ‫ ְבּ‬+ infinitive construction. So, this chapter will attempt to demonstrate the functional
differences, from a discourse pragmatic perspective, in the use of these other renderings.
Other texts in LXX Genesis will be analyzed, and the IS that was encoded will be noted.
Understanding the way ἐν τῷ + infinitive was used in contrast to these other renderings will
further elucidate its default function within the AG linguistic system.
The sixth chapter, in conclusion, will offer a summary of the findings. The chapter will
also consider two final issues related to the syntactic construction. First, the register to which
the construction belongs will be discussed. Due to the infrequency of usage in the nonliterary papyri and the more abundant usage in literary texts, the question of whether ἐν τῷ +
infinitive belongs to a more literary register will be considered. Second, the issue of
frequency will also be discussed. This syntactic construction was used quite often in the
LXX. Scholars have often given even conventional Greek syntactic constructions the label
Hebraism due to their unusually frequent usage in the LXX compared to compositional texts.
Third, the conclusion will outline areas of further inquiry.

60

These texts are considered together because καὶ ἐγένετο / ἐγένετο δὲ / καὶ ἔσται have a structural
effect on ἐν τῷ + infinitive; however, there is no effect semantically.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

17

CHAPTER 2: SYNTACTIC AND LINGUISTIC ANALYSIS
"Greek syntax, not Hebrew, is the translators' starting point."61
– John A. L. Lee
2.1. Introduction
It is impossible to evaluate how ἐν τῷ + infinitive is used in LXX Genesis without having
certain presuppositions about the language of the Septuagint. That has always been true for
anyone attempting to analyze the Greek language used in the LXX. Like any scientific
enquiry, though, the existing data – in this case, textual data – ought either to confirm one's
presuppositions or the presuppositions need to be modified. This chapter on syntactic and
linguistic analysis will present three parallel discussions. First, this chapter will briefly
discuss how ἐν τῷ + infinitive has been described by several works on grammar and syntax.
It is important to realize that the grammarians also possessed certain presuppositions about
the language of the LXX. Many of them made judgments about the LXX's language based
upon their educated intuition about AG, which was often shaped by CG usage, rather than by
the PC textual data. Their methodology was understandable because they analyzed the Greek
language at a time when technology did not yet exist to properly conduct corpus linguistics.
They also only examined the Greek language from a structuralist perspective. This
perspective is necessary, but it can never explain the functional use of language within
discourse. These grammarians, though, have been influential regarding the view that ἐν τῷ +
infinitive is a Hebraism.62 That view continues to be repeated based upon the authority of
these previous influential scholars without being properly investigated in light of PC literary
and non-literary corpora. Second, the underlying presuppositions in this thesis about the
LXX's language will be outlined. These guiding assumptions will be justified to some degree
and contrasted with other approaches to the language of the LXX. The relevance of these
guiding assumptions for ἐν τῷ + infinitive will also be articulated. Third, the structural
semantics and discourse pragmatics of ἐν τῷ + infinitive in the Greek corpora will be
61

Lee, The Greek of the Pentateuch, 262.
For an example, see Takamitsu Muraoka, A Syntax of Septuagint Greek (Leuven: Peeters, 2016),
334-35. Muraoka cites and seemingly misunderstands Turner when he concludes that ἐν τῷ +
infinitive is a Hebraism without any comparison between LXX usage and compositional Greek. See
discussion below in §2.2.9.
62

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

18

described. This linguistic description will be used in the subsequent chapters to analyze the
Greek texts from the PC era. It is important to note that this description was composed in
dialogue with the textual data rather than in isolation from it.
2.2. Grammatical and Syntactic Literature
This section outlines the way ἐν τῷ + infinitive has been described in some works on
Greek grammar and syntax. The discussion is not comprehensive but representative of a few
influential scholars who have written on Greek syntax. A few of the following scholars
express certain conclusions about the way ἐν τῷ + infinitive was used in compositional Greek
without providing textual data to support those conclusions. Instead, they cite other scholars
who came to the same conclusions. This is understandable to a certain degree because, as
mentioned above, many of these works were written at a time when computer technology did
not exist to carry out corpus-based linguistic studies. However, it was also reminiscent of the
way lexicographical studies were done in previous generations. Lee has stated, "Dependence
on predecessors over several centuries is a clear feature of New Testament lexicography, and
one that has potentially serious consequences."63 The lexica were not completely in error;
much of the content in the best lexicons helpfully defines lexemes and provides an invaluable
basis for lexical semantics. But Lee argues that there were serious consequences because the
lexica have been produced with "faulty material that has been simply handed on and not
adequately tested."64 The same could be stated about many of the syntactic explanations in
works on Greek grammar and syntax, including explanations of ἐν τῷ + infinitive. This point
will be clear in some of the statements made by the following authors.
2.2.1. Friedrich Blass, Albert Debrunner, and Robert Funk (1896-1961)65
BDF is concerned with NT Greek morphology and syntax from a structuralist
perspective. It helpfully notes that ἐν τῷ + infinitive is consistently temporal in the NT.
However, the authors state that Attic Greek did not use the syntactic construction to indicate

63

John A. L. Lee, A History of New Testament Lexicography, Studies in Biblical Greek, ed. D. A.
Carson (New York: Peter Lang, 2003), 9.
64
Ibid., 9.
65
Friedrich Blass, Albert Debrunner, and Robert W. Funk, A Greek Grammar of the New Testament
and Other Early Christian Literature (Chicago: The University of Chicago, 1961), 208.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

19

a temporal semantic relationship with its predicative.66 They do not provide a relevant
citation or any textual data to support that statement.67 Instead, they express the assumption
that Classical authors preferred participial constructions (unaware perhaps that those authors
used ἐν τῷ + infinitive more than 300x times). They argue, for example, that ἐν τῷ σπείρειν
αὐτόν in Matthew 13:4 would have been written as σπείροντος αὐτοῦ in CG.68 They do note
that the Hebrew ‫ ְבּ‬+ infinitive construction was used temporally.69 Their conclusion, then, is
that the LXX used the construction as a syntactic loan translation. Rather than using ἐν τῷ +
infinitive according to Greek idiom, the translators decided to depart from conventional
usage and use the construction as a rendering for ‫ ְבּ‬+ infinitive for the sake of formal
equivalence. This will be a common explanation in some of the following authors, and it
confirms the limitation of any linguistic explanation that does not consider corpus-based
textual evidence. It also does not explain the texts in which ἐν τῷ + infinitive is used as a
rendering for other Hebrew constructions.
2.2.2. A. T. Robertson (1914)70
A. T. Robertson’s discussion of ἐν τῷ + infinitive is embedded within a much larger
discussion on the use of prepositions with infinitives. He states generally that preposition +
article + infinitive is idiomatic Greek. According to him (contra BDF), it was "Attic in origin
and literary."71 He argues that the article was necessary because there was no division
between words in the orthography of AG. So, the article distinguished preposition +
infinitive from an infinitive that had a prepositional prefix. Robertson cites several classical
authors who attest ἐν τῷ + infinitive, but he does not discuss any of these authors' texts in
which the syntactic construction is used. Nor does he cite any PC authors or texts. Even
though he realizes that the construction was used in CG literary texts, he argues that it was

66

Ibid., 208. The authors do not use this metalanguage.
Debrunner cites Martin Johannessohn, Der Gebrauch der Präpositionen in der Septuaginta (Berlin:
Weidmannsche Buchhandlung, 1925), 335. However, Johannessohn does not discuss Attic usage at
all, but only discusses the LXX.
68
Ibid., 208. There is no doubt that CG authors used participle constructions more frequently than ἐν
τῷ + infinitive. But frequency does not absolutely exclude using another construction.
69
Ibid., 208.
70
A. T. Robertson, A Grammar of the Greek New Testament in the Light of Historical Research
(Logos Bible Software, 2006), 1069-72.
71
Ibid., 1069. His intuition seems to be correct.
67

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

20

not used with temporal semantics by those authors. This is also a statement that has often
been recirculated. Robertson ends his discussion by stating that until Attic parallels are found
containing the sense "during," his tentative conclusion – citing J. H. Moulton’s second
edition – is that it is "possible but unidiomatic Greek."72 Therefore, though aware that ἐν τῷ
+ infinitive was used by CG authors, he seemingly is not familiar with the actual texts in
which the syntactic construction is used. Had he analyzed those texts, he would have found
many examples where ἐν τῷ + infinitive contained the sense "during." Robertson also lived
in an era in which attention to PC usage was in its early stages.
2.2.3. Maximilian Zerwick (1963)73
Maximilian Zerwick attributes the frequency with which the syntactic construction was
used in Luke to the LXX’s influence; Luke was, in a sense, imitating the style of the LXX
texts.74 He states that the usage of ἐν τῷ + infinitive “is of itself quite Greek,” but then states
that “its regular use in the temporal sense may be attributed to Hebrew influence.”75 So, like
Robertson, he regards the temporal semantics so characteristic of the construction's usage in
biblical Greek to be Hebraic. However, Zerwick does not cite any relevant CG or PC texts to
demonstrate that usage in "the temporal sense" was Hebraic. He does cite secondary
literature on Semitisms in Luke's Gospel.76
2.2.4. Nigel Turner (1963)77
Nigel Turner, like BDF and Zerwick, was primarily interested in NT Greek. Most of his
discussion of the construction, then, centers on examples in Luke's corpus. He echoes
Robertson, BDF, and others by stating that ἐν τῷ + infinitive is "non-classical,"78 which
ostensibly means that its usage in biblical Greek was different from that in CG rather than
that it is not used at all in CG (which is how Muraoka seemingly interprets Turner). He

72

Ibid., 1072. See James Hope Moulton, A Grammar of New Testament Greek, Vol. 1: Prolegomena,
2nd ed. (Edinburgh: T&T Clark, 1906), 249.
73
Maximilian Zerwick, Biblical Greek: Illustrated by Examples, trans. Joseph Smith S. J. (Rome:
Editrice Pontificio Istituto Biblico, 1963), 133-35.
74
Ibid., 134.
75
Ibid., 133.
76
Ibid., 134.
77
Nigel Turner, A Grammar of New Testament Greek: Syntax, vol. 3 (Edinburgh: T. & T. Clark,
1963), 144-46.
78
Ibid., 144.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

21

argues that as a Hebraism it is “the usual LXX rendering of ‫ ב‬c. infin. (Heb).”79 Turner was
writing around the same time that Henry Gehman hypothesized that the LXX was an
exemplar of a Jewish Greek dialect (albeit a transitional one). Turner himself embraced the
same thesis and considered it to be the linguistic milieu in which the NT was written.80 Since
his fundamental perspective on the language of the LXX and NT was that they were
thoroughly Hebraistic and not conventional Greek, he tended to have a maximal view of
Hebraic influence on the Greek language employed in the biblical corpora.
2.2.5. Ilmari Soisalon-Soininen (1965)81
Ilmari Soisalon-Soininen focuses his scholarship on the LXX’s syntax. He notes that
even though ἐν τῷ + infinitive appeared very early in the Greek corpora, it was never used
frequently.82 He argues that its meaning in Classical literature was causal.83 But the
Ptolemaic era, he adds, also provides evidence for a temporal usage: "The classical meaning
is initially causal, but, for example, in the papyri of the Ptolemaic period, the temporal
meaning is present."84 From a translation-technical perspective, he notes that ἐν τῷ +
infinitive was natural Greek (“natürliche”) and at the same time a literal rendering of the
Hebrew ‫ ְבּ‬+ infinitive.85 He accurately points out that the LXX translators only used the

79

Ibid., 144. It is uncertain what he means by "usual rendering." There are some books, like Psalms,
where it is used as a rendering 75% of the time. However, the data show that it is very difficult to see
ἐν τῷ + infinitive generally as the "usual rendering" for the Hebrew ‫ ְבּ‬+ infinitive since it is only used
28% of the time in the Pentateuch as a rendering for the Hebrew construction and less than half the
time in the LXX as a whole. Therefore, rather than making general statements such as the one above,
it is more helpful to make statements based upon a particular book in the LXX, especially when there
is so much diversity in translation technique throughout the LXX.
80
Stanley Porter, “Introduction,” in The Language of the New Testament: Classic Essays, ed. Stanley
Porter, Journal for the Study of the New Testament Supplement Series 60 (Sheffield: Sheffield
Academic, 1991), 28-29.
81
Ilmari Soisalon-Soininen, Die Infinitive in der Septuaginta (Helsinki: Suomalainen Tiedeakatemia,
1965), 81-83.
82
This is generally true; however, some authors did use it frequently, such as the Pythagorean
philosopher pseudo-Theages (who used it more than 10x per 1,000 words compared to the LXX's
0.81x per 1,000 words). So, frequency is an issue of idiolect, having great variation from one author
to the next even in compositional texts. See Appendix 1 for frequencies in PC Greek authors/texts.
83
The data do not substantiate this claim. Temporal semantics, as noted in the introduction, is always
in view when the syntactic construction is used, either temporal scope or temporal cause.
84
Soisalon-Soininen, Die Infinitive in der Septuaginta, 81. ("Die klassische Bedeutung ist zunächst
kausal, aber z.B. in den Papyri der Ptolemäerzeit ist die temporale Bedeutung vorhanden.").
85
Ibid., 81.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

22

construction as a translation equivalent for the Hebrew ‫ ְבּ‬+ infinitive part of the time.
Therefore, it was not used as a stereotyped rendering. In his assessment, this shows that the
translators were familiar with the syntactic construction and clearly understood how the
phrase was used in idiomatic Greek, i.e., in a causal manner (from his perspective). So, they
refrained from using it as a stereotyped rendering for the Hebrew construction. He states, "I
would rather see the reason for the frequent rendering with other expressions in the fact that
ἐν τῷ + Inf was initially understood as causal."86 In other words, they realized the
construction primarily had a causal semantic value, so they were reluctant to use it as a
stereotyped rendering for the Hebrew construction.87
2.2.6. Basil Mandilaras (1973)88
Basil Mandilaras focuses on PC non-literary papyri. However, he does interact with the
LXX and NT in light of his analysis of the papyri. He notes that most NT grammars argue for
Hebrew influence on the syntactic construction.89 He cites Moulton (who also influenced
Robertson's analysis). Mandilaras recognizes that the syntactic construction occurred in CG
texts and that it was "fairly well-attested in the papyri" in the PC era.90 He provides data for
the use of thirty-seven prepositions in preposition + article + infinitive constructions. His
data are presented in a table. He lists the prepositions with their article in the left column,
then notes whether the syntactic constructions are attested in Herodotus, Thucydides,
Xenophon, Hellenistic writers, papyri, and the New Testament.91 Only εἰς τὸ + infinitive and
ἐν τῷ + infinitive are attested in the works of each of the authors or corpora. He concludes,
In particular, εἰς τὸ with the infinitive has gained a greater flexibility in the papyri, and
presents a wider usage than is shown in the NT. This construction has been considered by
NT scholars as a unique idiosyncrasy due to Semitic influence, but this assumption must
86

Ibid., 81-82. ("Den Anlass zu der häufigen Wiedergabe mit anderen Ausdrücken würde ich eher
darin sehen, dass ἐν τῷ + Inf, zunächst kausal aufgefasst wurde.").
87
The thesis is arguing that it was not used by the translators as a stereotyped rendering because it
encoded a different pragmatic information structure than the Hebrew construction, not because the
translators knew it really functioned causally in the Greek linguistic system and therefore only
intermittently and reluctantly used it due to its formal similarities with the Hebrew construction.
88
Basil G. Mandilaras, The Verb in the Greek Non-literary Papyri (Athens: Ministry of Culture and
Sciences, 1973), 308-351.
89
Ibid., 344.
90
"Well-attested" may not be the case. It was used 8x in the papyri before the CE, which seems fairly
infrequent.
91
Mandilaras, The Verb in the Greek Non-literary Papyri, 339-40.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

23

be rejected as the papyri also show this usage. Nevertheless, ἐν τῷ with the infinitive
seems to have suffered from Semitic influence as both its treatment and its meaning differ
from that of Classical Greek and papyri.92
But he adds a footnote stating that Moulton and Turner were "inclined to give Semiticism
much greater significance than it really had with regard to this construction."93
2.2.7. Stanley Porter (1989)94
Stanley Porter's analysis is thorough and rooted in his view of verbal aspect, which is
operative in infinitive constructions. Aspectual issues are important because they are a part of
the syntactic construction's semantics. Porter notes that the construction was a natural
translation of the Hebrew ‫ ְבּ‬+ infinitive; additionally, he finds no Hebrew interference on the
construction's verbal aspect. He lists several reasons for this conclusion. First, biblical Greek
contains present and aorist infinitives. This choice was non-existent in Hebrew, therefore the
LXX translators (when using the construction as a rendering for the Hebrew construction)
made a conscious choice as to which aspect to grammaticalize. Second, Porter recognizes
that the construction's usage with temporal semantics is attested "occasionally" in older
Greek, i.e., CG. He notes that this sense became more frequent during the PC era to which
biblical Greek belonged.95 Third, the construction's function in biblical Greek is not only
temporal but can be semantically related to its predicator in other ways, as well (i.e., cause,
result). Additionally, Porter comments on the way the construction was used in the NT,
"Thus even the apparently stereotyped and LXX-dependent ἐν τῷ + Infinitive is better seen

92

Ibid., 340-41. Chapter three will examine the papyrus texts and chapter four will examine LXX
Genesis texts. These texts, it will be argued, use the syntactic construction in the same way to
pragmatically structure information. Therefore, Mandilaras's conclusion must be rejected.
93
Ibid., 341. This is true with regard to ἐν τῷ + infinitive; however, Moulton and Turner did not share
the same view about the Greek language in the LXX and NT. Moulton tended to have a more positive
view. See Stanley Porter, "History of Scholarship on the Language of the Septuagint," in Die Sprache
der Septuaginta / The Language of the Septuagint, vol. 3 of Handbuch zur Septuaginta, eds. Eberhard
Bons and Jan Joosten (Gütersloh: Gütersloher Verlagshaus, 2016), 16-22.
94
Stanley Porter, Verbal Aspect in the Greek of the New Testament, with Reference to Tense and
Mood, ed. D. A. Carson, Studies in Biblical Greek (New York: Peter Lang Publishing, 1989), 124-25.
95
He notes Mandilaras's papyri list in support of the temporal use! While the texts themselves do
demonstrate "the temporal use," Mandilaras – as mentioned above – does not come to the same
conclusion from those texts as Porter does. It is also debatable that the "temporal sense" became more
frequent in the PC era; my own examination of Classical texts has found no difference in usage from
either the perspective of information structure analysis or structural semantics.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

24

as an instance of Semitic enhancement, not interference" (emphasis added).96 Fourth, Porter
discusses Luke's usage of the construction, which accounts for 38 out of the 52 attestations in
the NT.97 Luke, he notes, has some of the most "polished" Greek in the NT. Therefore, it
would be unusual if he were simply "slavishly imitating Semitic style."98 Porter then
discusses how Luke often alters a syntactic pattern away from the Hebraic model; he gives
several illustrations from Luke-Acts. His conclusion, then, seems to be that the syntactic
construction – when used by itself – is natural Greek; however, when used with ἐγένετο,
Luke seems to be capturing the "flavor of the LXX (enhancement)" by writing in a biblical
style.99
2.2.8. Daniel Wallace (1996)100
Like several other scholars mentioned above, Wallace focuses his analysis on NT Greek.
He only discusses the way ἐν τῷ + infinitive is used in the NT without any indication of
whether or not it was conventional Greek or an example of Hebraic influence. His analysis is
very helpful, though, for understanding the way the construction was used in the NT from a
structuralist perspective. First, concerning prepositions in general, he states that they are,
"extended adverbs."101 Then, he goes on to say, "But, unlike adverbs, they govern a noun and
hence can give more information than a mere adverb can."102 When a preposition is
combined with an infinitive, the adverbial force is further strengthened.103 Second, he
discusses three senses that the construction signifies in the NT: result (1x in Heb. 3:12),
contemporaneous time, and means. He notably does not mention cause. Concerning time, the
construction answers the question "when," indicating a contemporaneous temporal
relationship between the syntactic construction's verbal action and its controlling verb
(predicator). This is fairly common in the NT. He also notes that when translating it into
English, the temporal force should be brought out by using "while" (for present infinitive)
96

Porter, Verbal Aspect in the Greek of the New Testament, 124.
See Appendix 1 for statistics.
98
Ibid., 125.
99
Ibid., 125. This conclusion seems to be accurate.
100
Daniel Wallace, Greek Grammar Beyond the Basics: An Exegetical Syntax of the New Testament
(Grand Rapids: Zondervan, 1996), 595-96.
101
Ibid., 356.
102
Ibid., 356.
103
Ibid., 589.
97

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

25

and "when" (for aorist infinitive) plus a verb.104 Concerning means, it describes how an
action is accomplished, answering the question "how."105 The NT contains several examples
(Acts 3:26; 4:30; Rom. 15:13; Eph. 6:17; Heb. 2:8, 8:13). Some of these examples, he notes,
might be referring to contemporaneous time.
2.2.9. Takamitsu Muraoka (2016)106
Takamitsu Muraoka's analysis is very brief for a syntactic construction that is used 534
times in the LXX. His discussion begins with a few verses that contain temporal semantics,
then proceeds to a verse that could be taken as a conditional, and finally discusses three
verses in which the construction indicates means. In connection with this discussion, he has a
footnote that states, "On the statistics of ἐν τῷ + inf. in various corpora, see Turner 145. This
syntagm is unknown to CG, most likely a Hebraism..." (emphasis added).107 The statement,
"This syntagm is unknown to CG," is ostensibly based on Turner's assessment that the
syntactic construction was "non-classical" (mentioned above). However, as stated above,
Turner most likely means that the construction was used differently in biblical Greek than it
was in CG. But Muraoka interprets Turner's words to indicate that it did not exist in CG,
which is mistaken. TLG contains more than 300 attestations of the syntactic construction
from Classical literary texts. Additionally, Muraoka labels the syntactic construction a
Hebraism without any textual data to substantiate his statement. Statements such as these that
are not based on textual data make it imperative to investigate the language of the LXX using
sound linguistic methodologies in the context of corpus linguistics.
2.3. Presuppositions about the Septuagint's Language
2.3.1. Hebrew-Priority vs. Greek-Priority
It is not surprising that divergent views about LXX syntax occur within a larger
discussion about the LXX's language. Although debate over the LXX's language began in the
16th century, Porter notes that the topic began to receive more comprehensive focus in

104

Ibid., 595.
Ibid., 598.
106
Takamitsu Muraoka, A Syntax of Septuagint Greek (Leuven: Peeters, 2016), 334-35.
107
Ibid., 334.
105

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

26

academic literature in the 19th century.108 The history of the debate and the names of the
scholars involved are beyond the scope of this thesis, although some will be mentioned.
Porter's essay entitled "History of Scholarship on the Language of the Septuagint" (cited
below) gives a helpful overview of the topic and the different scholars involved. Two
opposing perspectives developed early, and each perspective evaluated the textual data in the
LXX using different approaches. These two perspectives continue to persist presently in one
form or another. The first perspective approaches the data by giving greater significance to
the Hebrew source text, perhaps due to certain presuppositions about translation. This
perspective argues that Hebrew syntax and lexical usage determine meaning in the LXX.
This view has been labeled recently by William Ross as "Hebrew-priority."109 The second
perspective approaches the data with greater emphasis on the Greek language, arguing that
conventional Greek syntax and lexical usage determine meaning in the LXX. This view has
been labeled "Greek-priority."110 As mentioned above, presuppositions influence how
syntactic constructions such as ἐν τῷ + infinitive are interpreted. F. C. Conybeare and St.
George Stock's analysis of the LXX, for example, was greatly shaped by their presupposition
that "the vocabulary is Greek and the syntax is Hebrew."111 This presupposition affected the
way they read the text and explained its syntax. As Lee has insightfully noted, "What we
expect to see will have an impact on what we find when we try to analyze the Greek of the
Pentateuch and the translators’ methods."112 Scholars who think the LXX's syntax is Hebrew
will have a greater tendency to label constructions as Hebraisms. However, those who
maintain that the LXX is generally conventional Greek with intermittent, even regular,
Hebraic influence due to translation technique, exhibit a more cautious approach in using the

108

Porter, "History of Scholarship on the Language of the Septuagint," 16.
See Ross, Post-classical Greek and Septuagint Lexicography, 49. See especially pp. 40-62 for a
helpful discussion of the two perspectives.
110
Ibid., 52.
111
F. C. Conybeare and St. George Stock, Grammar of Septuagint Greek (Boston: Ginn, 1905), 41.
This is likewise the assessment in LEH: "The syntax of the Septuagint is Hebrew rather than Greek”
(Johan Lust, Erik Eynikel, and Katrin Hauspie, A Greek-English Lexicon of the Septuagint, viii-ix).
112
John A. L. Lee, "Back to the Question of Greek Idiom," in The Legacy of Soisalon-Soininen:
Towards a Syntax of Septuagint Greek, eds. Tuukka Kauhanen and Hanna Vanonen (Göttingen:
Vandenhoeck & Ruprecht, 2020), 24.
109

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

27

label Hebraism. This cautious approach takes for granted that LXX syntax is conventional
Greek until one finds data to the contrary.
2.3.2. Conventional Post-classical Greek
The first presupposition of this thesis is that LXX Genesis is a specimen of conventional
PC Greek. It has often been underestimated in this regard. However, according to Aitken and
Dhont, the LXX is the largest extant corpus of PC Greek prose, although it is not
significantly larger than some other corpora.113 The LXX, then, has much to teach scholars
about the way the Greek language was used during the PC era. Since Hebrew Genesis was
translated into Greek to be read by Greek speakers, the implication is that it was mostly
intelligible to those who knew Greek. The intelligibility and coherence of certain parts of the
LXX have perhaps been rightly questioned.114 Albert Pietersma, for example, is helpful to
insist that examples of unintelligibility and incoherence should not be "swept under the
rug."115 But scholars should never assume, no matter how profound their knowledge of AG,
that unintelligibility and incoherence in their estimation is indicative that the text was
unintelligible and incoherent to those who first received it. Unintelligibility must be
demonstrated rather than assumed. As a specimen of PC Greek, the LXX is "in essence
Greek with Hebrew interference, rather than Hebraic Greek into which idiomatic Greek
occasionally intrudes."116 Even most examples of Hebrew interference are intelligible,
making grammatical sense.
Affirming that LXX Genesis is a specimen of conventional PC Greek implies that there is
some diachronic connection with the Greek manifested in the Classical era. Although errors
have been made, especially in the 19th century, when comparing biblical Greek with CG,
there is a close relationship between the varieties of Greek spoken during the Classical and
PC eras. PC Greek did not arise in a vacuum. This means that scholars can often study

113

Aitken and Dhont, "The Septuagint within the History of Greek,” 444. The total word count of the
LXX is 623,782 compared to Polybius's Histories (316,866), Diodorus Siculus's historical work
(464,305), and Dionysius Halicarnassus’s history of Rome (415,573).
114
Albert Pietersma, "The Society of Biblical Literature Commentary on the Septuagint: Basic
Principles," in The SBL Commentary on the Septuagint: An Introduction, ed. Dirk Büchner (Atlanta:
SBL Press, 2017), 7.
115
Ibid., 7.
116
Lee, "Back to the Question of Greek Idiom," 24.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

28

Classical usage to shed light on conventional language usage during the subsequent era.
About the development of PC Greek during the Hellenistic era, Horrocks explains that the
language employed for the administration of the Greek territories after Alexander's death was
a developed form of Attic Greek.117 The Greek elites who administered these territories
passed their language on to the inhabitants of the territories. Horrocks states, "We should
therefore think of the PC Greek not simply as the written and spoken language of the GrecoMacedonian elite, but as a superordinate form of Greek standing at the pinnacle of a dialectal
pyramid of lower-register varieties that now evolved under its influence and ultimately owed
their identity to it."118 Although this statement only articulates part of the history of the Greek
language, a diachronic connection between CG and PC Greek is a key element of that story.
CG is upstream from PC Greek. Although the environment may be very different upstream, it
belongs to the same stream. Lee's principle is helpful in this regard: "As a general principle,
all evidence of Greek is potentially relevant and useful."119 Although the Greek in LXX
Genesis possesses many differences compared to CG, both are specimens of the same
language and therefore utilize the language in the same way in many regards.
However, affirming that LXX Genesis is conventional PC Greek also implies that there
has been diachronic development from the Classical era to the PC era. Like every language,
Greek experienced various changes over time. In addition to natural language change, there
were a variety of other factors that contributed to change. This is why anyone familiar with
CG literature will notice differences between Plato’s Greek and the documentary sources
from the Hellenistic era that occupy a middle to low register. But language change is less
apparent in PC literary texts since these texts occupied a similar register. These texts were
less affected by interference from other languages, as well. Language change is also less
apparent in administrative texts that occupy a middle register, the register to which the LXX
Pentateuch belongs. However, since there was still a great deal of development, the texts

117

Geoffrey Horrocks, "Phases of the Greek Language," in Die Sprache der Septuaginta / The
Language of the Septuagint, vol. 3 of Handbuch zur Septuaginta, eds. Eberhard Bons and Jan Joosten
(Gütersloh: Gütersloher Verlagshaus, 2016), 77.
118
Ibid., 77.
119
Lee, The Greek of the Pentateuch, 4-5. This statement does not ignore that diachronic language
change occurred. See the discussion below where it is argued that there was in fact language change
that occurred in the PC era and that textual data from that era is most relevant.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

29

most relevant for comparison with LXX Genesis come from the PC era. Lee has also noted
in this regard that "certain priorities and restrictions apply. The most obvious is that the
evidence closest in time to the LXX carries the greatest weight.”120
Furthermore, affirming that LXX Genesis is conventional PC Greek is a denial that the
Greek of the Pentateuch is a specimen of an Alexandrian Jewish dialect. Moisés Silva traces
the Jewish-Greek dialect view to Edwin Hatch, who published Essays in Biblical Greek in
1889.121 Hatch seemed irritated that many scholars in his day were critiquing biblical Greek
because it did not conform to Classical standards. Hatch pointed out that development had
taken place in Greek since Plato’s philosophical treatises and Herodotus’s historical
accounts, but he also argued that the cause for the distinctive nature of LXX Greek was that
it was written by Greek speakers of a different ethnicity. He argued that the LXX and the NT
"afford clear internal evidence that their writers, in most cases, were men whose thoughts
were cast in a Semitic and not in a Hellenic mould."122 This view has seemed plausible to
many.
There have been others, some of them prominent Greek scholars, who have been
convinced by this thesis and have consequently viewed LXX Greek as an exemplar of an
Alexandrian Jewish dialect that was vigorous in Egypt and Palestine. Despite the push-back
Hatch received, first by T. K. Abbott123 and then by H. A. A. Kennedy,124 others followed in
the 20th century with the same basic presupposition. Turner, who finished Moulton’s multi-

120

Ibid., 5. This principle is generally true but may not be absolute. There are literary elements that
made their way into PC Greek texts that have their origin in CG; this is due in part to the great
influence that the Attic dialect had on Alexander the Great and his administrators. Therefore, in order
to validate certain arguments about the register of particular lexemes and syntactic constructions, it
would be appropriate to use evidence from CG.
121
Moisés Silva, Biblical Words and Their Meaning: An Introduction to Lexical Semantics, Revised
and Expanded Edition (Grand Rapids, MI: Zondervan, 1994), 57.
122
Edwin Hatch, Essays in Biblical Greek (Oxford: Clarendon Press, 1889), 10.
123
T. K. Abbott, Essays, Chiefly on the Original Texts of the Old and New Testaments (London:
Longmans, Green & Co., 1891), 65-109.
124
H. A. A. Kennedy, Sources of New Testament Greek: The Influence of the Septuagint on the
Vocabulary of the New Testament (Edinburgh: T&T Clark, 1895). Kennedy perhaps mistakenly placed
both the LXX and NT in a colloquial register. This was no doubt due to the similarities he found in
the language employed in biblical Greek and the language of the papyri. Kennedy writes, "But as
regards the respective vocabularies, they are both children of the same parent, namely, the colloquial
Greek of the time. This is the secret of their striking resemblance" (146).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

30

volume work on New Testament Greek grammar and syntax, as mentioned above,
approached biblical Greek with the same basic presupposition.125 While Turner’s primary
focus was NT Greek, Princeton’s Henry Gehman made the same argument about LXX
Greek. But he argued that the dialect was temporary as Jews transitioned from speaking
Aramaic to become Hellenized.126
This view is plausible because biblical Greek is distinctive in many ways, both in its
syntax and in the usage of its lexical stock. But the distinctive character of LXX Greek is not
because there was an Alexandrian Jewish dialect before the LXX was translated, something
akin to a Jewish creole. There is no external evidence outside biblical Greek to support that
claim. Additionally, many of the “Semitisms” and “Hebraisms” in the LXX’s syntax and
vocabulary have been found in compositional Greek corpora. These facts were articulated by
Abbott and Kennedy (mentioned above), and most emphatically demonstrated by Adolph
Deissmann’s work comparing ancient papyri, inscriptions, and ostraca from the PC era with
the language of the LXX and NT.127 Deissmann summarized the main points of this
argument in an article published in Realencyklopädie für protestantische Theologie und
Kirche. He argued that scholars have overstated the distinctive character of LXX Greek.
After he had thoroughly studied the papyri, he considered them to be the richest examples of
the way non-literary Hellenistic Greek was written and spoken. It was Greek “triggered by
the thousand necessities and situations in the daily life of common people.”128 The papyri
include many administrative documents that were written in non-literary official language.
These papyri provide definitive evidence, Deissmann argued, that biblical Greek was a
specimen of conventional PC Greek. He acknowledged that the LXX and NT contain novel
terms and ways of using Greek vocabulary. However, the novelty is not related to the type of
language the Jews spoke, i.e., “history-of-language,” but to their religion, i.e., “history-of-

125

Porter, “Introduction,” The Language of the New Testament, 28-29.
Ibid., 29.
127
See Adolph Deissmann, Bibelstudien (Marburg: Elwert, 1895): idem, Neue Bibelstudien (Marburg:
Elwert, 1897); idem, Licht vom Osten (Tubingen: Mohr, 1908).
128
Adolph Deissmann, “Hellenistic Greek with Special Consideration of the Greek Bible,” in The
Language of the New Testament: Classic Essays, ed. Stanley Porter, Journal for the Study of the New
Testament Supplement Series 60 (Sheffield: Sheffield Academic, 1991), 52. This does not mean that
Deissman considered the language of biblical Greek to be colloquial.
126

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

31

religion.”129 Lee is one scholar who has built on Deissmann’s work. He has shown in two
monographs that “the language of the LXX translators was essentially the Greek of their
time.”130 The usage of ἐν τῷ + infinitive in PC Greek corpora will likewise provide evidence
for that same argument.
2.3.3. Middle-level Register with Literary Elements
The second presupposition is that LXX Genesis occupies a middle-level linguistic
register and contains some literary elements. According to M. A. K. Halliday, register can be
defined as a “functional variety of language.”131 Language is used in diverse ways according
to social situation: bedtime stories, communication with a cashier, an academic paper, a
sermon, and the President’s inaugural speech all require different language levels. People
face numerous social settings in which communication occurs, and those settings have
different expectations for language use. Syntax and lexical usage in a love letter, for
example, will be very different than those in a master's thesis. The lexical stock in any given
language will have a wide utility; the same is true for syntactic structures. However, certain
words and structures will be more register-specific, and therefore indicative of the register. If
those words and ways of speaking are used in a social context that is incongruous, they will
sound strange to native speakers.
The LXX contains a variety of registers. However, it is not always easy to determine the
register of specific books or isolated pericopes since register is a sociolinguistic reality.
Present scholars are far removed from the socio-cultural realities of the Jewish Hellenistic
world; therefore, they struggle to see clearly when it comes to making observations about
register since they lack the sociolinguistic intuitions that native speakers possess. The only
way to proceed is to identify elements that are indicative of a particular register, which can
be difficult, and search for those elements. Aitken and Dhont assert, "We are only in the early
stages of understanding register in the PC period, thus complicating any absolute or
generalizing statements made about the Septuagint."132 That may be true, but it has not
129

Ibid., 54.
John A. L. Lee, A Lexical Study of the Septuagint Version of the Pentateuch, Septuagint and
Cognate Studies 14 (Chico, CA: Scholars Press, 1983), 1.
131
M. A. K. Halliday, Halliday’s Introduction to Functional Grammar, 4th ed., revised by Christian
M. I. M. Matthiessen (Abingdon: Routledge, 2014), 29.
132
Aitken and Dhont, "The Septuagint with the History of Greek: An Introduction," 438.
130

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

32

stopped scholars (nor should it!) from attempting to identify various registers in the LXX.
Henry Thackeray classified entire books according to register at the beginning of the 20th
century.133 Though not denying the possibility of multiple registers even within a single
book, he classified books based on their overall register (what he called "style"): 1 Esdras,
Daniel, Esther, Job, Proverbs, Wisdom, Epistle of Jeremiah, Baruch (3:9-end), and 2, 3, 4
Maccabees are all literary or Atticistic.134 Furthermore, he stated that Tobit belongs to a
vernacular register and the Pentateuch is somewhere in between. He called this middle
register, “good κοινη Greek.”135 Scholars have not accepted every aspect of Thackeray's
conclusions, but most have found them generally helpful.136
Lee has reaffirmed Thackeray's assessment that generally the Pentateuch occupies a
middle register. Lee’s The Greek of the Pentateuch contains some description of the
Pentateuch's register, providing examples of different types of variation. After decades of
scholarly research on the language of the Pentateuch, he concludes that the Pentateuch
translators “in general adopted a middle-level Koine Greek of their time, moderately
educated but not literary and not colloquial or informal.”137 However, Lee nuances his
argument, comparing the Greek in the Pentateuch with the middle-level administrative
documents and personal correspondence in the Graeco-Egyptian papyri beginning in the third
century BCE. The register can be called “formal,” and it carries with it a certain dignity. He
writes, “[W]e can observe features of vocabulary that reflect what may be termed
‘officialese,’ that is, the official style of the third century BCE, used in administration and in
ordinary communication between educated Greek speakers.”138

133

Henry Thackeray, A Grammar of the Old Testament in Greek: According to the Septuagint
(Cambridge: Cambridge University, 1909), 12-16.
134
Ibid., 14. Atticistic refers to the Atticist movement that began in the 1st century BCE; it was a
literary movement that sought to imitate the Attic manner of speech during the Classical era,
ostensibly in a pretentious way. Lee also adds Sirach, Isaiah, and "probably Ecclesiastes." See Lee,
The Greek of the Pentateuch, 90, fn. 40.
135
Ibid., 14.
136
Evans, "The Grammarian Cannot Wait: Thackeray, Muraoka, and the Analysis of Septuagint
Syntax," 578. See also William Ross, "Some Problems with Talking about Septuagint Greek," Journal
for the Study of Judaism 53 (2022): 18.
137
Lee, The Greek of the Pentateuch, 63.
138
Ibid., 64.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

33

Some logical inferences can be made about the Pentateuch translators' choice of register.
First, they chose to translate the documents that governed the Jews’ civic139 and religious life
into the same register as the documents that governed the civic life of people in Ptolemaic
Egypt. The language level, then, would help signal its intended function to anyone who read
the Pentateuch. In this way, even apart from its content, the register was intended to engender
a certain response. Second, the translators wanted to use a register that was appropriate to the
dignity of the Pentateuch’s content, a difficult task if the vernacular language level was used.
Discussing "order words," Lee writes, "[T]he Pentateuch translators chose the officialsounding (but not literary) terms of their day, as being appropriate to the generally formal,
dignified character of the text they were translating."140 Third, the translators wanted the
Pentateuch to be clearly understood by their social and religious community. This would
have been more difficult if the Pentateuch had been entirely rendered in a literary register
that was only used by the most educated.141
Lee has also shown that the Pentateuch contains scattered literary elements in its
syntactic structures and lexical stock. These elements demonstrate the translators' education
because, as mentioned above, it was in the educational process in antiquity that students
came into contact with literary texts, especially Classical literary texts. It is not unusual, then,
that certain expressions would be taken up in an individual translator’s written idiolectal style
despite not being used in the vernacular. Regarding LXX Genesis, Lee identifies several
lexemes and phrases that belong to a literary register. The first is the rare syntactic
construction, attested in Genesis 37:35, that contains εἰς + genitive nominal: "εἰς ᾅδου." Lee

139

Joseph Mélèze Modrzejewski, "The Septuagint as Nomos: How the Torah Became a 'Civic Law'
for the Jews of Egypt," in Critical Studies in Ancient Law, Comparative Law and Legal History
Studies in Honour of Alan Watson, eds. John Cairns and Olivia Robinson (Oxford: Hart Publishing,
2001), 192.
140
Lee, The Greek of the Pentateuch, 65.
141
There were doubtless many well-educated Jews in Ptolemaic Egypt who, like Philo of Alexandria
during the Roman Egyptian era, were competent in literary Hellenistic Greek. The LXX books of
literary quality also demonstrate that reality. The point here is that the middle register was more
clearly understood by a broader range of Jewish people with diverse educational backgrounds. While
some have disputed the clarity and intelligibility of certain parts of the LXX, few argue that the
Pentateuch lacks intelligibility. (See Cameron Boyd-Taylor, "Toward an Analysis of Translation
Norms: A Sighting Shot," in BIOSCS 39 (2006): 27-46, for an example of a scholar questioning the
clarity of Deut. 19:16-21).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

34

refers to this construction as a "literary archaism," and cites several textual references in EG
(Homer, Od. 11.164) and CG (Sophocles, OT 1372; Aristophanes, Frogs 69; Euripides, HF
1101; Plato, Apol. 41a.1).142 The patterned expression of using the preposition εἰς with a
nominal in the genitive could only have been learned by reading literary texts, as it was not
conventional either in the Classical or PC eras. Additionally, certain literary particles found a
place in the text of the LXX. Lee discusses γε (Gen. 18:13; 26:9; 37:10),143 δή (Gen. 18:4;
27:34; 27:38),144 and μέν (Gen. 43:14; 44:8; 44:26).145 He then ends the chapter on
"Educated Language" by discussing some lexemes that belong to a literary register.146 These
are some linguistic elements that are particularly relevant to register analysis in LXX
Genesis. In the conclusion of this thesis, we shall return to the question of the register of ἐν
τῷ + infinitive. Nothing can be stated definitively at this point, but this construction was used
on occasion in documentary sources and used much more frequently in literary texts.
2.3.4. Translation with Hebraic Interference
One of the most obvious aspects of Septuagint studies is the fact that the Pentateuch,
indeed most of the LXX, was a translation from Hebrew. This fact must be carefully

142

Lee, The Greek of the Pentateuch, 91-92. His discussion of literary elements is in a chapter titled,
"Educated Language." Therefore, some of the discussion concerns linguistic features that are not
necessarily literary but require education to be proficient to use them. Listed above are only features
that are specifically identified as belonging to a literary register.
143
He writes, "The particle γε has all the signs of being a higher-register feature and stylistic flourish
that one learnt how to use. It is found often in literary texts of the post-Classical period..." (Ibid., 93).
The particle is used with ἆρα in the three texts cited above from Genesis. Lee refers to this as "old
and formulaic" (Ibid., 94; he cites LSJ).
144
Lee notes that this particle became a stereotyped rendering in later LXX books for the Hebrew
emphatic particle ‫ָנא‬. But this did not subtract from its literary quality. Therefore, it became a
"component of the literary Greek" of some LXX authors (Ibid., 96), as well as a stereotype. This is
one linguistic feature that demonstrates that common LXX renderings – even stereotypes – can
belong to a literary register. This question will be taken up with reference to ἐν τῷ + infinitive in the
conclusion.
145
The usage of this particle has no relationship to the Hebrew Vorlage but conforms to Greek
convention (Ibid., 98). Lee explains that the Classical usage was emphatic and that its common
function, attested in Gen. 43:14, was to mark off the pronoun not simply as the subject of the verb but
as introducing a change in topic from the previous sentences (Ibid., 99). This, in fact, is connected to
how the particle could function to encode a certain information structure.
146
Lee, The Greek of the Pentateuch, 111-20.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

35

considered since it had a profound effect on the LXX's text-linguistic character.147 This fact
does not conflict with the reality that the LXX Pentateuch is a 3rd-century BCE specimen of
PC Greek. But it is no ordinary specimen. Even Lee, whose scholarship has reinforced and
nuanced Deissmann’s and Thackeray's arguments that the Pentateuch is good Koine Greek,
has stated, “The language of the LXX is plainly not normal Greek in many places.”148
Though the Pentateuch translators' education and competence are apparent in the text, even
demonstrating excellent facility in using Greek idiomatic expression,149 the text still contains
Hebraisms. Lee and others have argued that many scholars have given an inordinate amount
of weight to Hebraisms. They have reiterated Deissmann’s hypothesis that certain syntactic
structures and lexemes, though once thought to show Hebraic interference, are found in the
Graeco-Egyptian papyri from the third century BCE. Trevor Evans, for example, writes,
"The number of 'Hebraisms' or (more vaguely) 'Semitisms' is much more limited than has
often been asserted."150 Even with that qualification, the Semitic flavor of the LXX
Pentateuch cannot be ignored.
It is, however, the Greek linguistic system that was used to translate the source text. So,
every syntactic structure should be considered an ordinary part of the Greek linguistic system
unless evidence can be offered to the contrary. Since it is Greek, it should be treated as
conventional Greek unless it can be shown that it was not. On the topic of LXX
lexicography, Emanuel Tov has stated that the "rule of thumb we follow is that as long as
possible we record the words of the LXX as if that text were a regular Greek text, explaining
the words – conjecturally – in the way which a Greek reader would have taken them."151
These comments are likewise relevant to LXX syntax. These comments are also, admittedly,
more concerned with the "text-as-received" than with the "text-as-produced." However, the
147

Cameron Boyd-Taylor, "In a Mirror, Dimly— Reading the Septuagint as a Document of Its
Times," in Septuagint Research: Issues and Challenges in the Study of the Greek Jewish Scriptures,
ed. Wolfgang Kraus and R. Glenn Wooden (Atlanta: SBL Press, 2006), 15-31, here 16-17.
148
Lee, A Lexical Study, 1.
149
Lee, The Greek of the Pentateuch, 123-72.
150
Evans, “The Nature of Septuagint Greek: Language and Lexicography,” in The Oxford Handbook
of the Septuagint, eds. Alison Salveson and Timothy Michael Law (Oxford: Oxford University, 2021),
96.
151
Emanuel Tov, "Greek Words and Hebrew Meanings," in Melbourne Symposium on Septuagint
Lexicography, ed. Takamitsu Muraoka (Atlanta: Scholars Press, 1990), 117. Cited in Ross, Postclassical Greek, 52.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

36

process of translation can never so neatly separate the two perspectives. When translators
produce a text, they never completely lose sight of how their renderings will be understood
by readers, especially since the translators themselves are its first readers. The fact that they
produced a text that possesses numerous examples of idiomatic Greek demonstrates that
point. Modern translation methodology is perhaps more concerned than the LXX translators
with characteristics like clarity and naturalness, so that a translation's anticipated reception
plays a role in the translation's production. However, it cannot be assumed that the LXX
translators were purely concerned with a quantitative, stereotyped production of their source
text without any concern for conventional Greek usage. The text shows otherwise.
Robert Hiebert has written extensively on LXX Genesis, both concerning its textlinguistic character as an ancient translation and concerning the translation principles that
governed his work translating LXX Genesis into English for the NETS project. His
conclusions about the Greek language in Genesis have come as the result of scrutinizing the
LXX's text in comparison with the MT to discern the "text-as-produced," that is, the rationale
behind each rendering when the translator produced the text. Methodologically, for Hiebert,
the underlying Hebrew Vorlage functions as the "arbiter of meaning"152 between possible
Greek meanings when the LXX is ambiguous. Also, his discussion on the Greek language in
LXX Genesis often concerns Hebraic interference. As the LXX provides many examples of
unconventional Greek, scholars and students of the LXX must identify what is conventional
and what demonstrates Hebraic influence. Hiebert, then, does an excellent job helping
scholars identify various ways in which the Hebrew linguistic system has influenced the
textual realities of LXX Genesis. He identifies evidence of Hebraic influence in "literalistic
translation"153 (cf., Genesis 11:10, Σημ υἱὸς ἑκατὸν ἐτῶν, "Sem was a son of one hundred
years;" he also discusses Genesis 14:15; 12:8; 19:29; 20:13), and "isolate translation"154 (cf.,
152

This phrase is taken from the introduction to the NETS and is also cited by Hiebert; see Albert
Pietersma and Benjamin Wright, "The New English Translation of the Septuagint (NETS)," BIOSCS
31 (1998): 26-30, here 27. For Robert Hiebert, see "Translation Technique in the Septuagint of
Genesis and Its Implications for the NETS Version," BIOSCS 33 (2000): 76-93, here 79.
153
Robert Hiebert, "Translation Technique in the Septuagint of Genesis and Its Implications for the
NETS Version," 80-82.
154
Ibid., 83. Hiebert, in part quoting Albert Pietersma, defines isolate translation as, "a rendering that
is based on the perceived meaning of an individual word 'in (virtual) semantic isolation' with
etymology playing a key role."

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

37

Genesis 7:4, καὶ ἐξαλείψω πᾶσαν τὴν ἐξανάστασιν, ἣν ἐποίησα, "and everything that rises
up, which I have made, I will wipe out"). Additionally, he identifies some other Hebraic
elements: the usage of ἀνὰ μέσον...ἀνὰ μέσον as a rendering for ‫ֵבּין‬...‫ֵבּין‬,155 parataxis,156 καὶ
ἐγένετο as a rendering for ‫ַו ְיִ֗הי‬,157 infinitive absolute + cognate finite verb,158 εἰ as a rendering
for ‫ ִאם‬in oaths,159 pleonasms,160 προστίθημι + infinitive as a rendering for ‫ יסף‬+ infinitive,161
ποιέω ἔλεος + prepositional phrase or dative as a rendering for ‫ִﬠם‬/‫ֲﬠֵשׂה־ֶחֶסד ֶאת‬,162 ἐκ χειρὸς as
a rendering for ‫ִמַיּד‬,163 ἐξιλάσομαι τὸ πρόσωπον as a rendering for ‫( כפר ָפָ֗ניו‬pi'el),164 χεῖλος as
a rendering for ‫ָשָׂפה‬,165 and finally Οὐλαμλοὺζ as a rendering for ‫אוָּ֛לם ֥לוּז‬.166
Since LXX Genesis provides many such examples where the Hebrew linguistic system
has influenced the text-linguistic character of the LXX, it is necessary to identify the
Hebraisms in the text and distinguish them from conventional Greek. This gives insight into
translation technique and also isolates the usage of conventional Greek in the LXX.
Therefore, to identify Hebraic interference in the LXX's syntax, Lee has argued that every
rendering belongs to one of three groups: “(a) the Greek matches the Hebrew but is not
natural Greek, and interference from Hebrew is certain; (b) the Greek matches the Hebrew,
but Greek and Hebrew syntax coincide, so the result may or may not be due to interference;
and (c) natural Greek is used contrary to the Hebrew, and interference is not possible.”167 ’Εν
τῷ + infinitive falls under the second group. It is well attested in texts before, during, and
after the period of the production of the LXX. The biggest question, then, concerns usage:

155

Robert Hiebert, "Linguistic Interference in Septuagint Genesis," JSCS 55 (2022): 55-73, here 5859. Cf. Genesis 1:4b.
156
Ibid., 59-62. This feature is ubiquitous.
157
Ibid., 62-63. This feature is also quite common.
158
Ibid., 63-65. Cf. Genesis 2:16; 16:10; 18:10; 44:15.
159
Ibid., 65-66. Cf. Genesis 14:22b-24
160
Ibid., 66-67. Pleonasm can involve the redundant use of pronouns or adverbs. Cf. Genesis 1:11;
10:14; 20:13.
161
Ibid., 67-68. Cf. Genesis 4:2, 12; 8:12, 21 (2x); 18:29; 37:8; 38:26; 44:23.
162
Ibid., 68-69. Cf. Genesis 24:12; 24:14; 24:49; 40:14.
163
Ibid., 69-70. Cf. Genesis 9:5.
164
Ibid., 70-71. Cf. Genesis 32:21 (20).
165
Ibid., 71. Cf. Genesis 11:1.
166
Ibid., 72. Cf. Genesis 28:19.
167
Lee, The Greek of the Pentateuch, 262. Lee reproduces the same schema for identifying Hebraic
interference in Lee, "Back to the Question of Greek Idiom," 17.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

38

how did it function and what were its semantics in the compositional texts compared to those
of the LXX? The PC compositional textual data in chapter three, compared with the LXX
Genesis data in chapter four, will demonstrate that Gen neither altered the construction’s
function nor its semantics. Therefore, there is no legitimate reason to attribute its usage in
LXX Genesis to Hebraic interference, regardless of the frequency with which it was used. In
the conclusion to this thesis, the issue of frequency will be addressed.
On the surface, ἐν τῷ + infinitive closely coincides with the Hebrew construction it most
often renders. But it possesses some key differences when examined structurally. Each time
ἐν τῷ + infinitive renders ‫ ְבּ‬+ infinitive, the LXX contains an addition. In Hebrew
morphological conventions, the Hebrew construction combines the preposition and infinitive
into one word; Greek has two separate words with the addition of an article. The Greek
preposition ἐν renders the prepositional morpheme ‫ְבּ‬, the article τῷ renders nothing,168 and
the Greek infinitive renders the Hebrew infinitive. Additionally, each time the Greek
construction is used as a translation equivalent, the translators have had to choose the most
appropriate tense. The Hebrew linguistic system has no tense for infinitives. The result is that
of the 23 times that the Greek construction appears in LXX Genesis (only 18 of which render
‫ ְבּ‬+ infinitive), the present tense is used 10 times and the aorist is used 13 times. Though the
employment of different verb tenses is not always significant, these two do convey different
verbal perspectives (i.e., verbal aspect) that could potentially affect semantics. Though ἐν τῷ
+ infinitive may structurally coincide with the Hebrew ‫ ְבּ‬+ infinitive construction to some
degree, the similarities are arguably only superficial. There is even less coincidence with the
other Hebrew constructions for which ἐν τῷ + infinitive is the counterpart. These issues will
be returned to in chapter five.
2.4. Linguistic Analysis of 'Εν Τῷ + Infinitive
The following discussion has to do with a description of the construction from two
different but complementary perspectives: structural semantic analysis and pragmatic IS
analysis. As mentioned in the introduction, IS analysis is based on a structural semantic

168

The article has a discourse pragmatic function that signals the construction's referent as
identifiable. See discussion below.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

39

analysis. Structural semantic analysis is fundamental and makes use of the metalanguage
often used by grammarians to describe how grammatical structures encode meaning. That
analysis will be followed by a pragmatic IS analysis to elucidate how the syntactic
construction was used in discourse to create coherence and cohesion. The IS analysis will
make use of the terminology outlined in the introduction.
2.4.1. Structural Semantic Analysis of 'Εν Τῷ + Infinitive
As mentioned in the previous section, the syntactic construction ἐν τῷ + infinitive is
composed of three grammatical constituents: the preposition ἐν, the article τῷ, and an
infinitive. In many texts, the infinitive possesses a subject and/or an object, both in the
accusative case. But neither are necessary: the infinitive's subject can be the same as its
predicator and the infinitive can be intransitive. The article's function will be discussed more
extensively below in section 2.4.2. It was noted above that Robertson views the article as
orthographically necessary. He argues that because only the majuscule script was used in AG
texts, the article was needed to distinguish between a prepositional phrase with an infinitive
and an infinitive with a prefixed preposition. Dennis Burk, though accepting Robertson's
argument, also maintains that there are two additional structural reasons for the article: "(1)
to mark the case of the infinitive and (2) to mark the infinitive as object of the
preposition."169 These might be legitimate reasons, but they seem somewhat suspect on the
surface, even if the infinitive should be viewed completely as a substantive.170 The only
obvious reason why an infinitive would need to be associated with a particular case is to
clarify its semantic relationship with its predicator when a preposition is used that takes more
than one case as its object. However, this would be superfluous with prepositions that only
take one case as their object. It is clear, as well, when an infinitive is the object of a
preposition in texts that have no article between the preposition and the infinitive. There is
no article because the prepositional phrases are generally encoding newly asserted

169

Dennis Ray Burk, “A Linguistic Analysis of the Articular Infinitive in New Testament Greek”
(Phd. diss., The Southern Baptist Theological Seminary, 2004), 109.
170
Recent research on PC articular infinitives indicates that "the articular infinitive retains its verbal
character in the papyri, and is thus best considered ‘verbo-nominal,’" rather than a substantive. See
Klaas Bentein, "Going Nominal: the Ancient Greek Articular Infinitive Between Syntax and
Context," in Subordination and Insubordination in Post-Classical Greek: From Syntax to Context, ed.
Klaas Bentein, Eleonora Cattafi, and Ezra la Roi, 1–41 (Berlin: De Gruyter, 2025), 1.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

40

information. For example, the prepositional phrase εἰς ὑμνεῖν καὶ αἰνεῖν (to sing hymns and
to praise) is found in 2 Esdras 22:24.171 The information communicated by the phrase is
newly asserted in its context. It will be argued below, based on IS analysis, that the article
possesses a certain discourse pragmatic function. The structuralist explanations provided by
Robertson and Burk should not be discounted or ignored because they do provide some
insight. However, pragmatic IS does perhaps provide greater explanatory power for the
article's use in the syntactic construction.
As noted above, Wallace states, "Prepositions are, in some respects, extended
adverbs."172 So, generally, ἐν τῷ + infinitive possesses an adverbial function, clarifying
something about its predicator. Structurally, the syntactic construction is most often an
adjunct (which is typical for prepositional phrases); however, there are some attestations of
the construction used as the complement of an equative verb.173 As an adjunct, it can occupy
a pre-verbal or post-verbal position without changing the sentence's meaning. Its position,
however, does affect its discourse function. Furthermore, as an adjunct, it can modify the
predicator in a nuclear clause (i.e., the sentence’s main clause) or the predicator in an
embedded clause. As both an adjunct and complement, ἐν τῷ + infinitive encodes a temporal
semantic relationship with its predicator's state of affairs (either temporal scope or temporal
cause).174 This specific temporal semantic relationship can be demonstrated in usages during
the Classical and PC eras.
As noted above, some grammarians have stated that ἐν τῷ + infinitive is used causally in
CG literature and that only language contact – in this case, contact with the Hebrew source
text – changed its semantics in the LXX to indicate temporality. It is unclear how this notion
developed, but it cannot be supported by textual data in the compositional Greek corpora. For
171

There is textual data that εἰς + infinitive was used with no article in CG (Philemon Comic.
Fragment 193-94, εἰς θἀμαρτάνειν), PC Greek (Choliambica Adespota, Anonymus in turpilucrum,
line 85, εἰς ζῆν) and LXX Greek (Judges 6:11, εἰς ἐκφυγεῖν; 2 Esdras 22:24, εἰς ὑμνεῖν).
172
Wallace, Greek Grammar beyond the Basics, 356.
173
For an example, see Nicomachean Ethics 1169b.30-31: εἰ δὲ τὸ εὐδαιμονεῖν ἐστὶν ἐν τῷ ζῆν καὶ
ἐνεργεῖν (“But if happiness consists in life and activity”). Translation taken from
Aristotle, Nicomachean Ethics, trans. H. Rackham, Loeb Classical Library 73 (Cambridge, MA:
Harvard University Press, 1926), 529. See also Polybius in §3.2.11 below.
174
In texts where ἐν τῷ + infinitive indicates cause, temporality is still not absent in the sense that
cause is a "dynamic temporal relationship." See Jacques Moeschler, "Causality, Lexicon, and
Discourse Meaning," Rivista di Linguistica, 15.2 (2003): 34.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

41

preposition + article + infinitive constructions, διὰ τὸ + infinitive and παρὰ τὸ + infinitive are
typically used to indicate cause. In contrast, the default usage of ἐν τῷ + infinitive is intended
to indicate the predicator's occasion in time (either temporal scope or temporal cause), the
precise value being dependent on the lexical semantics of the infinitive and its predicator.
Figure 2.4.1.1. Occasion in Time

Occasion in time describes the relationship that ἐν τῷ + infinitive grammaticalizes with
its predicator. More specifically, occasion in time can encode the semantic value of temporal
scope (a state of affairs that provides the temporal boundary for another state of affairs) or
temporal cause (a state of affairs that causes another state of affairs to occur in time). As the
dark area indicates in Figure 2.4.1.1, there is an overlap between the two specific semantic
values. This is the reason why, in some texts, an argument can be made for either. For
example, in the sentence: "When the sun set, the temperature dropped," the temporal
boundary for the temperature drop is when the sun set. However, the sunset also caused the
temperature to drop. The similarities can be further illustrated by the following diagrams.
Figure 2.4.1.2. Temporal Scope 1

This diagram illustrates temporal scope, or what has been labelled "limitative" by some
scholars. Coulter George, for example, writes concerning limitative adverbials that, "the
limitative expression of time sets a temporal boundary and confines the action to a narrower

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

42

point or set of points that lies within that boundary."175 In other words, the predicator's state
of affairs is contained in the temporal boundary of the state of affairs grammaticalized by ἐν
τῷ + infinitive. Hence, the predicator happens during ἐν τῷ + infinitive. In George's analysis
of various Greek corpora, "ἐν + D[ative] is typically limitative"176 and is the "default
limitative construction."177 Compare the above diagram with the following:
Figure 2.4.1.3. Temporal Cause 1

Like the previous diagram, this diagram also illustrates that the predicator is contained in
the temporal scope of the state of affairs grammaticalized by ἐν τῷ + infinitive. But the
predicator is also caused by ἐν τῷ + infinitive (symbolized by the

). This diagram, then,

would most clearly represent the example sentence above. Consider also the following text
from an Atticist grammarian during the PC era:
(1) Aelius Herodianus, Περὶ παθῶν 3.2.206.26
καὶ γὰρ ἡ Δημήτηρ γῆ ἐστι· ἡ δὲ γῆ διακόπτεται ἐν τῷ ἀροτριᾶσθαι.
For Demeter is the earth. And the earth is broken up when / because it is ploughed.
In the text, the construction ἐν τῷ + infinitive arguably gives both temporal scope and
cause, illustrating Figure 2.4.1.3. As a result, the construction ἐν τῷ ἀροτριᾶσθαι could be
translated by "when it is ploughed" or "because it is ploughed." Therefore, the predicator
διακόπτεται ("broken up") occurs not only within the temporal boundary of ploughing, but
also because of ploughing. This example also helps illustrate the importance of lexical
semantics and verbal semantics to determine the exact relationship between ἐν τῷ + infinitive
and its predicator. The lexeme that is translated "broken up" (διακόπτω) can refer to the
175

Coulter H. George, Expressions of Time in Ancient Greek (Cambridge: Cambridge University,
2014), 12.
176
Ibid., 11.
177
Ibid., 106. George also demonstrates that it can encode other semantic relationships, as well.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

43

activity of breaking up the earth or the state of being broken up. In AG, the state of being
broken up would typically be encoded using the perfect tense. Since the text above uses the
present tense, it is referring to the activity.178
If the perfect tense were used and it referred to the state of being "broken up," then Figure
2.4.1.3 would fail to capture the semantic relationship with ἐν τῷ + infinitive in text (1), since
the state of being "broken up" would continue after the state of affairs encoded by ἐν τῷ +
infinitive ceases. It would be better illustrated by the following diagram:
Figure 2.4.1.4. Temporal Cause 2

Figure 2.4.1.4 indicates that the predicator’s state of affairs begins within the temporal
boundary of the state of affairs grammaticalized by ἐν τῷ + infinitive and is caused by that
state of affairs (

). But its duration continues after the state of affairs grammaticalized by ἐν

τῷ + infinitive ceases. The perfect tense, then, would have indicated that the state of being
"broken up" began in the temporal boundary of ploughing but continues after the ploughing
ceases.
Figure 2.4.1.5. Temporal Scope 2

178

This demonstrates the interaction between verbal aspect and lexical aspect (sometimes referred to
as Aktionsart, actionality, or event type). Zeno Vendler's classification of lexical aspect has been
influential. See Zeno Vendler, "Verbs and Time," Philosophical Review 66 (1957): 143-60. See also
the discussion of actionality in Klaas Bentein, Verbal Perphrasis in Ancient Greek: Have– and Be–
Constructions (Oxford: University of Oxford, 2016), 40-45.

44

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

This type of semantic relationship is also quite similar to another type of temporal scope.
The only distinction between Figure 2.4.1.4 and Figure 2.4.1.5 is causation (

). The

semantic relationship that this diagram illustrates is that the predicator begins within the
temporal boundary of the state of affairs grammaticalized by ἐν τῷ + infinitive, but the
predicator continues after the state of affairs grammaticalized by ἐν τῷ + infinitive ceases.
Finally, there is one other type of temporal cause semantic relationship that can be visualized
using a diagram. There is no corresponding temporal scope semantic relationship for this
type of temporal causal event type as far as the usage of ἐν τῷ + infinitive is concerned.
Figure 2.4.1.6. Temporal Cause 3

This diagram illustrates that the state of affairs grammaticalized by ἐν τῷ + infinitive
causes the state of affairs of its predicator, but there is no overlap in time. The predicator
occurs in direct causation sequentially after the state of affairs grammaticalized by ἐν τῷ +
infinitive. The textual data provide some rare examples of this type of semantic relationship.
However, as mentioned above, there is no corresponding temporal scope relationship because
it would not rightly be "scope" but "sequence." For temporal sequence, the AG linguistic
system uses μετὰ τὸ + infinitive.179 However, it was still appropriate to use ἐν τῷ + infinitive
to indicate the type of causal relationship illustrated in Figure 2.4.1.6 because the cause itself
is contained in (ἐν) the state of affairs grammaticalized by ἐν τῷ + infinitive.
2.4.2. Information Structure Analysis of ’Εν Τῷ + Infinitive
The textual data indicate that ἐν τῷ + infinitive is prototypically presuppositional
information. In other words, the syntactic construction refers to information that the author
assumes is held in common with the reader or expects to be taken for granted. Therefore, the
state of affairs communicated by ἐν τῷ + infinitive is generally not newly asserted. It is
179

The LXX uses this syntactic construction 101x. An example from Genesis 13:14: Ὁ δὲ θεὸς εἶπεν
τῷ Ἀβρὰμ μετὰ τὸ διαχωρισθῆναι τὸν Λὼτ ἀπʼ αὐτοῦ (NETS: And God said to Abram after Lot had
separated from him).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

45

information that is already available to the reader in the text-internal world or the textexternal world. As an illustration, consider the following example from a CG literary text:
(2) Xenophon, Hellenica 1.6.4.4
ὅτι Λακεδαιμόνιοι μέγιστα παραπίπτοιεν ἐν τῷ διαλλάττειν τοὺς ναυάρχους
…that the Spartans had greatly erred when changing the admirals.180
In the text, Lysander had been forced to give up his post as the Spartan fleet's admiral;
the less experienced Callicratidas had replaced him. This change is reported in the sentences
that precede the citation above. So, ἐν τῷ διαλλάττειν communicates presuppositional
information that has only recently been activated in the discourse and is therefore still freshly
lit up in the reader's mind. By inserting the presuppositional information (which is
redundant), Xenophon provides given information that helps bring coherence to the text
because the newly asserted information ("greatly erred") is anchored ideationally to what has
already been activated.
Second, the textual data indicate that ἐν τῷ + infinitive is prototypically an identifiable
textual constituent. This is closely related to its presuppositional nature. However, it is
necessary to distinguish between the two categories because presuppositional information
concerns the ideas in the text, and identifiability concerns the cognitive recognition of textual
constituents that have been used to grammaticalize those ideas. To grammaticalize
identifiable information, ἐν τῷ + infinitive uses an article. This means that the syntactic
construction is marked as an identifiable textual constituent. Scholars have plausibly made
the argument that the Greek article's primary function is not to signal that a substantive is
definite, but that it is identifiable.181 There is a distinction between the concept of
definiteness and that of identifiability. Runge explains, "Definiteness is a binary grammatical
category; something is either definite or not. Identifiability is a cognitive category concerned

180

My own translation.
Evert Van Emde Boas, Albert Rijksbaron, Luuk Huitink, and Mathieu De Bakker, Cambridge
Grammar of Classical Greek (Cambridge: Cambridge University Press, 2019), 328-30. The
grammarians list a number of reasons why a referent might be identifiable: 1) previously mentioned,
2) "obvious from the context or made specific by the immediate context," 3) "generally well-known,"
4) "a species or class in its entirety," 5) "abstract concept," 6) name. They add the caveat regarding
reason number 6 that often when the article is not present, it is not easy to account for but depends on
the author's style (330).
181

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

46

with the degree of shared knowledge between a speaker and a hearer."182 This clarifies the
article's function in general and accounts for how the article is functioning in the construction
ἐν τῷ + infinitive. Stephen Levinsohn and Mark Dubis likewise state that "the presence of the
article with infinitival expressions...marks them as identifiable and instructs the recipients to
relate them to something in the prior context."183 In Xenophon's text above, the phrase ἐν τῷ
διαλλάττειν would be identifiable to anyone reading the text because the state of affairs it
grammaticalizes has just been activated in the discourse register. Readers would associate the
action expressed by the infinitive, marked as identifiable with an article, with what had
happened previously. Levinsohn's and Dubis’s statement underscores the general anaphoric
function of ἐν τῷ + infinitive. As mentioned above, the construction as a whole refers to a
constituent that has recently entered the discourse register in the text-internal world or the
text-external world.184 In the same way that a pronoun makes anaphoric reference to an
antecedent, ἐν τῷ + infinitive makes anaphoric reference to an antecedent that was either
explicitly activated in the text, implicitly activated as part of a cognitive schema, or readily
available to author and reader in the world outside the text.
Identifiable information can be further divided into various levels of activation status
within the discourse register. Identifiable information can be active (therefore having no need
to be accented or placed in a prominent position due to its recentness), accessible (may or
may not need to be accented or placed in a prominent position, but still recently activated), or
inactive (indicating a relatively greater need to be placed in a prominent or accented position
since many events with much information have been communicated since its activation).185
Inactive information could also be information that may not have been explicitly activated in
the previous discourse but is connected systemically to other active and accessible ideas in

182

Steven E. Runge, "Towards a Unified Understanding of the Greek Article from a Diachronic,
Cognitive Perspective," in The Article in Post-Classical Greek, ed. Daniel King (Dallas: SIL
International, 2019), 129.
183
Stephen H. Levinsohn and Mark Dubis, "The Use of the Greek Article in 1 Peter: A Case Study,"
in The Article in Post-Classical Greek, ed. Daniel King (Dallas: SIL International, 2019), 120.
184
This is generally true about all cases of a preposition + article + infinitive.
185
Lambrecht, Information Structure and Sentence Form, 106-8. Lambrecht states that active
information could be accented or placed in a prominent position, but there is no such need (106).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

47

the discourse register. Prototypically, ἐν τῷ + infinitive pragmatically encodes identifiable
information that is active within the discourse register.
Third, the textual data indicate that ἐν τῷ + infinitive prototypically communicates
something about the topic. This construction can either be a part of the topic expression or it
can give background information about the topic.186 Stated negatively, ἐν τῷ + infinitive is
normally not a part of an utterance's focus expression. The topic, as a whole, concerns the
“aboutness” of an utterance;187 the focus is what is being asserted concerning the topic.
Therefore, as a textual constituent that is presuppositional and identifiable, ἐν τῷ + infinitive
contributes relevant information about the topic. Using the text from Xenophon cited above,
the topic would be something like when the Spartans changed admirals. The focus would be
on what is asserted about that topic: they erred greatly.
As mentioned previously, ἐν τῷ + infinitive gives important temporal background
information. In the text above, it is used to localize the Spartans’ error to a point in time:
when they changed admirals (ἐν τῷ διαλλάττειν τοὺς ναυάρχους). Cognitive linguistics has
emphasized that the Greek preposition ἐν "refers to the basic notion of containment."188
Using this insight, then, it can be stated that every textual utterance using ἐν τῷ + infinitive
shows that this syntactic construction expresses the occasion in time for its predicator's state
of affairs or, rarely, expresses the occasion in time for a noun it modifies.189 This occasion in
time can be described as temporal scope (i.e., time within which something occurs) or
temporal cause (i.e., one action/state in time that causes something else). Whether it contains
temporal scope or temporal cause is dependent upon the semantics of the lexemes in the
utterance.

186

For a discussion of this terminology, see Matić, "Topic, Focus, and Discourse Structure: Ancient
Greek Word Order," 579.
187
Ibid., 15.
188
Richard A. Rhodes, "Greek Prepositions: A Cognitive Linguistic View," in Postclassical Greek
Prepositions and Conceptual Metaphor, eds. William A. Ross and Steven E. Runge (Berlin: Walter de
Gruyter, 2022), 12-13.
189
In chapter three, one text will illustrate the fact that the syntactic construction can modify a noun
phrase. However, the noun phrase has an implicit, presupposed predicate and argument. See §3.2.5.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

48

CHAPTER 3: ANALYSIS OF POST-CLASSICAL COMPOSITIONAL TEXTS
"To study the Greek of the LXX 'as Greek' we need to know what that Greek is."190
– John A. L. Lee
3.1. Introduction
Early PC Greek, often referred to as Hellenistic Koine Greek,191 is a developed form of
the Attic dialect.192 Due to historical circumstances following the Persian War and cultural
developments in Attica, Attic Greek enjoyed prestigious status among the Greek dialects.
This was the reason it was "adopted by the all-conquering Macedonians and then carried
throughout the East as an administrative and cultural language by the campaigns of
Alexander the Great."193 This does not mean that idiosyncratic features of other Greek
dialects, such as the Ionic dialect, did not find their way into the Koine, nor that there were
no developments in the language's lexical stock.194 But it does mean that, by and large, PC
Greek can trace its lexical usage, morphology, and syntax back to Attica. However, as Chrys
Caragounis has noted, Koine Greek was a form of the language that replaced the nuance and
complexity of its predecessor with a greater degree of simplicity and explicitness.195 As the
language of an empire, spoken not only by educated native speakers but also by those using it
as a second language, Koine Greek also contained different registers. The term Koine refers
to the language commonly spoken throughout an entire empire, namely the Greek language
of the Hellenistic era, rather than a common dialect or register of the language, i.e., low-level
colloquial language. PC Greek usage of various levels is found in literary texts,
administrative documents, legal petitions, personal letters, and extant texts more indicative of
everyday communication. This means that the Koine, like most languages, was used in
various formal and informal ways.

190

Lee, The Greek of the Pentateuch, 4.
PC Greek is dated to roughly 330 BCE - 600 CE while Hellenistic Koine Greek falls within the
range of 330 BCE - 300 CE.
192
Horrocks, Greek: A History of the Language and Its Speakers, 4.
193
Ibid., 4.
194
Chrys C. Caragounis, The Development of Greek and the New Testament: Morphology, Syntax,
Phonology, and Textual Transmission (Grand Rapids: Baker Academic, 2006), 39.
195
Ibid., 39.
191

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

49

As noted above, Lee has demonstrated that the LXX generally contains middle-level PC
Greek, occupying the same register as the thousands of Graeco-Egyptian administrative and
legal documents that have been preserved in the sands of Egypt.196 Therefore, the
documentary papyri from the PC era provide the best texts to illuminate the usage of
conventional Greek to compare to the Greek of the LXX. However, it has also been shown
that the LXX translators could use "more educated expressions of literary quality."197
Therefore, when necessary, the LXX must also be compared to literary texts to help
illuminate its language. The literary texts were still used pedagogically during the era when
the Pentateuch was translated. They functioned as "schoolroom texts" in the education of
Greek speakers.198 So, any discussion of the translators must take into consideration that
literary texts from the Classical era were the basis for their education. Hence, the syntactic
patterns and lexical usage in the literary texts no doubt left an imprint, whether conscious or
not, on the educated class in Ptolemaic Egypt. The following analysis of Greek texts, then,
will examine both literary texts and non-literary documentary sources. The literary texts
provide an abundance of attestations of ἐν τῷ + infinitive (see table 3.1 below), which raises
questions concerning the register to which the syntactic construction belongs. However, to
limit the discussion to a manageable set of texts, this chapter will only examine ten texts
from Polybius. As for documentary sources, they only occasionally attest to this syntactic
construction. Therefore, all ten examples from these sources will be discussed below. In the
discussion of each text, it will be necessary to clearly articulate the broader context in which
the syntactic construction is found. This is necessary because the key to understanding how
ἐν τῷ + infinitive was used is often within the discourse context and not simply within the
sentence (although sometimes the construction refers anaphorically to a constituent in the
same sentence). So, the discussion associated with each text will take into consideration as
much or as little of the context needed to illuminate the IS.

196

Lee, The Greek of the Pentateuch, 64.
Ross, Postclassical Greek and Septuagint Lexicography, 42.
198
Boyd-Taylor, "In a Mirror, Dimly— Reading the Septuagint as a Document of Its Times," 26-27.
See also Lee regarding the translators' education in The Greek of the Pentateuch, 120. Lee states,
"Reading Classical texts was part of the training, and that is revealed clearly in the evidence." See
also Horrocks, Greek: A History of the Language and Its Speakers, 4-5.
197

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

50

Table 3.1: 'Εν Τῷ + Infinitive in PC Literary Authors before CE
Period of time199
IV-III BCE
IV-II BCE
IV-I BCE
IV BCE-II CE
III BCE
III-II BCE
III-I BCE
II BCE

II-I BCE
II BCE - II CE
II BCE - III CE
II BCE - IV CE
I BCE
I BCE - I CE
Total

Author (Number of Attestations)
Dinarchus (1x), Theophrastus (21x), Epicurus (6x), Menander (7x),
Cleanthes (1x), Clearchus (2x), Duris (2x), Hecataeus (5x), Persaeus
Phil. (1x), Demetrius Phalereus (1x), Timaeus Hist. (1x), Damoxenus
Comic. (1x)
Charondas (1x)
Dioscurides (1x)
Diotogenes (1x)
Erasistratus (2x), Lycon (2x), Chrysippus (47x), Euryphamus (3x),
Metopus (5x), Straton (2x), Hieronymus Hist. (1x), Bryson (1x),
Callicratidas (1x), Theages (12x)
Polybius (24x), Aristophanes Gramm. (4x), Heraclides Criticus (1x),
Philo Mechanicus (2x), Demetrius Gramm. (2x), Anonymus Photii
(3x), Timaeus Phil. (1x), Aristocles Paradox. (1x)
Agatharchides Geogr. (2x), Ezechiel (1x), Apollodorus Gramm. (1x),
Antipater (2x), Demetrius Lacon (1x), Diogenes Phil. (2x), Anonymus
Epicureus (1x), Heracleodorus Gramm. (1x), Comanus (3x), Dionysius
Scytobrachion (1x)
Posidonius Phil. (6x), Antiochus Phil. (1x)
Ptolemaeus Gramm. (2x)
Testamenta XII Patriarcharum (7x)
Diogenis Sinopensis Epistulae (4x)
Diodorus Siculus (36x), Dionysius Halicarnassensis (31x), Arius
Didymus (13x), Tryphon I Gramm. (1x), Philodemus (35x),200
Philoxenus Gramm. (4x), Socrates (1x), Philo Judaeus (53x)
Strabo (3x), Aristonicus Gramm. (5x), Dorotheus (1x), PseudoArchytas (9x), Antiochus Astrol. (1x), Vitae Adam et Evae (4x)
62 authors (397x)
3.2. Post-classical Literary Texts

From the beginning of the PC era (ca. 330 BCE) until the Common Era, ἐν τῷ + infinitive
was used 397 times by 62 authors in extant compositional texts as documented in Thesaurus
Linguae Graecae. In this period, the corpus of literature in TLG contains 12,694,837 words
(excluding the LXX). Therefore, this syntactic construction occurs 0.03 times per 1,000

199

The period of time in the left column of Table 3.1 indicates either that the lifetimes of the authors
straddled the centuries mentioned (in the case of two successive centuries), or that their precise
lifespans is unknown.
200
The total number of usages in Philodemus includes one usage not found in TLG's database but
found in the Herculaneum papyri on papyri.info (P.Herc. 26).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

51

words in the corpus as a whole. However, some authors employed it much more frequently
than others, such as in the c. 2nd century BCE pseudepigraphal document On the Virtues,
attributed to the Pythagorean philosopher Theages, who did so more than ten times per 1,000
words.201 The issue of frequency, which is often used as an argument for Hebraic interference
in LXX studies, must take into consideration that even in compositional texts, there is a great
amount of variation from one author to the next. Conventional Greek does not exhibit any
consistency in the frequency with which certain syntactic constructions are used. Frequency,
then, is an issue of idiolect.202
Polybius's The Histories attests twenty-four occurrences of ἐν τῷ + infinitive. Polybius
was not the first to write prose during the PC era, but his historiography is a good exemplar
of prose during that era. About him, Horrocks writes, "Polybius is in general very much a
man of his times, both in his choice of vocabulary...and in his overall style, which exhibits
the typical verbosity of the Hellenistic chancellery, most particularly in the complex sentence
constructions which make characteristically heavy use of nominalized infinitives as an
instrument of subordination."203 For these reasons, especially because of his use of infinitives
for grammatical subordination, Polybius can provide insight into the prototypical function of
the construction in the PC Greek linguistic system.
3.2.1. Polybius, The Histories 1.23.8
τὸ δὲ λοιπὸν πλῆθος τῶν Καρχηδονίων ἐποιεῖτο μὲν τὸν ἐπίπλουν ὡς εἰς ἐμβολήν, ἐν
δὲ τῷ συνεγγίζειν θεωροῦντες τὸ συμβεβηκὸς περὶ τὰς προπλεούσας ναῦς, ἐξέκλινον
καὶ διένευον τὰς τῶν ὀργάνων ἐπιβολάς.204
The rest of the Carthaginian force was making the advance as if for ramming. But in
the approach,205 seeing what had happened to the leading ships, they veered off and
dodged the attacks of the war engines.

201

This is far more frequent than LXX Genesis's 0.71x per 1,000 words. There were 14 other authors
of compositional texts that used the construction with a greater frequency than LXX Genesis.
202
See Appendix 1 for frequencies in PC Greek authors/texts.
203
Horrocks, Greek: A History of the Language and Its Speakers, 97.
204
Polybius, The Histories, Volume I: Books 1-2, trans. W. R. Paton, rev. F. W. Walbank, Christian
Habicht, Loeb Classical Library 128 (Cambridge, MA: Harvard University Press, 2010), 70.
205
To reflect the fact that ἐν τῷ + infinitive is an identifiable constituent, English renderings will –
even awkwardly – use "in" to translate the preposition and the direct article "the." All translations are
my own.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

52

Context:206 This text narrates a naval battle during the Punic War in 260 BCE.207 Under
Hannibal Gisco's command, the Carthaginians had previously destroyed a Roman fleet under
the command of Gnaeus, as they possessed greater experience and swifter vessels.
Subsequently, the Roman commander Gaius Duilius took command of the Roman navy, and
he constructed novel war machines (afterwards referred to as "ravens"). When Gnaeus sailed
to meet the Carthaginians, they attacked him eagerly and confidently due to the superiority of
their navy. But they were unaware of the Romans' novel war engines; as a result, the first
thirty Carthaginian ships were taken. This was the first sea battle won by the Romans in the
Punic Wars.
SS and IS:208 Structurally, the constituent ἐν τῷ συνεγγίζειν (in the approach) is an
adjunct that modifies two finite verbs: ἐξέκλινον (they veered off) and διένευον (dodged). It
provides the occasion in time (temporal scope) for these predicators.209 Therefore, it was in
the temporal boundary of their approach that the Carthaginians veered off and dodged. The
construction refers anaphorically to the prior sentence: "The rest of the Carthaginian force
was making the advance as if for ramming" (ὡς εἰς ἐμβολήν). So, ἐν τῷ συνεγγίζειν should
be viewed as presuppositional since advance for ramming (which necessarily involves an
approach) has just been activated in the discourse register. The information has been
grammaticalized with a constituent marked for identifiability. The syntactic construction
belongs to the sentence's topic domain.210
3.2.2. Polybius, The Histories 1.51.9
διεκπλεῖν μὲν οὖν διὰ τῶν πολεμίων νεῶν καὶ κατόπιν ἐπιφαίνεσθαι τοῖς ἤδη πρὸς
ἑτέρους διαμαχομένοις, ὅπερ ἐν τῷ ναυμαχεῖν ἐστι πρακτικώτατον, ἀδυνάτως εἶχον,
διά τε τὴν βαρύτητα τῶν πλοίων, προσέτι δὲ καὶ τὴν ἀπειρίαν τῶν πληρωμάτων.211
206

In the analysis of each text, the "Context" section will give background information (often textexternal) and text-internal contextual information necessary for making sense of the IS.
207
Polybius, The Histories, Volume I, 63.
208
The analysis of SS and IS has been combined since IS is built upon syntactic structure and
structural semantics.
209
See Félix-Marie Abel, Grammaire du Grec Biblique: Suivie d'un Choix de Papyrus (Paris:
Lecoffre-Gabalda, 1927), 314. Abel cites this text as an example of the temporal usage. He writes,
"l'hébreu en particulier pour les locutions où entre καὶ ἐγένετο, mais l'usage de ce ἐν temporel n'est
pas absent de la Koine. Cf. Polyb. I, 28, 8." Credit belongs to Dirk Büchner for finding this reference.
210
Sentences can possess any number of constituents in the topic domain with various levels of
pragmatic salience.
211
Polybius, The Histories, Volume I, 156.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

53

So, on the one hand, to sail through the enemy ships and then appear behind those
already fighting against others, which in fact is most effective in the waging of naval
warfare, was impossible both because of the heaviness of the ships and also the
inexperience of the crews.
Context: This text likewise concerns a naval battle narrative between the Romans and
Carthaginians during the First Punic War in 249 BCE.212 Under Publius Claudius Pulcher,
the Roman fleet had attacked the Carthaginians at Drepana, Sicily, and had made a blunder.
The ships came too far into the harbor before forming a line. Thus, they had unintentionally
allowed the Carthaginians to sail past them out into the open sea. The ships of the
Carthaginian fleet, under the command of Adherbal, were then able to form their line against
the Romans with their back toward the open sea; this gave them significantly more space to
maneuver. This text, then, is a description of the Roman fleet; the particular maneuver it
describes was not only impossible for the Romans in this battle but was never accomplished
by them in any battle.213
SS and IS: Structurally, ἐν τῷ ναυμαχεῖν is a complement to ἐστι (to be) within a
dependent, explanatory clause. It provides the occasion in time (temporal scope) for ἐστι, its
predicator. The utterance can be perceived as gnomic, so it does not refer to an actual event
during the naval battle. But it still indicates that within the temporal boundary of naval
warfare, gnomically speaking, the most effective maneuver is to sail through enemy ships
engaged in battle and appear behind them. The entire context is concerned with narrating a
naval battle, so ἐν τῷ ναυμαχεῖν encodes presuppositional information that is marked for
identifiability. Likewise, it belongs to the topic domain of the clause in which it is located.
3.2.3. Polybius, The Histories 1.62.4
μέχρι μὲν γὰρ ἐκ τῶν κατὰ λόγον ἦν τις ἐλπὶς ἐν τοῖς ὑποκειμένοις, οὐδὲν τῶν
παραβόλων ἢ δεινῶν δοκούντων εἶναι παρέλιπεν, ἀλλὰ πάσας τὰς τοῦ νικᾶν ἐν τῷ
πολεμεῖν ἐλπίδας, εἰ καί τις ἄλλος ἡγεμόνων, ἐξήλεγξεν.214
For as long as there was some reasonable basis for hope in the situation, he left out
nothing, however seemingly dangerous or dire. But as much as any other general, in
the waging of war, he tested every hope of victory.
212

Ibid., 151.
F. W. Walbank, A Historical Commentary on Polybius, vol. 1 (Oxford: Clarendon Press, 1957),
114. Walbank says that this was, "a classic maneuver, but never employed by the Romans. Indeed, in
Hellenistic times it tended to disappear with the adoption of boarding tactics, though the Rhodians
used it successfully at Chios."
214
Polybius, The Histories, Volume I, 186.
213

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

54

Context: In 241 BCE, during the First Punic War, the Carthaginians had suffered another
naval defeat at the hands of the Romans. Fifty ships were sunk and seventy others captured,
along with their crews (numbering around 10,000 men). Hanno had led these ships in an
attempt to resupply Hamilcar and the troops at Eryx. The text cited above describes
Hamilcar. He was a feared and respected general. As the text indicates, his general pattern in
warfare was to try any possible way to win as long as there was still a reasonable chance of
victory.
SS and IS: Structurally, ἐν τῷ πολεμεῖν (in the waging of war) is an adjunct that modifies
the finite verb ἐξήλεγξεν (he tested). It denotes the occasion in time (temporal scope) for
ἐξήλεγξεν, its predicator. The predicate is asserting Hamilcar's habitus in warfare. So, it is
not referring to a single event. Nevertheless, the phrase ἐν τῷ πολεμεῖν still indicates that
within the temporal boundary of warfare, generally speaking, Hamilcar tested (ἐξήλεγξεν)
every hope of victory. Furthermore, similar to the text in §3.2.2, since the entire context of
the prepositional phrase is warfare, the phrase ἐν τῷ πολεμεῖν (in the waging of war) encodes
presuppositional information that is marked for identifiability. Additionally, the construction
belongs to the topic/background domain.215
3.2.4. Polybius, The Histories 2.29.3
τὸ δὲ μέγιστον ἀποκεκλειμένης πάσης τῆς εἰς τοὔπισθεν ἀναχωρήσεως καὶ τῆς ἐν τῷ
λείπεσθαι σωτηρίας. ἡ γὰρ τῆς ἀμφιστόμου τάξεως ἰδιότης τοιαύτην ἔχει τὴν
χρείαν.216
But the most important thing, they were cut off completely from retreat towards the
rear and safety in the [case of] coming up short;217 for this peculiar position of
facing both ways had this kind of disadvantage.218

215

Lisa Brunetti, Kordula De Kuthy, Arndt Riester define the topic as "an expression denoting a
special referential entity within the background, potentially the most salient one." See "The
Information-Structural Status of Adjuncts: A Question-under-Discussion-Based Approach," Discours
28 (2021), 6. In this thesis, topic domain refers to the grammatical topic (i.e., topic constituent) and
other constituents that provide background information about the topic constituent.
216
Polybius, The Histories, Volume I, 342.
217
The translation, "coming up short," is an attempt to render λείπω idiomatically.
218
My own translation interprets what is denoted by ἐν τῷ λείπεσθαι in a manner consistent with that
of W. R. Paton in the Loeb Classical Library volume: "while, above all, they were absolutely cut off
from retreat or any prospect of escape in the case of defeat, this being the peculiarity of this two-faced
formation." Paton's translation, like the one above, adds the words "case of" to help clarify the
conditional nature of what is communicated in the context.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

55

Context: Polybius details a battle that occurred during the Celtic219 invasion of Italy in
225 BCE.220 The Celts simultaneously fought two separate Roman military units at Telamon
after inflicting heavy losses on other Romans and taking a large quantity of plunder,
prisoners, cattle, and other spoils of war. In The Histories 2.26, Polybius records a decision
made by one of the Celtic kings: knowing their good fortune might come to an end and
satisfied with their haul, he decided it would be best to return to Gaul. However, their march
away from one pursuing Roman unit led them into another oncoming unit. They were
trapped between these two units and were forced to fight the Romans on two fronts, resulting
in a precarious strategic position. Although the position had the advantage of not allowing
any surprise attack behind them, it also cut them off from any ability to retreat if they were to
come up short in the battle.
SS and IS: The sentence that contains ἐν τῷ + infinitive in this passage is structurally and
semantically complex. The sentence’s core predication (nuclear predicate) is not a finite verb,
but a genitive absolute participle ἀποκεκλειμένης (cut off). This means that the participle
clause is structurally independent but semantically dependent on another clause with a finite
verb (the following clause with the finite verb ἔχει).221 ’Εν τῷ λείπεσθαι functions as an
adjunct that modifies the predicate and one of its arguments, ἀποκεκλειμένης πάσης... τῆς
σωτηρίας (cut off completely...from safety). The construction is nested in the constituent τῆς
σωτηρίας, indicating a close syntactic relationship. Semantically, the construction provides
the occasion in time (conditional temporal scope) for ἀποκεκλειμένης πάσης... τῆς σωτηρίας.
In other words, if the Celts were not victorious [i.e., "coming up short"], they would be
completely cut off from a path to safety within the temporal boundary of that defeat.
In the discourse context, Polybius discusses the battle (μάχη) between the Celts and the
Romans. When readers engage with any report of a battle, such as this account, they

219

Polybius used Κελτοί to refer to the Gauls and other ethnic groups in Western Europe. This was
typical of Greek authors. However, the Romans more specifically used the Latin term Galli to refer to
the Gauls and Celtae to refer to Celts. Polybius's usage has been retained in the comments in this
thesis.
220
Polybius, The Histories, Volume I, 327.
221
The genitive absolute clause does not have an explicit subject. Therefore, the subject is the same as
the clause before it resulting in the rendering using the third person plural pronoun "they." See
Levinsohn, Discourse Features of New Testament Greek, 136.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

56

generally take for granted that there were winners and losers, those who were killed and
those who survived. This pragmatic phenomenon is called a cognitive schema. The
constituent elements – denotata like winners, losers, casualties, and survivors – are made
available222 in the discourse register in conjunction with the discussion of war. Charles
Fillmore explains a cognitive schema (or frame-linked referent): "By the term 'frame' I have
in mind any system of concepts related in such a way that to understand any of them you
have to understand the whole structure in which it fits; when one of the things in such a
structure is introduced into a text, or a conversation, all of the others are automatically made
available."223 This is the reason why ἐν τῷ λείπεσθαι was encoded as a textual constituent
marked for identifiability. Polybius expects his readers to take it for granted as
presuppositional information in the cognitive schema of war. The construction provides
background information within the topic domain. Topic Domain = [marked topic: the most
important thing224 + temporal background: in the case of coming up short]; focus = [THEY
WERE CUT OFF COMPLETELY FROM RETREAT AND SAFETY.]

3.2.5. Polybius, The Histories 2.32.10
τοὺς δὲ τῶν Κελτῶν σφίσι συνόντας διαβιβάσαντες εἰς τὸ πέραν ἀνέσπασαν τὰς ἐπὶ
τοῦ ῥείθρου γεφύρας, ἅμα μὲν ἀσφαλιζόμενοι τὰ πρὸς ἐκείνους, ἅμα δὲ μίαν ἑαυτοῖς
ἀπολείποντες ἐλπίδα τῆς σωτηρίας τὴν ἐν τῷ νικᾶν, διὰ τὸ κατόπιν αὐτοῖς ἄβατον
ὄντα παρακεῖσθαι τὸν προειρημένον ποταμόν.225
But sending those of the Celts with them to the other side, they pulled up226 the
bridges on the river, both as a safeguard against those [Celts] and at the same time
leaving for themselves one hope of escape, that is, in the [achievement of] victory,
because the said river lying at their backs was not fordable.

222

Being made "available" indicates that the constituent would be identifiable as a result of another
concept being activated in the discourse register. To illustrate this point, Lambrecht uses the sentence,
"Every time I go to the clinic, the doctor is someone different." The definite article is used with doctor
because it is an identifiable textual constituent due to the word "clinic" being activated in the
discourse register. When people hear "clinic," automatically doctor is made available.
223
Charles J. Fillmore, “Frame semantics,” in Linguistics in the Morning Calm, ed. Linguistics
Society of Korea (Seoul: Hanshin Publishing Co., 1982), 111. Cited from Lambrecht, Information
Structure and Sentence Form, 90.
224
The word order pragmatically marks it as the topic. For a discussion, see Dik, Word Order in
Greek Tragic Dialogue, 38.
225
Polybius, The Histories, Volume I, 350, 352.
226
Ostensibly, they removed the wooden boards (τὰς σανίδας) on the bridges like they did in The
Histories 2.5.5.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

57

Context: In The Histories 2.32, Polybius narrates the exploits of a Roman expeditionary
force led by Publius Furius and Gaius Flaminius in 223 BCE. The Romans were attempting
to completely remove the Celts from the Po Plain. The Cenomani, a tribe that had become
their ally, accompanied the Romans to fight at the foot of the Alps against another tribe, the
Insubres. However, being outnumbered, the Romans were wary of fighting with the
Cenomani as their ally against the Insubres because both tribes came from the same nation.
The Celts were notoriously treacherous, as well. So, the Romans sent the Cenomani across a
river and pulled up the wooden boards on the bridges to prevent them from possibly turning
against them in the heat of battle.
SS and IS: Structurally, ἐν τῷ νικᾶν functions slightly differently than in any previous
text. While ἐν τῷ νικᾶν seems to be in apposition to μίαν…ἐλπίδα τῆς σωτηρίας (encoded
using the article τὴν,227 which precedes ἐν τῷ νικᾶν), the phrase is also functioning as an
adjunct (i.e., it can be moved to different positions in its clause without affecting meaning).
Therefore, it is best to view this example as an adjunct that modifies not only the noun phrase
to which it is in apposition, but the entire clause with its predicate and arguments: μίαν
ἑαυτοῖς ἀπολείποντες ἐλπίδα τῆς σωτηρίας (leaving for themselves one hope of escape).
Concerning abstract nouns such as ἐλπίς and σωτηρία, Civilleri notes that, "Such nouns are
semantically complex because their semantic-pragmatic function is not to refer to objects
(‘reference,' which is normally a property of nouns) but to predicate something."228
Therefore, the entire clause is lexically dense with words that predicate some state of affairs:
leaving, hope, escape. Semantically, it could be stated that the Carthaginians are leaving
(PRESENT) hope in victory; they have hope (PRESENT) for victory; and they have hope to
escape (FUTURE) because of victory. The semantic relationship between the entire clause and
ἐν τῷ νικᾶν is conceptually complex. It could be described as contingent temporal cause:
escape will be achieved, hope will be realized, and leaving hope will cease if and when
victory is achieved.

227

The article is a grammaticalized textual anaphora. See Cristina Guardiano, "The History of Greek
Articles: A Syntactic Approach," in The Article in Post-Classical Greek, ed. Daniel King (Dallas: SIL
International, 2019), 62.
228
Germana Olga Civilleri, "Abstract Nouns," in Encyclopedia of Ancient Greek Language and
Linguistics, vol. 1, ed. Georgios K. Giannakis, 4-7 (Leiden: Brill, 2014), 4.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

58

As in previous texts, Polybius reports a military battle in The Histories 2.32. Also similar
to §3.2.4, ἐν τῷ νικᾶν is marked for identifiability because the idea of victory belongs to the
cognitive schema of warfare. Reading any narrative concerning warfare would implicitly
make available certain ideas in the discourse register (cf. fn 225, 226). Therefore, the
constituent ἐν τῷ νικᾶν grammaticalizes information that ought to be taken for granted (i.e.,
presuppositional information). Discerning the topic structure in this text is a bit more difficult
since the sentence structure does not fall neatly into the topic/focus schema articulated by
Dik. It can be concluded reasonably that ἐν τῷ νικᾶν is a non-obligatory element of the
sentence's structure and provides background information, which generally belongs to the
topic domain.
3.2.6. Polybius, The Histories 3.79.10
τῶν δ᾿ ὑποζυγίων αὐτοῦ τὰ πλεῖστα πίπτοντα διὰ τοὺς πηλοὺς ἀπώλλυντο, μίαν
παρεχόμενα χρείαν ἐν τῷ πεσεῖν τοῖς ἀνθρώποις· καθεζόμενοι γὰρ ἐπ᾿ αὐτῶν καὶ τῶν
σκευῶν σωρηδὸν ὑπὲρ τὸ ὑγρὸν ὑπερεῖχον, καὶ τῷ τοιούτῳ τρόπῳ βραχὺ μέρος τῆς
νυκτὸς ἀπεκοιμῶντο.229
Falling because of the mud, the majority of the Celts' [αὐτοῦ] pack animals perished,
serving one use for the men in their fall [i.e., when they fell]: sitting on them and the
heaped-up baggage, the men were above the water and in such a manner were able to
get some sleep for a short part of the night.
Context: Hannibal led his troops south of Florence through a marshland between the
Arno and the Apennines during the Second Punic War in 217 BCE. His army consisted of
soldiers from various regions conquered by the Carthaginians. Leading the march through the
wetlands were the Iberians and Libyans, who had experience in long, difficult marches. Due
to their experience and the fact that they were the first to pass over the muddy ground, they
were able to pass through the marshland with only a little misery. However, the Celts
following behind them experienced overwhelming difficulty because of their lack of
experience on long marches, and the wet ground had already been thoroughly broken up by
the Iberians and Libyans.
SS and IS: Structurally, ἐν τῷ πεσεῖν is an adjunct that modifies the participial phrase
μίαν παρεχόμενα χρείαν (serving one use). Semantically, it clarifies the occasion in time for

229

Polybius, The Histories, Volume II: Books 3-4, trans. W. R. Paton, Loeb Classical Library 137
(Cambridge, MA: Harvard University Press, 2010), 214.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

59

the pack animals to serve a specific use. So, in the temporal boundary of falling, they
provided a use. ’Εν τῷ πεσεῖν refers anaphorically to a textual referent in the sentence:
πίπτοντα. The participle lit-up "falling" in the reader's consciousness, so ἐν τῷ πεσεῖν is
presuppositional – encoded using a textual constituent marked for identifiability – and a part
of the topic domain in its clause.
3.2.7. Polybius, The Histories 4.64.7
τῶν δὲ πειθαρχούντων, ἅμα τῷ τὴν πρώτην διαβῆναι σημαίαν βραχέα ταύτης
καταπειράσαντες οἱ τῶν Αἰτωλῶν ἱππεῖς, ἐν τῷ ταύτην τε μεῖναι συνασπίσασαν καὶ
τὴν δευτέραν καὶ τρίτην διαβαινούσας συμφράττειν τοῖς ὅπλοις πρὸς τὴν
ὑφεστῶσαν.230
After his [Philip] orders were obeyed, as soon as the first company crossed, the
Aetolian cavalry briefly skirmished with them; while the first company [ταύτην]
stood fast fighting with shields interlocked, the second and third companies also
crossed to close ranks with the hoplites facing resistance.
Context: In the text, Polybius narrates an incident in the Aetolian War (220-217 BCE) in
which Philip V of Macedon led the Hellenic league against the Aetolians, Elians, and
Spartans. The text itself describes a military maneuver by Philip's soldiers. The Aetolian
cavalry attempted to attack their company after they had just crossed a river. The Aetolians
were unsuccessful and forced to retreat. The soldiers who were attacked managed to hold
their ground and even reinforced their position with additional companies, causing the
attackers to withdraw.
SS and IS: Structurally, ἐν τῷ ταύτην τε μεῖναι συνασπίσασαν (while the first company
stood fast fighting with shields interlocked) is an adjunct that modifies the participle
διαβαινούσας (crossed). It gives the occasion in time (temporal scope) for the participle, its
predicator. In other words, it was within the temporal boundary of the first company standing
fast that the second and third companies crossed. The construction refers anaphorically to the
previous sentence, which states, "The king, perceiving their design [i.e., attacking them while
they were crossing the river], ordered his peltasts to enter the river first and land on the other
bank in close order shield to shield [συνησπικότας] and company by company."231 The
sentence explains that Philip had ordered his troops to cross the river in close formation with

230
231

Ibid., 494.
Ibid., 495.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

60

their shields interlocked because he knew they would be attacked by the Aetolian cavalry as
soon as they reached the other side. Therefore, they "stood fast" in that formation after being
attacked on the other side. So, like other attestations of ἐν τῷ + infinitive, the syntactic
construction is presuppositional information that is grammaticalized using a textual
constituent marked for identifiability.
3.2.8. Polybius, The Histories 5.52.8
οἳ τὸ μὲν πρῶτον συνεπλέκοντο καὶ κατεπείραζον ἀλλήλων, ἐν δὲ τῷ συνάψαι τὰς
παρ᾿ ἀμφοῖν δυνάμεις ἀπέστησαν.232
At first, they were engaged in combat and were testing each other. But when the two
main armies from both sides rejoined them, they separated.
Context: This text is set in 221 BCE during Antiochus III's advance through
Mesopotamia to stop a rebellion led by a Mede named Molon. The text itself concerns a
skirmish between the advance guard of Antiochus's army and the advance guard of those
rebelling under Molon.
SS and IS: In this text, structurally, ἐν τῷ + infinitive is a part of an adjunctive clause
that modifies the finite verb ἀπέστησαν (they separated) in the main clause (nuclear clause).
It provides the occasion in time (temporal cause) for the predicator (see Figure 2.4.1.3). In
the previous sentence, Polybius referred to the light infantry of both armies that had been sent
as an advance guard: "He [Molon] was approaching the district in question at the very time
that the king [Philip III] had left Apollonia with the whole of his army, and the light infantry
of both, which had been sent on in advance, came into contact in crossing a certain range of
hills."233 It is implicitly understood in warfare that if an advance guard is sent ahead of an
army, the army will rejoin the advance guard at some point in time. Again, the anaphoric
reference made by ἐν τῷ συνάψαι is to information that is a part of a cognitive schema that
was activated in the discourse register through the assertion that the light infantry of both
armies had been sent in advance. The information is, therefore, presuppositional and is
grammaticalized using a textual constituent marked for identifiability. Additionally, it
belongs to the sentence’s topic domain.

232

Polybius, The Histories, Volume III: Books 5-8, trans. W. R. Paton, Loeb Classical Library 138
(Cambridge, MA: Harvard University Press, 2011), 138.
233
Ibid., 139.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

61

3.2.9. Polybius, The Histories 6.42.2
οἱ μὲν γὰρ Ἕλληνες ἐν τῷ στρατοπεδεύειν ἡγοῦνται κυριώτατον τὸ κατακολουθεῖν
ταῖς ἐξ αὐτῶν τῶν τόπων ὀχυρότησιν234
For the Greeks, when setting up camp, primarily considered adapting it to the
strengths of its location.
Context: This example is in a section that makes general observations about the Romans
and the Greeks. In The Histories 6.41-42, Polybius described how the Romans and Greeks
set up their camps in different ways. The Romans always set up camp in the same manner,
with each person in the company in the same location relative to others. However, the Greeks
set up camp differently, as the above text indicates.
SS and IS: Structurally, ἐν τῷ στρατοπεδεύειν is an adjunct that gives the occasion in
time (temporal scope) for its predicator, ἡγοῦνται (consider). Therefore, semantically, it is in
the temporal boundary (see Figure 2.4.1.2) in which they are setting up camp that the Greeks
primarily consider adapting the camp to the strengths of the topography. The syntactic
construction ἐν τῷ στρατοπεδεύειν makes anaphoric reference to a text-internal constituent in
the previous two sentences. These sentences read, "It is very much the same thing in a
Roman camp [στρατοπεδείας]. The Romans, by thus studying convenience in this matter,
pursue, it seems to me, a course diametrically opposite to that usual among the Greeks.”235
Therefore, ἐν τῷ στρατοπεδεύειν communicates presuppositional information that was
grammaticalized using a grammatical constituent marked for identifiability. In this text, the
syntactic construction forms part of the topical expression (topic = the Greeks setting up
camp).
3.2.10. Polybius, The Histories 21.4.5
εἰ δὲ μὴ συνυπακούοιεν, πάντως διειλήφει παραλιπὼν ταῦτα διαβαίνειν εἰς τὴν
Ἀσίαν, σαφῶς γινώσκων διότι τὸ τέλος ἐστὶ τοῦ πολέμου καὶ τῆς ὅλης ἐπιβολῆς οὐκ
ἐν τῷ χειρώσασθαι τὸ τῶν Αἰτωλῶν ἔθνος, ἀλλ᾿ ἐν τῷ νικήσαντας τὸν Ἀντίοχον
κρατῆσαι τῆς Ἀσίας.236
"But if they [Aetolians] did not submit, he [Lucius Cornelius Scipio] was determined
in any case to leave these matters aside and cross over into Asia, knowing well that
the goal of the war and the entire campaign was not in subjugating the Aetolian
League, but in conquering Asia by defeating Antiochus."
234

Polybius, The Histories, Volume III, 404.
Ibid., 405.
236
Polybius, The Histories, Volume V: Books 16-27, trans. W. R. Paton, Loeb Classical Library 160
(Cambridge, MA: Harvard University Press, 2012), 278.
235

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

62

Context: This text is set in 190 BCE during Roman territorial expansion to occupy lands
that were formerly ruled by Greeks during the Roman-Seleucid War. The Antiochus
mentioned is Antiochus III the Great.
SS and IS: Structurally, the two syntactic constructions are complements. Even though
they occupy a different structural position than what is typical, they still semantically indicate
the occasion in time for a single predicator, ἐστὶ. The syntactic constructions communicate
what was not the temporal cause and what was the temporal cause for reaching “the goal of
the war” (see Figure 2.4.1.3). The first example tells what the goal is not: οὐκ ἐν τῷ
χειρώσασθαι τὸ τῶν Αἰτωλῶν ἔθνος (not in subjugating the Aetolian League). The second
example explains what the goal is: ἐν τῷ…κρατῆσαι τῆς Ἀσίας (in conquering Asia).
Semantically, this relationship might be described as contingent temporal realization. In other
words, the goal of the war (from the Roman perspective) would only be realized if/when they
conquered Asia.
As far as IS is concerned, the construction encodes presuppositional information.
Although neither use of ἐν τῷ + infinitive has a direct textual antecedent, the context in its
entirety describes how the Romans were waging war against the Aetolians and Seleucids in
an attempt to gain dominance in the Ancient Mediterranean. Therefore, like several texts
cited above, the two uses of ἐν τῷ + infinitive refer to a cognitive schema that has remained
active throughout the discourse concerning Roman conquests. This is the reason the
construction was encoded using a textual constituent marked for identifiability.
3.3. Post-classical Papyri
The following discussion will examine six texts with eight examples of ἐν τῷ + infinitive
in papyri. These are the only six texts that have been found containing the syntactic
construction in the papyri.237 This is quite infrequent when the total number of papyri is
considered. Even though the papyri do not provide an abundance of examples, they are
sufficient to demonstrate how the construction was used in conventional PC Greek.
237

See Appendix 1 for a comparison of the frequency in papyri with that in other texts from the PC
era. As far as the papyri are concerned, the construction was used approximately 0.01 times per 1,000
words before the CE. This is much lower in frequency than the 0.71x per 1,000 words in LXX
Genesis.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

63

3.3.1. PSI. 4.354, lines 9-11 (254 BCE)
καλῶς οὖν ποιήσεις ἀποστείλας τὰς ἡμιόνους ὅπως συναχθῇ εἰς Μοιθῦμιν ἐν τάχι, ἵνα
μὴ ἐν τῶι παραπορεύεσθαι τὸν βασιλέα διαφορηθῇ ὑπὸ τῆς δυνάμεως.
Therefore, you will do well by sending the mules so that it [hay] might be taken
quickly to Meidum so that, in the king’s passing, it will not be carried off by the
troops.238
Context: This text is an excerpt of a letter written in Philadelphia, Egypt, by Boubalos.
He is writing to Zenon to discuss logistical details about Ptolemy's upcoming visit to
Philadelphia. To prepare the place of assembly (ἀγορά), Boubalos had moved bundles of hay
from the grounds where the assembly would gather to the side of the main road (ἐστὶν δὲ ἐν
παρόδωι). He asked Zenon to send mules to haul away the grass to Meidum so that when the
king passed on his way to the assembly, the grass would not be carried off by the troops
accompanying the king. The troops were opportunists and would snatch up free fodder for
their livestock any chance they got.
SS and IS: Structurally, ἐν τῶι παραπορεύεσθαι is an adjunct in a dependent, purpose
clause that modifies its predicator, the finite verb διαφορηθῇ. Semantically, the construction
provides the occasion in time for its predicator. This relationship is best viewed as temporal
scope, the king passing providing the temporal boundary.
’Εν τῶι παραπορεύεσθαι τὸν βασιλέα was encoded as a textual constituent marked for
identifiability. The construction makes direct anaphoric reference to what had been stated at
the beginning of Boubalos's letter: the king was coming (παραγείνεσθαι τὸν βασιλέα) and he
would be going to the place of assembly (ἀγορὰν), arriving via the main thoroughfare.
Therefore, it was implicit that he had to pass where the grass bundles were stacked. Using ἐν
τῷ + infinitive, Boubalos grammaticalizes information that would be presuppositional to
Zenon. The clause's topic expression is "in the king passing;" the focus239 is on what is
communicated by the predicate: grass being carried off (διαφορηθῇ). The information
encoded by ἐν τῶι παραπορεύεσθαι τὸν βασιλέα was redundant. However, its usage helped to
provide coherence to the communication with Zenon, anchoring asserted information to
presuppositional information.
238

My own translation.
Lambrecht refers to this type of focus as predicate-focus structure. See Lambrecht, Information
Structure and Sentence Form, 222ff.
239

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

64

3.3.2. P.Cair.Zen. 3.59379, lines 1-3 (ca. 254-51 BCE)
Ζήνωνι χαίρειν Ἀμεννεὺς ὑοφορβός. ἐν τῶι συνταγῆναι ἡμῖν ἀναθρέψαι τὰ εἰς τὰ
Ἀρσινόεια ἱερεῖα θεὶς οὖν τὸ ἱμάτιον ἐνέχυρον ἐπριάμην καὶ ἀνέθρεψα, ὅπως μηθέν
μοι ἐγ̣καλῆις.
To Zenon: greetings. Amenneus, a swineherd. When it was ordered for us to raise
pigs for the Arsinoe festival sacrifices, having therefore a cloak as a pledge, I
bought240 pigs and raised them, so that you might not find fault with me.
Context: This text is also one of the documents belonging to the Zenon archives. It has
approximately the same date as the text in §3.3.1. After Amenneus, the pig herder, had
obeyed Zenon’s instructions, buying pigs to raise for an upcoming festival, two of the pigs
had been stolen. When the thief was confronted, he claimed that a crocodile had eaten the
pigs. So, Amenneus petitions Zenon to write letters to the village and the nomarch,
Timotheos, asking that the issue be resolved.
SS and IS: Structurally, ἐν τῷ συνταγῆναι is an adjunct on the left periphery, framing the
clauses that follow. The construction modifies the two finite verbs that are predicators,
ἐπριάμην and ἀνέθρεψα. Semantically, the relationship between them is causal, i.e., "because
it was ordered," however, a temporal semantic value is not completely absent. There was a
temporal sequence between the construction and its predicators (see Figure 2.4.1.6). It was
appropriate to use ἐν τῷ + infinitive to indicate the type of temporal causal relationship
because the cause itself is contained in (ἐν) the state of affairs grammaticalized by ἐν τῷ
συνταγῆναι. Zenon's order was the occasion (in time) for Amenneus to buy and raise pigs.
The syntactic construction ἐν τῷ συνταγῆναι was used at the beginning of the letter. So, it
could not possibly be identifiable because of the text-internal world. It is identifiable due to
Zenon’s and Amenneus's previous communication, something external to the text above. It is
not specified whether the communication was a previous letter or face-to-face oral
communication. Either way, Zenon's instruction to Amenneus was still active in their
conversational discourse register. So, the idea denoted by ἐν τῷ συνταγῆναι is
presuppositional, common ground information. The construction ἐν τῷ συνταγῆναι was used

240

The form is imperfect, but LSJ, 1463, states that it was used as an aorist in Attic Greek. Since the
Greek in Ptolemaic Egypt would have been influenced by Attic Greek, then it is perhaps best viewed
as aorist. It also best fits the context.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

65

in the topic domain; the focus is the assertion that buying and raising pigs had occurred, so
that Zenon would not find fault. This is a predicate-focus structure.
3.3.3. P. Koeln. 8.327 (250 - 150 BCE)
(1) Column 1.31-2.3
τὴν δὲ θερμασίαν ὀξεῖαν εἶναι εὔλογον πυρετοῦ συμπεσόντος, ἐπειδὴ ὁ σφυγμὸς ἐκ
τῆς θερμασίας τῆς ἐν τῶι πυρέσσειν γινομένης ἐστὶν̣ μείζων τοῦ κατὰ φύσιν.
It is reasonable that the heat is acute when fever occurs, since the pulse, from the heat
generated during the fever, is greater than what is natural.
Context: This papyrus was found in Herakleopolis, Egypt. It is an unknown medical
treatise that was cut in a half-circle and placed on the chest of a mummy. The content of the
papyrus discusses causes of fevers and related symptoms, which have led some to suspect
that it was "a lost treatise Περὶ καυσωδῶν πυρετῶν or, more simply, Περὶ πυρετῶν."241 It is
perhaps debatable whether or not the papyrus is a PC documentary text, since it may have
been written by the Greek physician Erasistratus (c. 304 – c. 250 BCE), and belongs to the
genre of medical prose. The text contains three examples of ἐν τῷ + infinitive. The first
example is found in a sentence that begins in column 1.31 and ends in column 2.3 (these
lines are all fragmentary, therefore they have been reconstructed). The second and third
examples are in more complete sections of the papyrus, though they contain some
fragmentation.242
241

https://papyri.info/dclp/59762; accessed 16 September 2024. The quotation is taken from the
papyrus's introduction.
242
My own translation of column 1.8–3.16 (which contain all three uses of ἐν τῷ + infinitive): "And
also, something is always carried along from the veins in the urine, and they fall into a coma, and
since they do not retain urine in the bladder, they are constantly unable to urinate, and if they do
urinate, they are unaware of it. Similarly, there are contractions around the stomach. Additionally,
their mouths become phlegmatic, and because certain colors from the veins are carried to the mouth,
and a black coating appears on their tongues, it is not unreasonable. In cases of burning fevers, it is
like erysipelas around the liver, and in this condition, the heat generated by the infection is very
intense. It is reasonable that the heat is acute when fever occurs, since the pulse, from the heat
generated during the fever, is greater than what is natural. The pulse becomes more frequent, and
there is also extreme dryness around the mouth, the tongue, and the skin all over the body in these
fevers, since heat naturally causes dryness. And because the parts around the lungs and heart become
more contracted and the dilation of the lungs increases and becomes more frequent, breathing also
becomes more intense and frequent. Then, thirst occurs and in the heating [of] the internal places, it is
also reasonable for them to become thirsty and desirous of a cold drink. When inflammation is
present, it continuously causes sensations similar to those perceived by the senses, and because of the
intense heat, a sense of perception follows, such that they become sleepless, and they become thin and

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

66

SS and IS: Structurally, ἐν τῶι πυρέσσειν is an adjunct within another prepositional
phrase. The construction is predicated on the adjectival participle τῆς γινομένης.
Semantically, ἐν τῶι πυρέσσειν communicates the occasion in time (temporal cause) for the
generation of heat. It was redundant to use, but its inclusion adds textual cohesion and
ideational coherence. The noun phrase "burning fevers" (καυσώδεσιν πυρετοῖς) was attested
previously in column 1.25 without the article. Therefore, it was encoded using an
unidentifiable textual constituent because the author was activating the discussion of fevers
in the discourse register.243 Thus, when ἐν τῶι πυρέσσειν is used in column 2.1, it refers
anaphorically to the noun phrase "burning fevers" in column 1.25, as well as to the prior
mention of fever in the same sentence (πυρετοῦ). The author had encoded presuppositional
information as a part of the sentence's topic (fever/heat).244
(2) Column 2.23-29
γινομένης δὲ δίψης καὶ ἐν τῶι θερμαίνεσθαι τοὺς ἐπιθυμητικοὺς τόπους,245 εὔλογον
καὶ δίψας αὐτοῖς γίνεσθαι καὶ ψυχροῦ ποτοῦ [drink] τὴν ἐπιθυμίαν·
Then, thirst occurs and in the heating [of] the internal places, it is also reasonable for
them to become thirsty and desirous of a cold drink.
SS and IS: Structurally, the construction ἐν τῶι θερμαίνεσθαι modifies the finite verb
εὔλογον as its predicator. Semantically, it communicates the occasion in time (temporal
cause) for its predicator. In other words, within the temporal boundary of the internal places
heating, it is reasonable to be thirsty (see Figure 2.4.1.3). Before this text, θερμασία (heat)
and its cognates were used 4x, and πυρετός (fever) and its cognates were also used 4x. It is

restless. This happens especially in cases of moist fevers, since the moisture in the veins is neither
properly mixed nor evenly heated. Shivering is not unreasonable, for when the heat escapes some of
the fluids will be cooled, and when these fluids fall into the arteries, they cause shivering."
243
Levinsohn and Dubis, "The Use of the Greek Article in 1 Peter: A Case Study," 101-126.
Levinsohn and Dubis are careful to explain that the absence of the article does not always indicate the
constituent is unidentifiable. The article is often not employed on identifiable constituents to indicate
pragmatic salience. The usage of πυρετοῦ in text (1) above is an example. So, while the presence of
the article does indicate +identifiability, its absence does not indicate –identifiability.
244
See Matić, "Topic, Focus, and Discourse Structure," 579. He uses topic to refer to the
denotatum/referent, topic expression to refer to a particular word or phrase within the utterance that
encodes the topic, and topic position to refer to the constituent that occupies the topic's slot in
information structure.
245
Τοὺς ἐπιθυμητικοὺς τόπους refer to the places where desires are located (see LSJ 634). The author,
then, is connecting heat in the seat of desire with the desire (τὴν ἐπιθυμίαν) for a cold drink. This is a
"reasonable" (εὔλογον) conclusion in his mind.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

67

safe to conclude, also, that the topic of the entire medical treatise is fevers or body heat.
Therefore, the topic is presuppositional, and it is active throughout the entire text. The
construction in 2.25 (ἐν τῶι θερμαίνεσθαι) grammaticalizes a textual constituent that would,
therefore, be cognitively identifiable to anyone reading the text, and it is marked as such
using the article.
(3) Column 3.13-16
ἐγ γὰρ τῶι τὴν θερμασίαν διεκπίπτειν ἔσονταί τινα τῶν ὑγρῶν κατεψυγμένα
For in the escaping [of] the heat, some of the fluids will be cooled.
SS and IS: Structurally, ἐγ τῶι τὴν θερμασίαν διεκπίπτειν modifies the paraphrastic
construction ἔσονταί κατεψυγμένα (will be cooled). Semantically, it communicates the
occasion in time for some of the bodily fluid cooling. In other words, it is within the temporal
boundary (scope) of escaping heat that fluids cool. Additionally, the context indicates that
escaping heat also causes the fluids to cool (see Figure 2.4.1.3). "Heat escaping" would be
active in the discourse register as a part of the cognitive schema of fevers and body heat.
Furthermore, readers' ordinary experience in the text-external world would establish the
presupposition that any time something is heated, the heat inevitably must escape. Therefore,
the presuppositional information was grammaticalized using a textual constituent marked for
identifiability. The ideational content communicated by the constituent is a part of the text's
topic domain.
3.3.4. P.Tebt. 3.1.703, section 2.40-43 (ca. 210 BCE)
ἅμα δὲ ἐν τῶι ἐφοδεύειν πειρῶ περιερχόμενος ἕκαστον παρακαλεῖν καὶ
εὐθαρ̣σεστέρους παρασκευάζειν.
And at the same time, in the inspection, try going around to encourage each one and
make them more courageous.
Context: This text was found in Tebtunis, Egypt. It has been given the title "Instructions
of a Dioecetes to a Subordinate."246 In the instructions, the διοικητής directs his subordinate
to inspect the water conduits and irrigation system in an agricultural area to ensure that
proper maintenance is done and the entire system continues to operate smoothly.
SS and IS: Structurally, ἐν τῶι ἐφοδεύειν is an adjunct that modifies the finite imperative
verb πειρῶ (try) and its complementary infinitives παρακαλεῖν and παρασκευάζειν.

246

https://papyri.info/ddbdp/p.tebt;3.1;703; accessed 18 June 2024.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

68

Semantically, it communicates the occasion in time (temporal scope) for the imperative. It is,
then, within the temporal boundary of making inspections that encouragement was to occur
(see Figure 2.4.1.2). The syntactic construction makes anaphoric reference to a text-internal
constituent, the instructions that were given previously. Therefore, the information
concerning the inspection is encoded using an identifiable textual constituent that is active in
the discourse register. The newly asserted information – the imperative to "try going around
to encourage each one and to make more courageous" – is anchored cognitively to
identifiable information, semantically to a temporal boundary, and ideationally to
information that is presuppositional. It is redundant information, but its inclusion adds
coherence to the utterance.
3.3.5. UPZ. 1.110, lines 94-97 (21 September 164 BCE)
τίς γὰρ οὕτως ἐστὶν ἀνάλιστος ἢ νωθρὸς ἐν τῶι λογίζεσθαι καὶ πράγματος διαφορὰν
εὑρεῖν, ὃς οὐδʼ αὐτὸ τοῦτό γε δυνήσεται συννοεῖν;
For who is so silly or dull in thinking and in ascertaining the distinction of a matter,
he will not even be able to understand this very thing?
Context: This text247 discusses a royal decree regarding agriculture that was issued
because of the need to increase grain production. Capable laborers were ordered to help
contribute. Also, fallow land was confiscated by the state and assigned to farmers willing to
cultivate it.248 But the decree was not unreasonable: it was not intended to confiscate and
redistribute the land of civil servants (i.e., officials, soldiers, manual laborers working on
state projects, etc.) and people unable to do manual labor, as some people had so foolishly
interpreted the decree. For these people were "worn out day and night" serving the state, and
they had a legitimate reason for not contributing to increased grain production.
SS and IS: Structurally, ἐν τῶι λογίζεσθαι...εὑρεῖν modifies the copula and its arguments:
τίς οὕτως ἐστὶν ἀνάλιστος ἢ νωθρὸς (who is so silly or dull). The construction gives the
247

My own translation of lines 84-97: "For perhaps what was said is thus fitting, inasmuch as it is
foolish making the intent of the ordinance to suppose it necessary that even those who are worn out
day and night in the city in public service and others who are unable, are forced to accept that their
farms and property are taken as through the ordinance, having the sum total divided with one accord
among everyone in the region who is a registered citizen. For who is so silly or dull in thinking and in
ascertaining the distinction of a matter, he will not even be able to understand this very thing."
248
See J. G. Manning, The Last Pharaohs: Egypt Under the Ptolemies, 305-30 BC (Princeton:
Princeton University, 2010), 124. He writes, "All land was either 'royal land,' directly managed by the
king, or it was 'conceded' to others to work, and could be confiscated by the king as he desired."

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

69

occasion in time (temporal scope) for the verb and its arguments (see Figure 2.4.1.2). In other
words, it was within the temporal boundary of thinking about and ascertaining the distinction
of a matter – or rather, the inability to do so – that people demonstrated they were so silly and
dull that they could not understand the decree's intent. The syntactic construction
communicates information that ought to be taken for granted, i.e., presuppositional
information, as it refers back to a previous sentence's description of how certain people were
interpreting the royal decree. The information, then, is presuppositional and encoded using a
textual constituent marked for identifiability.
3.3.6. P.Oxy. 4.743, lines 35-38 (4 October 2 BCE)249
ἐν τῷ δέ με περισπᾶσθαι οὐκ ἠδυνάσθην συντυχεῖν Ἀπολλωνίῳ τῷ Λιβικῷ ἵνα αὐτῷ
αὐτὰ ταῦτα ὑποδείξω.
But upon my being distracted, I was unable to meet Apollonius the Libyan in order
to inform him of these things.
Context: This papyrus was found in Oxyrhynchus and is titled "Letter to a Friend." The
letter's author informs his friend that he was upset because another man named Helenos had
lost his money. Due to dealing with issues related to the lost money, he was distracted and
therefore unable to meet with another man named Apollonius the Libyan.
SS and IS: Structurally, ἐν τῷ με περισπᾶσθαι (upon me being distracted) is an adjunct
that modifies the finite verb ἠδυνάσθην and its complementary infinitive συντυχεῖν (to meet).
The construction provides the occasion in time (temporal cause) for its predicate, the finite
verb ἠδυνάσθην and the infinitive συντυχεῖν (see Figure 2.4.1.3). Additionally, the
construction refers to the situation described in the previous sentences, the loss of money and
its repercussions in the author's life. Although it was never explicitly stated that the events
caused a distraction or ought to impede the author from other activities, there are
sociolinguistic implications for how the author is using a syntactic construction marked for
identifiability. He wants the reader to understand why he might be unable to meet with

249

My own translation of lines 22-38: "For I am also completely upset that Helenos lost money. For
when Damas arrived in Alexandria, we came to Epaphroditus, and it was discovered that he had
neither received nor paid anything. So then, I want you to know this that I had given him orders to go
to Takona for the sake of the produce [used as rent], and now I have sent him to collect them all and
have entrusted to him the care of the whole matter. Whatever service he may require from you, assist
him, as he will make an agreement in everything for you just as for me. But when I was distracted, I
was unable to meet Apollonius the Libyan in order to inform him of these things."

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

70

Apollonius the Libyan, so he frames the utterance in such a way that its discourse pragmatic
status will induce the reader to take it for granted as presuppositional information. James
Paul Gee explains this linguistic phenomenon, "The grammatical choice as to what
information to assert and what to assume sets up a social contract between the writer and
reader. The reader is supposed to address any comments or disagreements with the asserted
information and simply take the assumed information for granted...And, of course, speakers
and writers can seek to use this social contract to manipulate listeners and readers into not
bringing up (or not thinking too much about) what they don’t want brought up or thought
about."250 Therefore, in this text, the author does not want the reader to question why he
would not be able to meet with Apollonius, but he wants the reader to assume it as
presuppositional, given the extenuating circumstances. The information, then, was encoded
using a grammatical constituent mark for identifiability. The constituent is topical.
3.4. Post-classical Inscriptions
Two inscriptions have been found that contain the ἐν τῷ + infinitive construction, one in
Pergamum dating to the Hellenistic era and the other in Egypt dating to the Ptolemaic era.
Like the papyri, the construction was not used in inscriptions very frequently, but the two
examples are sufficient to demonstrate how it was used in conventional PC Greek,
particularly in more formal Greek.
3.4.1. IvP II 252, lines 8-9 (2nd-1st century BCE)
χορηγίαν ἔθη̣κεν δὲ καὶ ἐν τῶι κατασκευάσασθαι λουτῆρας δημοσίους δύο.
And he also provided funds during the construction [of] two public baths.
Context: This example is a part of an inscription found in Pergamum from the second or
first century BCE. The inscription honors Metrodorus, son of Herakleon. He had served
Pergamum well as its gymnasiarch (γυμνασίαρχος). He had been a patron for various civic
projects, common for wealthy gymnasiarchs, and had taken care of young men's education at
the gymnasium. This role brought him honor, evident by the inscription dedicated to him.
The text above is contained in the description of his role as patron. There are three projects
listed, with the third project's description containing the syntactic construction. The first

250

Gee, Introducing Discourse Analysis, 65.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

71

project repaired (ἐπισκευῆς) temples, the second brought water channels into the city to
provide water for public baths, and the third, discussed in conjunction with the phrase ἐν τῶι
κατασκευάσασθαι, provided funds for constructing two public baths.
SS and IS: Structurally, ἐν τῶι κατασκευάσασθαι is an adjunct that modifies ἔθη̣κεν as
its predicator. Semantically, the construction communicates the occasion in time (temporal
scope) for its predicator. In more specific terms, ἐν τῶι κατασκευάσασθαι communicates the
temporal boundary for Metrodorus's provision of funds. It would therefore be appropriate to
use "during" as a rendering: And he also provided funds during the construction of two
public baths. He may have provided funds for the other projects mentioned, but this specific
utterance only discusses funds provided for the construction of the public baths.
Ἐν τῶι κατασκευάσασθαι was encoded using a textual constituent marked for
identifiability. Therefore, it grammaticalized information expected to be taken for granted
(presuppositional), because of prior information in the inscription that discussed building,
restoration (ἐπισκευῆς), and construction in the domain of civic projects. As an inscription
displayed publicly, it would also be expected that its readers would be conscious of the
public baths that had been constructed in the city. Therefore, its IS status as presuppositional
information has both a text-internal rationale and a text-external rationale. Additionally, the
syntactic construction belongs to the utterance's topic domain. If topic and focus were posed
as a question (topic) and answer (focus):251 "What did he provide during the construction
(topic)? FUNDS FOR TWO PUBLIC BATHS (focus)."
3.4.2. I.Prose. 30 (96 BCE)
τὸ ἐν Ἀθρίβει τοῦ Ἀρκεντεχθαὶ τὸ μὲν πρῶτον καὶ λόγιμον ἀρχαιότατόν τε καὶ
ἐνδοξότατον τῶν πλείστων ὑπάρχον τῶν μὲν ἄλλων τιμῶν τετευχέναι λείπεσθαι δὲ ἐν
τῷ μὴ εἶναι ἄσυλον, προστετάχαμεν ἐπιχωρῆσαι καὶ τούτῳ τῷ ἱερῶι ἐντὸς αὐτοῦ
περιβόλου τὴν ἀσυλίαν καθάπερ ἐστὶ καὶ τῶι ἐν Μέμφει καὶ Βουσίρει καὶ ἑτέροις καὶ
τῶν ἄλλων ἱερῶν·
while the temple in Athribis of the god Horus is on the one hand the first notable one
to have been built – being both oldest and most glorious and possessing most other
honors – but on the other hand was deficient in not being a [legally sanctioned]
asylum, we determined to grant the right of asylum to this temple and its precincts
just as also in Memphis, Bousiris, and other temples in other cities.
251

In information structure analysis, this is referred to as a "question-under-discussion" approach. See
Lisa Brunetti, et al., "The Information-Structural Status of Adjuncts: A Question-under-DiscussionBased Approach," 16-17.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

72

Context: This example is part of an inscription found on a stele in Athribis, Egypt.252 It
contains the grant of asylum to the temple of Horus (Ἀρκεντεχθαὶ) by Ptolemy X Alexander
(r. 107–88 BCE).
SS and IS: Structurally, ἐν τῷ μὴ εἶναι ἄσυλον is an adjunct that modifies the infinitive
λείπεσθαι (was deficient) as its predicator. Therefore, ἐν τῷ μὴ εἶναι ἄσυλον communicates
the occasion in time for the temple's deficiency. This is a causal semantic relationship that
only peripherally possesses any temporal semantic value; however, temporality is not
completely muted. One could argue that within the temporal boundary of the temple not
possessing the right of asylum, it had cause for deficiency (see Figure 2.4.1.3 in chapter two).
The syntactic constituent ἐν τῷ μὴ εἶναι ἄσυλον was encoded using an article, which
marks it as an identifiable textual constituent. Therefore, the inscription's author expected his
readers to treat the information as presuppositional. It is an indirect textual anaphor, since the
prior context mentions other prominent temples being granted the right of asylum. It is
implicit that these temples were once also in the state of not being legally sanctioned asylums
before Ptolemy granted them the right of asylum.253 Additionally, there is a text-external
rationale for the information status of ἐν τῷ μὴ εἶναι ἄσυλον. The first readers of the stele
would have been aware that the temple formerly did not possess the right of asylum. The
language of the inscription was certainly composed with these readers in mind. Therefore,
there are text-internal and text-external reasons for understanding the syntactic construction's
usage in the text to be presuppositional. Additionally, it is a part of the topic domain in the
sentence.

252

My own translation of the entire inscription: "King Ptolemy, who is also called Alexander. After
Ptolemy, the Kinsman and governor of Egypt, reported to us that all the temples in Egypt have indeed
received great benefactions from our ancestors, including some of the prominent ones having been
bestowed the right of asylum, while the temple in Athribis of the god Horus is on the one hand the
first notable one to have been built – being both oldest and most glorious and possessing most other
honors – but on the other hand was deficient in not being a [legally sanctioned] asylum, we
determined to grant the right of asylum to this temple and its precincts just as also in Memphis,
Bousiris, and other temples in other cities. Therefore, let it be done accordingly."
253
This was done throughout the Ptolemaic era for various temples, even synagogues (cf. I.Asylia
228).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

73

3.5. Summary
Structurally, the construction is typically used as an adjunct in the texts discussed in this
chapter. However, the texts also provide examples of ἐν τῷ + infinitive in a complement
structural position. Semantically, though, the construction always indicates occasion in time
for its predicator. Its more specific semantic nuances are dependent on the contextual use of
lexical constituents. Sometimes the nuance is causal, sometimes limitative (providing
temporal scope); sometimes the contexts are simply reporting events, sometimes the contexts
are conditional, and sometimes the contexts are gnomic. Pragmatically, though, the
construction always communicates presuppositional information that was marked for
identifiability with the article.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

74

CHAPTER 4: ANALYSIS OF SEPTUAGINT GENESIS TEXTS
"[The] translators...tried to put into Greek dress what they believed God intended to say to
his people."254
–John William Wevers
4.1. Introduction
Gen produced a text that generally rendered the Hebrew Vorlage phrase by phrase with
very little divergence either in the constituent ordering of clause and sentence or in the
lexical semantics of its renderings. This translation technique was rooted in the translator's
view that the Hebrew book of Genesis, as the first book of the Torah (indeed the entire
Tanakh), was especially sacred. As Wevers has noted, this "meant that the translators
considered their task thoughtfully, did not simply put Hebrew words into equivalent Greek
lexemes, but tried to put into Greek dress what they believed God intended to say to his
people."255 The result was a text that was intelligible to those who read it, characterized by
close alignment with the source text on the one hand and conventional Greek usage on the
other hand. These two aspects of translation technique are not mutually exclusive. Wevers
asserts, "Their translation may not have been perfect, but it made sense to [the translators];
they did not create nonsense, and when the modern reader is puzzled, the fault must lie with
him or her, not with the translator. It means that the student of LXX must try to understand
what the translators meant rather than stand in arrogant judgment over their product."256
Due to the Greek translation’s close alignment with its Hebrew Vorlage, some scholars
have found support for the interlinear paradigm in the text-linguistic structure of LXX
Genesis. For example, Hiebert writes, "The overall assessment of Greek Genesis is that,
lexically and syntactically, it is a strict, quantitative representation of its source text. Thus,
the concept proposed in NETS discussions of the Septuagint (LXX) as an interlinear
translation is an apt metaphor for this book because of the significant degree of dependence
on the Hebrew that it exhibits."257 First, it is not self-evident that if a translation has a
254

John William Wevers, Notes on the Greek Text of Genesis, Society of Biblical Literature
Septuagint and Cognate Studies Series 35 (Atlanta: Scholars Press, 1993), xii.
255
Ibid., xii.
256
Ibid., xiii-xiv.
257
Robert J. V. Hiebert, "Genesis: To the Reader," in A New English Translation of the Septuagint,
eds. Albert Pietersma and Benjamin G. Wright (Oxford: Oxford University, 2007), 1.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

75

"significant degree of dependence" on its source text that it can be metaphorically
characterized as an interlinear. Every translation is significantly dependent on its source text.
If a translation was not, then it would not be considered a translation. Second, there are a
variety of linguistic features throughout LXX Genesis that can be used as evidence to support
the hypothesis of an interlinear paradigm. Hiebert helpfully identifies many of those features
in various essays.258 However, when those features are considered within the total lexical and
syntactical usage in LXX Genesis, it is perhaps not apparent that those features are indicative
of the entire text. The most dominant linguistic feature that supports the interlinear metaphor
is arguably serial fidelity, that is, the tendency of LXX Genesis to closely follow the Hebrew
word order. Any discussion of word order in the LXX, though, must also take into
consideration that AG, as an inflectional language, was quite flexible in the order and
arrangement of its grammatical constituents. Although Gen did not use word order to
pragmatically structure information, his serial fidelity did not affect sentence structure or
semantics. Given the great diversity in Greek word order from one author to the next and one
sentence to the next, the translator chose to closely follow the Hebrew word order because it
was just as viable as any other word order on the level of semantics, and it fulfilled
traditional assumptions about faithfulness in the translation of sacred texts.259 Therefore,
typically the word order in each sentence in LXX Genesis is both dependent on its source
text and ordinarily encodes meaning according to the conventions of AG.
An illustration might be appropriate for articulating the nature of the Greek in LXX
Genesis. In Bali, Indonesia, it is quite easy to tell when one enters a Hindu house. The
structure of the house and the materials used to build it are the same as those of any other
house. However, there are certain features of a Hindu house that make it distinct, such as the
decorations scattered throughout the house and the little altars where offerings are made. It is
the same with LXX Genesis. The syntactic structures and lexical material used to render the
text are, by and large, conventional Greek. For the most part, they do not differ from those in
PC compositional texts. However, there are certain features scattered throughout the text,
258

Ibid., 3-4. See also the discussion on pp 49-50.
Sebastian Brock, "Aspects of Translation Technique in Antiquity," in Greek, Roman, and Byzantine
Studies 20.1 (2004): 81. Brock writes, "In the case of the biblical translator, for whom 'the very word
order is a mystery,' reproduction of the word order of the source language is essential."
259

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

76

little Hebraic linguistic "decorations," that give LXX Genesis a distinctive biblical style.
These features are possibly an intentional aspect of the translator's strategy to "foreignize"
the text for his readers,260 so that when Jews in third-century BCE Alexandria read the text,
they realized that they were being led behind the text to a different time, culture, and
linguistic milieu.
The selective usage of ἐν τῷ + infinitive by Gen, however, is not one such Hebraic
linguistic "decoration." This construction appears twenty-three times in LXX Genesis. It is
used eighteen times as a rendering for the Hebrew ‫ ְבּ‬+ infinitive and five times as a rendering
for various other Hebrew constructions (see figure 4 below). In these texts, ἐν τῷ + infinitive
occurs ten times in subordination to ἐγένετο or ἔσται plus another finite verb in a dependent
clause. This is an idiosyncratic sentence structure that is only attested in biblical Greek (as
the following discussion will articulate). However, the discussion of the texts below using IS
analysis will indicate that Gen's usage of ἐν τῷ + infinitive was not idiosyncratic. He
employed it in a manner consistent with the PC Greek compositional texts in chapter 3. The
following discussion is divided into four groups of texts:
1) Texts that use ἐν τῷ + infinitive as a rendering for ‫ ְבּ‬+ infinitive and subordinate
the construction to ἐγένετο or ἔσται + finite verb [8x],
2) Texts that use ἐν τῷ + infinitive as a rendering for other Hebrew constructions and
subordinate the construction to ἐγένετο or ἔσται + finite verb [3x],
3) Texts that use ἐν τῷ + infinitive as a rendering for ‫ ְבּ‬+ infinitive [10x],
4) Texts that use ἐν τῷ + infinitive as a rendering for other Hebrew constructions
[2x].
In the discussion of each text, it will be necessary to clearly articulate the broader context
in which the syntactic construction is found. This is necessary because the key to
understanding how ἐν τῷ + infinitive was used is often within the discourse context and not
simply within the sentence (although sometimes the construction refers anaphorically to a
260

John Barton, The Word: How We Translate the Bible – And Why It Matters (New York: Basic
Books, 2023), 72-73. Barton writes, "Any translation inevitably acculturates the source text to some
degree within the target language’s norms, but [for Venuti] this is to be resisted as much as possible,
not affirmed. The translator’s task, he argues, is not to create the text the author would have produced
in our culture, but to render the text in all its alien character. Consequently, Venuti favors
‘foreignizing’ translations."

77

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

constituent in the same sentence). So, the discussion of each text will take into consideration
as much or as little of the context needed to illuminate the IS.
Table 4.1: 'Εν τῷ + Infinitive in Genesis
No.

Verse

Hebrew
Vorlage

1
2
3
4
5
6
7
8
9
10
11

4:8
9:14
11:2
19:16
19:29
19:29
19:33
19:33
19:35
19:35
24:52

‫ִבְּהיוָֹתם‬
‫ְבַּﬠ ְנ ִני‬
‫ְבָּנְסָﬠם‬
‫ְבֶּחְמַ֥לת‬
‫ְבּ ַשֵׁחת‬
0‫ַבֲּה ֹפ‬
‫ְבּ ִשְׁכָבהּ‬
‫וְּבקוָּמהּ‬
‫ְבּ ִשְׁכָבהּ‬
‫וְּבֻקָמהּ‬
‫ַכֲּא ֶ֥שׁר ָשַׁ֛מע‬

12
13
14
15
16
17
18
19
20
21
22

28:6
32:20
32:26
34:15
34:22
35:1
35:7
35:17
35:18
38:28
42:35

‫ְבָּבֲרכוֹ‬
‫ְבּ ֹמ ַצֲאֶכם‬
‫ְבֵּהָאְבקוֹ‬
‫ְלִה ֹ֥מּל‬
‫ְבִּהמּוֹל‬
8‫ְבָּב ְרֲח‬
‫ְבָּב ְרחוֹ‬
‫ְבִּלְד ָתּהּ‬
‫ְבּ ֵצאת‬
‫ְבִלְד ָתּהּ‬
‫ְמ ִרי ִ֣קים‬

23

44:31

‫ִכּ ְר ֹאו ֛ ֹתו‬

Hebrew syntax
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ noun
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְכּ‬+ rel. pron. +
Qal verb
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְל‬+ infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
‫ ְבּ‬+ Infinitive
Hiphil ptc.
absolute
‫ ְכּ‬+ infinitive

+ καὶ ἐγένετο,
ἐγένετο δὲ, καὶ
ἔσται
+
+ (καὶ ἔσται)
+
+
+
+

Tense

+
+
+
+

Pres.
Aor.
Pres.
Aor.
Aor.
Pres.
Pres.
Pres.
Pres.
Pres.
Pres.

+ (καὶ ἔσται)

Aor.

Pres.
Pres.
Aor.
Aor.
Aor.
Aor.
Aor.
Aor.
Aor.
Aor.
Aor.

4.2. 'Εγένετο (ἔσται) + 'Εν τῷ + Infinitive (Rendering ‫ ְבּ‬+ Infinitive)
The existential use of ἐγένετο (ἔσται) to form a dependent clause with another finite verb
is a unique feature of biblical Greek and can rightly be labeled a Hebraism.261 Structurally, ἐν
261

Hiebert, "Genesis: To the Reader," 3. It is this use of ἐγένετο (ἔσται) and not the usage of ἐν τῷ +
infinitive that needs to be viewed as a Hebraism. In OG texts, ἐγένετο (ἔσται) + finite verb occurs 49x
with ἐν τῷ + infinitive. Slightly more than 20% of those usages are in LXX Genesis (10x). The
syntax, then, is more frequent in LXX Genesis than any other book in the LXX. Total uses in the
LXX: Gen (10x), Num (2x), Ies (1x), Judg (3x), Routh (2x), 1Rgns (2x), 2Rgns (6x), 3Rgns (3x),
4Rgns (4x), 1Suppl (2x), 2Suppl (5x), 2Esd (1x), Esa (1x), Iezek (4x), Sous (1x), Dan (2x).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

78

τῷ + infinitive modifies both ἐγένετο (ἔσται) and the finite verb. As an adjunct, it gives
temporal information about ἐγένετο (ἔσται) + finite verb. LXX Genesis attests a much higher
frequency of this sentence structure than any other book in the Pentateuch; the only other
examples are in Numbers (2x). In the NT, Luke was arguably imitating not only the style of
the LXX in general when he used the same sentence structure but also LXX Genesis
specifically. Although the Greek compositional corpora do provide examples of the
existential use of ἐγένετο,262 it was not used to subordinate another finite verb in a dependent
clause. In the compositional corpora, the existential use of ἐγένετο is found almost
exclusively in literary texts. Gen, then, was ostensibly attempting to give a quantitative
representation of ‫ ) ְוָהָיה( ַו ְיִהי‬in the Hebrew text while at the same time imitating the style
found in historical literary authors like Thucydides. Although the result with a finite verb is a
Greek construction that had never before been attested in compositional texts, the text itself
was still comprehensible within the Greek linguistic system. As Wevers states above, it was
not nonsense. About the Hebraic use of ἐγένετο, Liana Tronci says that, "the καὶ ἐγένετο
constructions were calqued on BH in the LXX and were used in the NT, since they met some
structural configuration of γί(γ)νομαι already existing in Classical Greek."263 Consider the
following texts that use ἐν τῷ + infinitive as a rendering for ‫ ְבּ‬+ infinitive and subordinate the
construction to ἐγένετο or ἔσται + finite verb:
4.2.1. Genesis 4:8
‫ַו ְיִהי ִבְּהיוָֹתם ַבּ ָשֶּׂדה ַוָיּ ָקם ַק ִין ֶאל־ֶהֶבל ָאִחיו ַוַיַּה ְר ֵגהוּ׃‬
LXX: καὶ ἐγένετο ἐν τῷ εἶναι αὐτοὺς ἐν τῷ πεδίῳ καὶ ἀνέστη Κάιν ἐπὶ Ἅβελ τὸν
ἀδελφὸν αὐτοῦ καὶ ἀπέκτεινεν αὐτόν.
NETS: And it came about when they were in the plain, that then Kain rose up against
his brother Habel and killed him.

262

Some examples in literary texts: Calliphon, Testimonia 1.22; Thucydides, Historiae 1.19.1.6,
1.82.6.2, 1.100.1.1; Herodotus, Historiae 1.190.5; TM 65795, column 17.3 (literary papyrus). One
possible textual example in non-literary documentary sources was found: PSI.6.601.
263
Liana Tronci, "Contact-induced change and language-internal factors: the καὶ ἐγένετο type as a
case-study," Papers On Ancient Greek Linguistics: Proceedings of the Ninth International
Colloquium on Ancient Greek Linguistics, eds. Martti Leiwo, Marja Vierros & Sonja Dahlgren
(Helsinki: Societas Scientiarum Fennica, 2020), 177-204, here 178.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

79

Context: Genesis 4:8 begins with the statement, καὶ εἶπεν Κάιν πρὸς Ἅβελ τὸν ἀδελφὸν
αὐτοῦ Διέλθωμεν εἰς τὸ πεδίον (NETS: "And Kain said to his brother Habel, ‘Let us go
through into the plain’”).264 The construction occurs in the following sentence.
SS and IS: Structurally, he two prepositional phrases ἐν τῷ εἶναι αὐτοὺς ἐν τῷ πεδίῳ are
adjuncts that provide the occasion in time (temporal scope) and the physical location for the
predicator καὶ ἐγένετο...ἀνέστη Κάιν...καὶ ἀπέκτεινεν (And it happened...that Kain rose
up...and killed). In other words, ἐν τῷ + infinitive indicates the temporal boundary within
which Kain rose up and killed his brother. Taken as a single informational unit, the two
prepositional phrases ἐν τῷ εἶναι αὐτοὺς ἐν τῷ πεδίῳ (when they were on the plain) therefore
refer anaphorically to Kain's statement in the previous sentence. Since Kain had invited his
brother to cross onto the plain, the fact that they would be on the plain is presuppositional to
those reading the narrative. Therefore, Gen used an identifiable textual constituent (τῷ εἶναι)
to encode that presupposition.
4.2.2. Genesis 9:14
‫ְוָהָיה ְבַּﬠ ְנ ִני ָﬠָנן ַﬠל־ָהָאֶרץ ְו ִנ ְרֲאָתה ַה ֶקּ ֶשׁת ֶבָּﬠָנן׃‬
LXX: καὶ ἔσται ἐν τῷ συννεφεῖν με νεφέλας ἐπὶ τὴν γῆν ὀφθήσεται τὸ τόξον μου ἐν
τῇ νεφέλῃ,
NETS: And it shall be that when I gather clouds over the earth, my bow shall be
seen in the cloud,
Context: This text contains words spoken by God to Noah and his family after they came
out of the ark. God renewed the creation covenant with Noah and gave the bow as a covenant
sign representing the promise to never again destroy the earth by flood.
SS and IS: Structurally, ἐν τῷ συννεφεῖν με is an adjunct that gives the occasion in time
(temporal scope) for its semantic predicator, the verb ὀφθήσεται ("shall be seen"; structurally
καὶ ἔσται...ὀφθήσεται). Therefore, within the temporal boundary in which the clouds were
gathered, the bow would be seen. The syntactic construction refers anaphorically to a textual
constituent in Genesis 9:13, "I set my bow in the cloud [ἐν τῇ νεφέλῃ]." The syntactic
construction with its object, ἐν τῷ συννεφεῖν με νεφέλας (when I gather clouds), denotes

264

A more idiomatic translation compared to that of the NETS (which renders the text according to its
own stated translation principles) that takes into consideration a cognitive linguistic perspective of
εἰς: "Let us go across onto the plain." See Rhodes, "Greek Prepositions: A Cognitive Linguistic
View,” 20.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

80

cloud formation. In the worldview communicated by the text, the existence of the cloud is an
indication that God has gathered the cloud. Therefore, in 9:14, the syntactic construction
would at least communicate information that is presuppositional due to its textual
representation in 9:13. However, its presuppositional status in 9:14 is only due in part to its
antecedence in the previous verse in 9:13.265 Notice that νεφέλη is also marked for
identifiability in 9:13. This is because the translators assumed every reader would know
about the existence of clouds because of their common experience in the text-external world.
This would also be true of many other things that are part and parcel of common human
experience. Textual anaphora and identifiable textual constituents do not necessarily have a
structural antecedent to which they refer.266 In 9:14, though, the syntactic construction is
marked for identifiability both due to its structural antecedent in 9:13 and the fact that Gen
assumes that its denotatum ought to be taken for granted.
4.2.3. Genesis 11:2
‫ַו ְיִהי ְבָּנְסָﬠם ִמ ֶקֶּדם ַו ִיְּמ ְצאוּ ִב ְקָﬠה ְבֶּאֶרץ ִשׁ ְנָﬠר ַוֵיּ ְשׁבוּ ָשׁם׃‬
LXX: καὶ ἐγένετο ἐν τῷ κινῆσαι αὐτοὺς ἀπὸ ἀνατολῶν εὗρον πεδίον ἐν γῇ Σενναὰρ
καὶ κατῴκησαν ἐκεῖ.
NETS: And it came about as they moved from the east that they found a plain in the
land of Sennaar and settled there.
Context: This attestation is found in the narrative about the Tower of Babel. Chapter 10
has already mentioned Sennaar being settled by Nebrod and his descendants (10:9-10).
Furthermore, chapter 10 – after listing the descendants of Sem, Cham, and Iapheth and the
places where they settled – closes with the summary statement, "from these the islands of the
nations were scattered abroad on the earth after the flood" (NETS).
SS and IS: Structurally, ἐν τῷ κινῆσαι αὐτοὺς is an adjunct that semantically provides
the occasion in time (temporal cause) for its predicators, the finite verbs εὗρον (they found)
and κατῴκησαν (they settled). While it is within the temporal boundary of moving eastward
265

Remember Lambrecht's definition of pragmatic presupposition in chapter 1: "The set of
propositions lexicogrammatically evoked in a sentence which the speaker assumes the hearer already
knows or is ready to take for granted at the time the sentence is uttered" (Lambrecht, Information
Structure and Sentence Form, 52).
266
For a discussion of anaphora in AG, see Paul Kiparsky, "Greek Anaphora in Cross-Linguistic
Perspective," Journal of Greek Linguistics 12 (2012): 84–117. Kiparsky states concerning discourse
anaphors that they "need not have a structural antecedent" (85). Several of the texts discussed in this
thesis demonstrate the validity of that statement.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

81

that they found a plain and settled there, their movement also caused them to find and settle
on the plain. However, the temporal causal relationship with each finite verb is slightly
different. The semantic relationship with εὗρον can be illustrated with Figure 2.4.1.3, while
the semantic relationship with κατῴκησαν can be illustrated with Figure 2.4.1.4 (since the
state of affairs denoted by κατῴκησαν continues after the state of affairs denoted by ἐν τῷ
κινῆσαι αὐτοὺς ceases). The syntactic construction communicates presuppositional
information due to what has already been added to the discourse register in Genesis 10. The
construction grammaticalizes this presupposition using a textual constituent that is marked
for identifiability. The construction also pragmatically encodes the sentence's topic.
4.2.4. Genesis 19:29
‫ ֶאת־‬¢‫ ַהֲהֵפָכה ַבֲּהֹפ‬0‫ִהים ֶאת־ַאְבָרָהם ַו ְי ַשׁ ַלּח ֶאת־לוֹט ִמתּוֹ‬E‫ִהים ֶאת־ָﬠֵרי ַה ִכּ ָכּר ַו ִיְּזֹכּר ֱא‬E‫שֵׁחת ֱא‬
ַ ‫ַו ְיִהי ְבּ‬
‫ֶהָﬠ ִרים ֲא ֶשׁר־ָי ַשׁב ָבֵּהן לוֹט׃‬
LXX: καὶ ἐγένετο ἐν τῷ ἐκτρῖψαι τὸν θεὸν πάσας τὰς πόλεις τῆς περιοίκου ἐμνήσθη
ὁ θεὸς τοῦ Ἀβραὰμ καὶ ἐξαπέστειλεν τὸν Λὼτ ἐκ μέσου τῆς καταστροφῆς ἐν τῷ
καταστρέψαι κύριον τὰς πόλεις, ἐν αἷς κατῴκει ἐν αὐταῖς Λώτ.
NETS: And it came about when God annihilated all the cities of the neighboring
region that God remembered Abraam and sent Lot out of the midst of the overthrow
when the Lord overthrew the cities which Lot used to dwell in.
Context: These two attestations of the syntactic construction are found in the narrative
about Sodoma and Gomorra’s destruction. In 19:24-25, it is stated that the Lord rained fire
and sulfur from heaven on the two cities to overthrow them. In 19:28, Abraam is described as
seeing the flame of the land going up like the vapor of a furnace.
SS and IS: The two syntactic constructions, ἐν τῷ ἐκτρῖψαι and ἐν τῷ καταστρέψαι, each
head temporal adjunctive clauses that report the same event. But the semantic relationship
that each construction has with its predicator is different. Both constructions grammaticalize
the occasion in time. The first one, ἐν τῷ ἐκτρῖψαι, provides the temporal scope for its
predicator ἐμνήσθη (remembered; see Figure 2.4.1.2) while the second one, ἐν τῷ
καταστρέψαι, furnishes the temporal cause for its predicator ἐξαπέστειλεν (sent; see Figure
2.4.1.3). The two prepositional phrases demonstrate that temporal scope and temporal cause
are based upon the lexical semantics of ἐν τῷ + infinitive and its predicator. This arguably
offers a clearer perspective than those of some of the grammarians discussed in chapter two.
They view the syntactic construction itself as absolutely encoding a particular semantic
relationship with its predicator in certain corpora of AG. Thus, the argument is that CG used

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

82

ἐν τῷ + infinitive to indicate cause, and the LXX employed it distinctively to indicate
temporality in imitating Hebrew syntax and semantics. A more nuanced view would be that
the syntactic construction only encodes meaning based on the lexical semantics of the
infinitive and predicator, constrained by the structural semantics of the prepositional phrase.
When lexemes are different, such as ἐμνήσθη (remembered) and ἐξαπέστειλεν (sent), then
the specific semantic relationship also changes accordingly.
Both ἐν τῷ ἐκτρῖψαι and ἐν τῷ καταστρέψαι refer anaphorically to the event recorded in
19:24-25. Therefore, both syntactic constructions communicate presuppositional information
that is grammaticalized using a textual constituent marked for identifiability. The
constructions also communicate topical information for their respective clauses that are
joined by καί.
4.2.5. Genesis 35:17
ֹ ‫ַו ְיִהי ְבַהְק‬
‫ ֵבּן׃‬0‫שָׁתהּ ְבִּלְד ָתּהּ ַותּ ֹאֶמר ָלהּ ַהְמַי ֶלֶּדת ַאל־ ִתּי ְרִאי ִכּי־ ַגם־ֶזה ָל‬
LXX: ἐγένετο δὲ ἐν τῷ σκληρῶς αὐτὴν τίκτειν εἶπεν αὐτῇ ἡ μαῖα Θάρσει, καὶ γὰρ
οὗτός σοί ἐστιν υἱός.
NETS: And it came about while she was giving birth with difficulty that the
midwife said to her, “Take courage, for this one also is a son for you.”
Context: This text is taken from the narrative of Rachel giving birth to Beniamin.
SS and IS: Structurally, ἐν τῷ σκληρῶς αὐτὴν τίκτειν is an adjunct that modifies both
ἐγένετο at the beginning of the sentence and εἶπεν (as is the case with every usage of the
construction with ἐγένετο); semantically, its primary predicator is εἶπεν. An argument could
be made for either temporal cause or temporal scope as the semantic relationship between ἐν
τῷ τίκτειν and εἶπεν. It is possible to perceive the predicator as being caused by the ἐν τῷ +
infinitive clause: the midwife spoke and said "Take courage..." because Rachel was giving
birth with difficulty (see Figure 2.4.1.3). It is also possible to perceive the predicator as
taking place within the temporal boundary of the state of affairs denoted by the ἐν τῷ +
infinitive construction: the midwife spoke and said "Take courage..." while Rachel was
giving birth with difficulty (see Figure 2.4.1.2). Whether the semantic relationship is best
described as temporal scope or cause, the syntactic construction communicates the occasion

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

83

in time for its predicator. The construction makes anaphoric reference267 to an event activated
in the discourse register in the previous verse: ἔτεκεν Ῥαχήλ καὶ ἐδυστόκησεν ἐν τῷ τοκετῷ
(Rachel gave birth, and she experienced severe birth pangs in the birth). The clause is used as
background in 35:17 to provide a temporal frame for the midwife's speech. Consistent with
the discourse pragmatic function of ἐν τῷ + infinitive, ἐν τῷ τίκτειν communicates
information that is presuppositional, grammaticalizing a textual constituent marked as
identifiable (active in the discourse register), and topical.268
4.2.6. Genesis 35:18
‫ַו ְיִהי ְבֵּצאת ַנְפ ָשׁהּ ִכּי ֵמָתה ַו ִתּ ְקָרא ְשׁמוֹ ֶבּן־אוֹ ִני ְוָאִביו ָקָרא־לוֹ ִב ְנָיִמין׃‬
LXX: ἐγένετο δὲ ἐν τῷ ἀφιέναι αὐτὴν τὴν ψυχήν, ἀπέθνῃσκεν γάρ, ἐκάλεσεν τὸ
ὄνομα αὐτοῦ Υἱὸς ὀδύνης μου· ὁ δὲ πατὴρ ἐκάλεσεν αὐτὸν Βενιαμίν.
NETS: And it came about as she was giving up her soul, for she was dying, that she
called his name Son-of-my-pain, but his father called him Beniamin.
Context: This text is the next verse in the narrative of Rachel giving birth to Beniamin.
SS and IS: Structurally, ἐν τῷ ἀφιέναι αὐτὴν τὴν ψυχήν is an adjunct that gives the
occasion in time (temporal scope; see Figure 2.4.1.2) for its predicator ἐκάλεσεν (called).
Within the temporal boundary in which she was giving up her soul, Rachel called the child's
name Son-of-my-pain. This attestation is perhaps the most difficult to explain from the
perspective of pragmatic IS. The ideational content communicated by the syntactic
construction has not been activated previously in the discourse register. However, it is
perhaps anticipated by the words in 35:16, "Rachel gave birth, and she experienced severe
birth pangs in the birth" (ἔτεκεν Ῥαχήλ καὶ ἐδυστόκησεν ἐν τῷ τοκετῷ). Although
experiencing severe birth pangs (ἐδυστόκησεν) in the ancient world did not explicitly
indicate the mother would not survive,269 it perhaps might be expected in a nomadic, pastoral
culture. Furthermore, the imperfective aspect of the present infinitive indicates that Rachel
was in the process of giving up her soul as it had not yet actually occurred. Considering the
typical function of ἐν τῷ + infinitive in discourse, the translator used the construction
because he wanted his readers to take it for granted (i.e., consider it presuppositional) that
267

’Εν τῷ + infinitive is not the only anaphora in the clause. The pronoun αὐτὴν and the adverb
σκληρῶς likewise point backwards to the finite verb ἐδυστόκησεν in the previous sentence.
268
The topic would be something like what happened while Rachel was giving birth with difficulty.
269
Hippocrates, Epidemics 2, 4-7, ed. and trans. Wesley D. Smith, Loeb Classical Library 477
(Cambridge, MA: Harvard University Press, 1994), 36-37. This mother survived.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

84

Rachel's severe birth pangs and difficulty giving birth would indicate that she was in the
process of giving up her soul.270 Her possible death would have been activated, therefore, as
a part of the cognitive schema of giving birth with severe pain and great difficulty in the
ancient world. This information was therefore grammaticalized using a textual constituent
marked for identifiability. This information also communicates something about the topic,
namely, "What Rachel did when she was giving up her soul."
4.2.7. Genesis 38:28
‫ַו ְיִהי ְבִלְדָתּהּ ַו ִיּ ֶתּן־ָיד ַו ִתּ ַקּח ַהְמַי ֶלֶּדת ַו ִתּ ְקֹשׁר ַﬠל־ָידוֹ ָשׁ ִני ֵלא ֹמר ֶזה ָי ָצא ִראֹשָׁנה׃‬
LXX: ἐγένετο δὲ ἐν τῷ τίκτειν αὐτὴν ὁ εἷς προεξήνεγκεν τὴν χεῖρα· λαβοῦσα δὲ ἡ
μαῖα ἔδησεν ἐπὶ τὴν χεῖρα αὐτοῦ κόκκινον λέγουσα Οὗτος ἐξελεύσεται πρότερος.
NETS: And it came about as she was giving birth that the one put forth his hand, and
taking it, the midwife bound scarlet material on his hand, saying, “This one will come
out earlier.”
Context: This attestation is found in the narrative about Ioudas impregnating Thamar. In
38:27, the text states, "Now it came about at the time she was giving birth, that then she had
twins in her womb."
SS and IS: Structurally, ἐν τῷ τίκτειν αὐτὴν is an adjunct that gives the occasion in time
(temporal scope) for its predicator προεξήνεγκεν (put forth). Within the temporal boundary in
which Thamar was giving birth, one of her twins put forth his hand. As is the case with many
other attestations, this is background information that provides a temporal frame for the
nuclear clause (i.e., main clause). This construction refers anaphorically to the constituent in
38:27, ἡνίκα ἔτικτεν (at the time she was giving birth). It contains presuppositional
information that is already active in the discourse register. Therefore, it has been
grammaticalized using a textual constituent marked for identifiability. The information is
also topical.

270

The collocation ἀφίημι + ψυχή was used in AG compositional texts as an idiom to indicate that
death had occurred, especially when used with the perfective aspect. The imperfective aspect – while
typically indicating death – could also indicate that someone was giving their life for or with another
without the implication of death. See Esdras A 4:21 and Josephus Antiquities 11:52 for two examples
in Jewish compositional texts. So, there was some ambiguity with its usage in the imperfective aspect.
However, the context in Gen. 35:18 does make it clear that the process of death was intended, most
obviously with the clause following ἐν τῷ + infinitive: ἀπέθνῃσκεν γάρ ("for she was dying"). This
clause uses a finite verb to explicitly communicate newly asserted information.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

85

4.3. 'Εγένετο (ἔσται) + 'Εν τῷ + Infinitive (Rendering Other Hebrew Constructions)
4.3.1. Genesis 24:52
‫שַׁ֛מע ֶ֥ﬠֶבד ַאְבָרָ֖הם ֶאת־ ִדְּבֵריֶ֑הם ַו ִיּ ְשׁ ַ֥תּחוּ ַ֖א ְר ָצה ַ ֽליה ָֽוה׃‬
ָ ‫ַו ְי ִ֕הי ַכֲּאֶ֥שׁר‬
LXX: ἐγένετο δὲ ἐν τῷ ἀκοῦσαι τὸν παῖδα τὸν Ἀβραὰμ τῶν ῥημάτων αὐτῶν
προσεκύνησεν ἐπὶ τὴν γῆν κυρίῳ.
NETS: Now it came about when the servant of Abraam heard their words that he
did obeisance upon the ground to the Lord.
Context: This text is found in the narrative about the journey of Abraam's servant to
Mesopotamia to find a wife for Isaak in Nachor's family, Abraam's people. After meeting
Rebekka, the servant then met her father Bathouel and brother Laban. He explained the
purpose of his journey and how, ostensibly, Rebekka was the answer to his prayer. Then,
Laban and Bathouel responded with the words, "This ordinance has come from the Lord;
therefore, we shall not be able to speak against you bad for good. Look, Rebekka is before
you; take her, leave quickly, and let her be wife to your lord’s son, as the Lord has spoken"
(24:50-51).
SS and IS: Structurally, the syntactic construction is an adjunct that provides the
occasion in time (temporal cause) for the predicator προσεκύνησεν (did obeisance).
Therefore, what Laban and Bathouel said caused the servant to do obeisance upon the ground
(see Figure 2.4.1.6). This should be viewed as background information that provides a
temporal causal frame for the nuclear clause (i.e., one state of affairs that causes another state
of affairs in a temporal sequence). The fact that 24:50 reports that Laban and Bathouel
replied to the servant implies that he heard their words. The construction ἐν τῷ ἀκοῦσαι in v.
52 thus signifies information that should be taken for granted due to the context; therefore, it
is presuppositional information that has been grammaticalized using a textual constituent
marked for identifiability. It encodes information that is a part of the utterance’s pragmatic
topic.
4.3.2. Genesis 42:35
‫ַו ְיִ֗הי ֵ֚הם ְמ ִרי ִ֣קים ַשׂ ֵקּי ֶ֔הם ְוִה ֵנּה־ ִ֥אישׁ ְצרוֹר־ ַכְּסֹ֖פּו ְבַּשֹׂ֑קּו ַו ִיּ ְר ֞אוּ ֶאת־ ְצ ֹר ֧רוֹת ַכְּס ֵפּיֶ֛הם ֵ֥ה ָמּה ַוֲאִביֶ֖הם‬
‫ַו ִיּי ָֽראוּ׃‬
LXX: ἐγένετο δὲ ἐν τῷ κατακενοῦν αὐτοὺς τοὺς σάκκους αὐτῶν καὶ ἦν ἑκάστου ὁ
δεσμὸς τοῦ ἀργυρίου ἐν τῷ σάκκῳ αὐτῶν· καὶ εἶδον τοὺς δεσμοὺς τοῦ ἀργυρίου
αὐτῶν, αὐτοὶ καὶ ὁ πατὴρ αὐτῶν, καὶ ἐφοβήθησαν.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

86

NETS: Now it came about as they were emptying their sacks that then each one’s
bundle of money was in their sack. And they saw their bundles of money, they and
their father, and they were afraid.
Context: This attestation of ἐν τῷ + infinitive is found in the narrative about Ioseph and
his brothers. The brothers had just returned home to their father, Iakob, in the land of
Chanaan.
SS and IS: Structurally, ἐν τῷ κατακενοῦν αὐτοὺς τοὺς σάκκους αὐτῶν is an adjunct that
communicates the occasion in time (temporal scope) for its predicator ἦν (was), which must
be understood from the perspective of the brothers (in reality, the money was there the entire
time, not just when they emptied the sacks). Within the temporal boundary in which the
brothers emptied their sacks, each one discovered his money was there. In the narrative, it is
implicit that when the brothers' sacks (σάκκους) were filled with grain and their money
returned (42:25), those same sacks would be emptied after the brothers returned home. So, ἐν
τῷ κατακενοῦν αὐτοὺς τοὺς σάκκους (as they were emptying their sacks) is the presupposed
context in which the brothers would discover what is anticipated in the flow of the narrative:
the discovery of their money. This anticipation is heightened by the mention of one brother’s
discovery during the journey home in 42:26-28. He opened his bag (μάρσιππος)271 to feed the
donkeys and found to his surprise that his money had been returned. His brothers were
likewise startled. Readers anticipate that the same discovery will take place when the other
brothers empty their sacks. Therefore, the syntactic construction ἐν τῷ κατακενοῦν
communicates presuppositional information that has been grammaticalized using a textual
constituent marked for identifiability. The information is topical in 42:35.

271

There are three different Greek lexemes that were inconsistently used to render three different
Hebrew lexemes in this context: ἀγγεῖον renders ‫ ְכִּלי‬in 42:25; σάκκος renders ‫ ַשׂק‬in 42:25; μάρσιππος
renders ‫ ַשׂק‬in 42:27; μάρσιππος renders ‫ אְַמַ֫תַּחת‬in 42:27; μάρσιππος renders ‫ אְַמַ֫תַּחת‬in 42:28; σάκκος
renders ‫ ַשׂק‬in 42:35. An ἀγγεῖον is a type of container used for holding grain or fluids; σάκκος
generally refers to a type of heavy fabric from which various objects can be made; μάρσιππος is a bag
or pouch perhaps made of flax or hemp. Each of these had a different function with respect to the
cargo that was packed on a donkey. The translator was ostensibly aware of the different function of
each and therefore used the rendering that he felt best fit the context. It is possible that an ἀγγεῖον and
a μάρσιππος were packed within a σάκκος, which would cover and secure the cargo on the donkey.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

87

4.3.3. Genesis 44:31
‫ ָא ִ֛בינוּ ְבָּי ֖גוֹן ְשׁ ֽ ֹאָלה׃‬8֥ ‫ ֶאת־ֵשׂי ַ֨בת ַﬠְב ְדּ‬8‫ְוָהָ֗יה ִכּ ְראוֹת ֛וֹ ִכּי־ֵ֥אין ַה ַ֖נַּﬠר ָוֵ֑מת ְוהוֹ ִ֨רידוּ ֲﬠָב ֶ֜די‬
LXX: καὶ ἔσται ἐν τῷ ἰδεῖν αὐτὸν μὴ ὂν τὸ παιδάριον μεθʼ ἡμῶν τελευτήσει, καὶ
κατάξουσιν οἱ παῖδές σου τὸ γῆρας τοῦ παιδός σου πατρὸς δὲ ἡμῶν μετʼ ὀδύνης εἰς
ᾅδου.
NETS: then it shall be that when he sees the youngster is not with us, he will perish,
and your servants will bring down the old age of your servant, our father, with grief to
Hades.
Context: This text is part of the conversation between Ioseph and his brothers on their
second visit to Egypt. They had brought Beniamin back with them and refilled their grain
stocks. Then, they set out for Chanaan. However, the servant in charge of Ioseph's house
pursued them, claiming that they had stolen Ioseph's silver cup (which he had placed in
Beniamin's baggage). So, the brothers were brought back to Ioseph and told that Beniamin
must remain in Egypt as a slave for his supposed crime. Ioudas then spoke directly to Ioseph
and attempted to explain that Beniamin must not remain in Egypt, for their father Iakob
would surely die of grief if that were to happen.
SS and IS: Structurally, ἐν τῷ ἰδεῖν αὐτὸν is an adjunct that communicates the occasion
in time (conditional temporal cause) for its predicator τελευτήσει (he will perish). In other
words, if Iakob sees that Beniamin is not with his brothers, it will cause him to perish (see
Figure 2.4.1.6). The phrase ἐν τῷ ἰδεῖν αὐτὸν μὴ ὂν τὸ παιδάριον μεθʼ ἡμῶν (when he sees
the youngster is not with us) refers anaphorically to the previous sentence. In 44:30, Ioudas
states, "So now if I go in to your servant, our father, and the youngster be not with us"
(NETS). As presuppositional information in 44:31, the Greek phrase cited above has been
encoded using a textual constituent marked for identifiability. It is the pragmatic topic of the
sentence.
4.4. 'Εν τῷ + Infinitive (Rendering ‫ ְבּ‬+ Infinitive)
4.4.1. Genesis 19:33
‫ַו ַתּ ְשׁ ֶקין ָ ֶאת־ֲאִביֶהן ַי ִין ַבּ ַלּ ְיָלה הוּא ַו ָתּב ֹא ַהְבִּכיָרה ַו ִתּ ְשׁ ַכּב ֶאת־ָאִביָה ְול ֹא־ָיַדע ְבִּשְׁכָבהּ וְּבקוָּמהּ׃‬
LXX: ἐπότισαν δὲ τὸν πατέρα αὐτῶν οἶνον ἐν τῇ νυκτὶ ταύτῃ, καὶ εἰσελθοῦσα ἡ
πρεσβυτέρα ἐκοιμήθη μετὰ τοῦ πατρὸς αὐτῆς τὴν νύκτα ἐκείνην, καὶ οὐκ ᾔδει ἐν τῷ
κοιμηθῆναι αὐτὴν καὶ ἀναστῆναι.
NETS: And they gave their father wine to drink on this night, and the elder, when she
had gone in, lay with her father that night, and he did not know when she lay down
and got up.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

88

Context: This text is a part of the narrative describing Lot and his daughters in the
aftermath of the destruction of Sodoma and Gomorra. Fearing they would never find
husbands and produce offspring, his daughters inebriated Lot and had sexual relations with
him. This text specifically narrates the episode regarding the older daughter, which explains
the origin of the Moabites.
SS and IS: Structurally, ἐν τῷ κοιμηθῆναι αὐτὴν καὶ ἀναστῆναι is the complement of its
predicator, the finite verb ᾔδει. Semantically, it communicates the occasion in time (temporal
scope) for ᾔδει. In other words, it was within the temporal boundary of his daughter lying
down and getting up (and what transpired in between) that Lot lacked knowledge. The Greek
prepositional phrase cited above refers anaphorically to information that is freshly activated
in the previous clause in 19:33, "and the elder, when she had gone in, lay with her father that
night..." (NETS). The fact that she "lay with her father that night" implies that she lay down
and got up; the phrase also euphemistically alludes to sexual relations. So, the prepositional
phrase communicates presuppositional information that has been grammaticalized using a
textual constituent marked for identifiability.
4.4.2. Genesis 19:35
‫ַו ַתּ ְשׁ ֶקין ָ ַגּם ַבּ ַלּ ְיָלה ַההוּא ֶאת־ֲאִביֶהן ָי ִין ַו ָתּ ָקם ַה ְצִּﬠיָרה ַו ִתּ ְשׁ ַכּב ִﬠמּוֹ ְול ֹא־ָיַדע ְבִּשְׁכָבהּ וְּבֻקָמהּ׃‬
LXX: ἐπότισαν δὲ καὶ ἐν τῇ νυκτὶ ἐκείνῃ τὸν πατέρα αὐτῶν οἶνον, καὶ εἰσελθοῦσα ἡ
νεωτέρα ἐκοιμήθη μετὰ τοῦ πατρὸς αὐτῆς, καὶ οὐκ ᾔδει ἐν τῷ κοιμηθῆναι αὐτὴν καὶ
ἀναστῆναι.
NETS: And they gave their father wine to drink on that night also, and the younger,
when she had gone in, lay with her father, and he did not know when she lay down
and got up.
Context: This text is also a part of the narrative describing Lot and his daughters in the
aftermath of the destruction of Sodoma and Gomorra. Fearing they would never find
husbands and produce offspring, his daughters inebriated Lot and had sexual relations with
him. This text specifically narrates the episode regarding the younger daughter, which
explains the origin of the Ammanites.
SS and IS: Structurally, ἐν τῷ κοιμηθῆναι αὐτὴν καὶ ἀναστῆναι is the complement of its
predicator, the finite verb ᾔδει. Semantically, it communicates the occasion in time (temporal
scope) for ᾔδει, even though it is a complement. In other words, it was within the temporal
boundary of his daughter lying down and getting up (and what transpired in between) that

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

89

Lot lacked knowledge. The phrase ἐν τῷ κοιμηθῆναι αὐτὴν καὶ ἀναστῆναι (when she lay
down and got up) refers anaphorically to information that has just been activated in the
previous clause in 19:35, "and the younger, when she had gone in, lay with her father..."
(NETS). The fact that she "lay with her father" implies that she lay down and got up; the
phrase also euphemistically alludes to sexual relations. So, the Greek prepositional phrase
cited above communicates presuppositional information that has been grammaticalized using
a textual constituent marked for identifiability.
4.4.3. Genesis 28:6
‫ ִי ְצָחק ֶאת־ַיֲﬠֹקב ְו ִשׁ ַלּח ֹאתוֹ ַפּ ֶדָּנה ֲאָרם ָל ַקַחת־לוֹ ִמ ָשּׁם ִא ָשּׁה ְבָּבֲרכוֹ ֹאתוֹ ַו ְי ַצו ָﬠָליו ֵלא ֹמר‬0‫ַוַיּ ְרא ֵﬠָשׂו ִכּי־ֵבַר‬
‫ל ֹא־ִת ַקּח ִא ָשּׁה ִמְבּנוֹת ְכָּנַﬠן׃‬
LXX: Εἶδεν δὲ Ἠσαὺ ὅτι εὐλόγησεν Ἰσαὰκ τὸν Ἰακὼβ καὶ ἀπῴχετο εἰς τὴν
Μεσοποταμίαν Συρίας λαβεῖν ἑαυτῷ ἐκεῖθεν γυναῖκα ἐν τῷ εὐλογεῖν αὐτὸν καὶ
ἐνετείλατο αὐτῷ λέγων Οὐ λήμψῃ γυναῖκα ἀπὸ τῶν θυγατέρων Χανάαν,
NETS: Now Esau saw that Isaak blessed Iakob and that he was on his way to
Mesopotamia of Syria to take a wife from there for himself, that as he was blessing
him he also commanded him, saying, “You shall not take a wife from the daughters
of Chanaan...”
Context: This text is a part of the narrative concerning Esau and Iakob. As the text
indicates, Isaak had blessed Iakob and sent him on his way to Mesopotamia to find a wife.
SS and IS: Structurally, ἐν τῷ εὐλογεῖν is an adjunct that articulates the occasion in time
(temporal scope) for its predicator καὶ ἐνετείλατο (he also commanded). Within the temporal
boundary in which Isaak blessed Iakob, he issued him a command concerning a wife (see
Figure 2.4.1.2). The prepositional phrase ἐν τῷ εὐλογεῖν makes anaphoric reference to the
blessing mentioned directly after the ὅτι in 28:6: εὐλόγησεν Ἰσαὰκ τὸν Ἰακὼβ (Isaak blessed
Iakob). Therefore, the phrase communicates presuppositional information that has been
grammaticalized using a textual constituent marked as identifiable.
4.4.4. Genesis 32:20 (19)
‫ַו ְי ַצו ַגּם ֶאת־ַה ֵשּׁ ִני ַגּם ֶאת־ַה ְשִּׁלי ִשׁי ַגּם ֶאת־ ָכּל־ַה ֹהְלִכים ַאֲחֵרי ָהֲﬠָד ִרים ֵלא ֹמר ַכּ ָדָּבר ַה ֶזּה ְתַּדְבּרוּן ֶאל־ֵﬠָשׂו‬
‫ְבֹּמַצֲאֶכם ֹאתוֹ׃‬
LXX: καὶ ἐνετείλατο τῷ πρώτῳ καὶ τῷ δευτερῳ καὶ τῷ τρίτῳ καὶ πᾶσιν τοῖς
προπορευομένοις ὀπίσω τῶν ποιμνίων τούτων λέγων Κατὰ τὸ ῥῆμα τοῦτο λαλήσατε
Ἠσαὺ ἐν τῷ εὑρεῖν ὑμᾶς αὐτόν.
NETS: And he commanded the first and the second and the third and all who were
going on behind these flocks, saying, “Speak in accordance with this word to Esau
when you find him.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

90

Context: This text is a part of the narrative of Iakob's return to Chanaan when he met
with his brother Esau after many years. He sent several herdsmen ahead of him after learning
that Esau was on his way to meet him (32:6). The herdsmen were driving flocks of sheep,
rams, and goats, and herds of cows, bulls, camels, donkeys, and foals as a gift for Esau. Iakob
entrusted them with a message, explaining that all the flocks and herds were intended for
him.
SS and IS: Structurally, ἐν τῷ εὑρεῖν ὑμᾶς αὐτὸν is an adjunct that communicates the
occasion in time (temporal scope) for its predicator λαλήσατε (speak). Within the temporal
boundary having to do with when they find Esau, the servants are to speak the message given
by Iakob (see Figure 2.4.1.2). The Greek prepositional phrase cited above contains
information that is presuppositional not only to the readers of Genesis, but also to the
servants who hear Iakob's words. They are aware that they are being sent to bring a gift and a
message to Esau. Naturally, they anticipate finding him. So, the prepositional phrase refers
anaphorically to something in the text-external world (from the perspective of the hearers of
Iakob's message). It refers to their task of finding Esau and offering him some of Iakob's
flocks and herds. The translator of Genesis has grammaticalized this presuppositional
information using a textual constituent that is active in the discourse register and marked for
identifiability. It is a part of the utterance's topic.
4.4.5. Genesis 32:26 (25)
‫ ַיֲﬠֹקב ְבֵּהָאְבקוֹ ִﬠמּוֹ׃‬0‫ַוַיּ ְרא ִכּי ל ֹא ָי ֹכל לוֹ ַו ִיּ ַגּע ְבַּכף־ ְיֵרכוֹ ַו ֵתּ ַקע ַכּף־ֶיֶר‬
LXX: εἶδεν δὲ ὅτι οὐ δύναται πρὸς αὐτόν, καὶ ἥψατο τοῦ πλάτους τοῦ μηροῦ αὐτοῦ,
καὶ ἐνάρκησεν τὸ πλάτος τοῦ μηροῦ Ἰακὼβ ἐν τῷ παλαίειν αὐτὸν μετʼ αὐτοῦ·
NETS: And he saw that he was not powerful against him, and he touched the flat part
of his thigh, and the flat part of Iakob’s thigh became numb as he wrestled with him.
Context: This text is taken from the narrative of the wrestling match with Iakob.
SS and IS: Structurally, ἐν τῷ παλαίειν is an adjunct that modifies the finite verb
ἐνάρκησεν and articulates the occasion in time when the flat part of Iakob's thigh became
numb (ἐνάρκησεν τὸ πλάτος τοῦ μηροῦ Ἰακὼβ). This is a temporal causal semantic
relationship. The wrestling match with Iakob caused the numbness in his thigh (see Figure
2.4.1.4). The phrase ἐν τῷ παλαίειν refers anaphorically to the previous verse, where the
same verb is used in finite form to communicate newly asserted information: καὶ ἐπάλαιεν
ἄνθρωπος μετʼ αὐτοῦ ἕως πρωί (and a man was wrestling with him until morning).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

91

Therefore, when ἐν τῷ + infinitive is used in 32:26 (25), referring to information freshly
activated in the discourse register in the previous verse, it is both presuppositional in terms of
its ideational content and identifiable in terms of its text-linguistic form. The information
belongs to the topic domain.
4.4.6. Genesis 34:22
‫־ְבּז ֹאת ֵי ֹאתוּ ָלנוּ ָהֲאָנ ִשׁים ָל ֶשֶׁבת ִא ָתּנוּ ִלְהיוֹת ְלַﬠם ֶאָחד ְבִּהמּוֹל ָלנוּ ָכּל־ָזָכר ַכֲּא ֶשׁר ֵהם ִנֹמִּלים׃‬0‫ַא‬
LXX: μόνον ἐν τούτῳ ὁμοιωθήσονται ἡμῖν οἱ ἄνθρωποι τοῦ οἰκεῖν μεθʼ ἡμῶν ὥστε
εἶναι λαὸν ἕνα, ἐν τῷ περιτέμνεσθαι ἡμῶν πᾶν ἀρσενικόν, καθὰ καὶ αὐτοὶ
περιτέτμηνται.
NETS: Only in this will the people become like us to live with us so as to be one
people, when every male of ours is circumcised, as they also have been circumcised.
Context: After the incident involving Iakob's daughter Dina, who was defiled by
Sychem, Hemmor sought to arrange for the marriage of the two of them. Iakob's sons –
particularly Symeon and Leui – in their response gave the impression that this could happen,
and that they would settle and intermarry with the inhabitants of the land on the condition
that the Chorrite (Hivite) men be circumcised. The terms were pleasing to Hemmor and
Sychem, and Sychem quickly fulfilled his obligation because of his devotion to Dina. The
text cited above, then, contains part of the speech by Hemmor and Sychem to the other
Chorrite men in their city.
SS and IS: Structurally, ἐν τῷ περιτέμνεσθαι has the finite verb ὁμοιωθήσονται as its
predicator. The construction possesses a temporal causal relationship with its predicator; the
relationship could also be described as conditional. In other words, if the Chorrite men
fulfilled the condition of circumcision, it would cause Iakob's people to become one people
with them. In this conditional causal relationship, there is also a temporal sequence (see
Figure 2.4.1.6). So, the syntactic construction provides the occasion in time for its predicator
with the more specific semantic relationship of temporal cause set within a conditional
context.
The use of the phrase ἐν τῷ περιτέμνεσθαι is certainly presuppositional to readers who
are privy to the agreement between Iakob's sons, Hemmor, and Sychem (cf. v. 15). However,
it appears in a speech by Hemmor and Sychem to the other men in their city. This then
indicates that the translator also considers this information to be presuppositional to the
Chorrite men who heard the speech, even though they were not present when the agreement

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

92

was made. From Gen's perspective, it is not hard to assume that the Chorrite men would have
heard the news of Sychem's circumcision. The Chorrites, then, would have possessed this
information due to the realities of their text-external world, i.e., news spreading among them
by word of mouth. So, when Hemmor and Sychem spoke, the other Chorrites would have
cognitively identified the referent of ἐν τῷ περιτέμνεσθαι because they already knew what
Sychem had done. However, the location of ἐν τῷ περιτέμνεσθαι within the structure of the
text might give some pause for viewing it as presuppositional. After all, it is a clausal
constituent in what is an assertion about how the Hebrews would become like Chorrites and
dwell among them as one people. So, the construction appears to be in the focal position
rather than in the topical position, which is unusual.
Lambrecht deals with this type of issue. In clarifying precisely what the focus of an
utterance is, he states that, "The focus is that portion of a proposition which cannot be taken
for granted at the time of speech."272 It is perhaps most plausible to view the information
communicated by ἐν τῷ περιτέμνεσθαι as information that ought to be taken for granted,
since the Chorrites had heard about Sychem's circumcision to obtain a bride. However, who
would be circumcised so that the Hebrews and the Chorrites would become one people was
not taken for granted. Therefore, this verse does not answer the question, "What must happen
for the Hebrews to become like the Chorrites?" It answers the question, "Who must be
circumcised for the Hebrews to become like the Chorrites?" It is a subtle distinction, but it
fits best with the discourse pragmatics of how ἐν τῷ + infinitive is consistently used. The
only information, then, that might be new is what follows ἐν τῷ περιτέμνεσθαι: every male of
ours (ἡμῶν πᾶν ἀρσενικόν), just as they also have been circumcised (καθὰ καὶ αὐτοὶ
περιτέτμηνται).273 Lambrecht refers to this type of focus-structure as argument-focus
structure, which has as its core the identification of a referent.274 However, it is just as
possible that the Chorrites were aware of this stipulation, as well. If that is the case, then it
would not be new information.

272

Lambrecht, Information Structure and Sentence Form, 207.
Ibid., 209-10. These pages give a more detailed explanation of the same issue.
274
Ibid., 222ff.
273

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

93

4.4.7. Genesis 35:1
‫ ִמ ְפֵּני‬À‫ ְבָּב ְרֲח‬8‫ִהים ֶאל־ַיֲﬠֹקב קוּם ֲﬠֵלה ֵבית־ֵאל ְו ֶשׁב־ ָשׁם ַוֲﬠֵשׂה־ ָשׁם ִמְזֵבַּח ָלֵאל ַה ִנּ ְרֶאה ֵאֶלי‬E‫ַויּ ֹאֶמר ֱא‬
‫׃‬8‫ֵﬠָשׂו ָאִחי‬
LXX: Εἶπεν δὲ ὁ θεὸς πρὸς Ἰακώβ Ἀναστὰς ἀνάβηθι εἰς τὸν τόπον Βαιθήλ, καὶ οἴκει
ἐκεῖ, καὶ ποίησον ἐκεῖ θυσιαστήριον τῷ θεῷ τῷ ὀφθέντι σοι ἐν τῷ ἀποδιδράσκειν σε
ἀπὸ προσώπου Ἠσαὺ τοῦ ἀδελφοῦ σου.
NETS: Now God said to Iakob, “Arise, go up to the place Baithel, and reside there,
and make an altar there to the God who appeared to you when you were fleeing from
the presence of your brother Esau.”
Context: This text is a part of the Genesis narrative about Iakob. After Symeon and Leui
had slaughtered the Chorrite men, God once again spoke to Iakob and commanded him to
return to Baithel and make an altar (Gen. 35:1).
SS and IS: Structurally, ἐν τῷ ἀποδιδράσκειν σε is an adjunct that communicates the
occasion in time (temporal scope) for its predicator, the participle ὀφθέντι (appeared). It is
within the temporal boundary in which Iakob was fleeing that God appeared to him (see
Figure 2.4.1.2).
The referent for the Greek prepositional phrase was activated in the discourse register in
Genesis 28. Therefore, ἐν τῷ ἀποδιδράσκειν σε refers to a verbal antecedent that, from the
perspective of the readers, is no longer active but is accessible several chapters earlier. But
even though several chapters have already passed since the event that took place at Baithel,
the speech continues a recurrent theme in the conversational register between God and Iakob:
Iakob’s fear of Esau. Each conversation between God and Iakob before 35:1 is set against the
background of flight from Esau in fear (28:2ff) or journeying towards Esau in fear (32:333:3). Iakob’s prayer mentions his fear of being killed by Esau the last time he spoke with
God (32:11[12]), a fear that was understandable because of Esau's murderous threat that had
caused Iakob to flee in the first place (27:41-28:5). So, when ἐν τῷ ἀποδιδράσκειν σε is used
in 35:1, though its referent was several chapters prior, it picks up on the theme that has come
to the fore in every conversation between God and Iakob. Iakob’s character develops and
matures throughout the narrative, but his fear of his brother has continued with little change.
In Genesis 35:1, then, fleeing from Esau is still active in the conversational register of God
and Iakob. This would be akin to running into a friend at the grocery store and hearing about
the damage caused by their recent house fire. The topic of the house fire would still be active
in subsequent conversations with the friend, regardless of how much time had passed. In

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

94

Genesis 35:1, then, the use of the ἐν τῷ + infinitive construction demonstrates that the
translator was tracking conversations over several chapters and encoding IS based upon the
pragmatic consideration of the interlocutors involved.
According to the IS encoded in the conversation between God and Iakob, God expected
the information grammaticalized in the phrase ἐν τῷ ἀποδιδράσκειν σε to be taken for
granted by Iakob. As presuppositional information, it has been grammaticalized using a
textual constituent that is active in the discourse register and marked for identifiability. The
information is topical in the utterance.
4.4.8. Genesis 35:7
‫ִהים ְבָּב ְרחוֹ ִמ ְפֵּני ָאִחיו׃‬E‫ַו ִיֶּבן ָשׁם ִמְזֵבַּח ַו ִיּ ְקָרא ַל ָמּקוֹם ֵאל ֵבּית־ֵאל ִכּי ָשׁם ִנ ְגלוּ ֵאָליו ָה ֱא‬
LXX: καὶ ᾠκοδόμησεν ἐκεῖ θυσιαστήριον, καὶ ἐκάλεσεν τὸ ὄνομα τοῦ τόπου Βαιθήλ·
ἐκεῖ γὰρ ἐπεφάνη αὐτῷ ὁ θεὸς ἐν τῷ ἀποδιδράσκειν αὐτὸν ἀπὸ προσώπου Ἠσαὺ
τοῦ ἀδελφοῦ αὐτοῦ.
NETS: And there he built an altar and called the name of the place Baithel, for there
God had revealed himself to him when he was fleeing from the presence of his
brother Esau.
Context: This text is also part of the narrative in Genesis about Iakob. Genesis 35:7
records the event of Iakob returning and building the altar. He built it there because that is
where God had revealed himself when Iakob was fleeing (ἐν τῷ ἀποδιδράσκειν αὐτὸν) from
Esau. This refers to the same event as that described in Genesis 35:1.
SS and IS: Ἐν τῷ ἀποδιδράσκειν is an adjunct that modifies its predicator ἐπεφάνη and
articulates its occasion in time (temporal scope). In other words, God revealed himself to
Iakob within the temporal boundary of when he was fleeing from Esau. The prepositional
phrase’s use in Genesis 35:7 is understandable because, from the reader's perspective, its
referent had just been reactivated in the discourse register in 35:1. So, it possessed an active
status in 35:7, making its usage appropriate. The phrase ἐν τῷ ἀποδιδράσκειν αὐτὸν, then,
refers to presuppositional information that has been grammaticalized using a textual
constituent that is active in the discourse register and therefore marked for identifiability. It is
topical, as well.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

4.5.

95

'Εν τῷ + Infinitive (Rendering Other Hebrew Constructions)

4.5.1. Genesis 19:16
‫ ְיה ָ֖וה ָﬠ ָ֑ליו ַו ֹיּ ִצ ֻ֥אהוּ ַוַיּ ִנּ ֻ֖חהוּ‬275‫ַֽו ִיְּתַמְה ָ֓מהּ׀ ַוַיֲּח ִ֨זקוּ ָהֲאָנ ֜ ִשׁים ְבָּי ֣ ֹדו וְּבַיד־ִא ְשׁתּ ֗וֹ וְּבַי ֙ד ְשׁ ֵ֣תּי ְב ֹנ ָ֔תיו ְבֶּחְמ ַ֥לת‬
‫ִמ֥חוּץ ָל ִֽﬠיר׃‬
LXX: καὶ ἐταράχθησαν· καὶ ἐκράτησαν οἱ ἄγγελοι τῆς χειρὸς αὐτοῦ καὶ τῆς χειρὸς
τῆς γυναικὸς αὐτοῦ καὶ τῶν χειρῶν τῶν δύο θυγατέρων αὐτοῦ, ἐν τῷ φείσασθαι
κύριον αὐτοῦ.
NETS: And they were confounded, and the angels seized his hand and his wife’s
hand and his two daughters’ hands, as the Lord spared him.
Context: This text is a part of the narrative about the destruction of Sodoma and
Gomorra. In the text's background is the discussion that took place between Abraam and God
in Genesis 18. Abraam pleaded with the Lord to spare the cities of Sodoma and Gomorra if a
certain number of righteous people were found in the cities. In the end, God agreed not to
destroy the cities if as few as ten righteous people could be found. In chapter nineteen, Lot's
reception of the angels contrasts with the Sodomites’ reception. Lot did obeisance before
them, welcomed them into his home, and prepared a feast. The Sodomites wanted to rape
them. Privy to the conversation between Abraam and God in chapter eighteen, readers take it
for granted that the Lord intended to spare Lot and destroy the Sodomites. Furthermore, the
angels’ words in 19:15 confirm that assumption: "Rise, take your wife and the two daughters
whom you have, and get out, lest you also be destroyed together with the lawlessness of the
city" (NETS).
SS and IS: Structurally, like other attestations of ἐν τῷ + infinitive, the prepositional
phrase is an adjunct that communicates the occasion in time for its predicator ἐκράτησαν
(seized). It is debatable whether it more specifically communicates temporal scope or
temporal cause. But the perfective verbal aspect of the infinitive (if it is an infinitive),
perhaps leads one to view the phrase as communicating the temporal scope of its predicator
(see Figure 2.4.1.2). Due to information that ought to be taken for granted as the narrative
unfolds, ἐν τῷ φείσασθαι should be viewed as encoding presuppositional information that is
already active in the discourse register. The construction most immediately makes anaphoric

275

This may not be an infinitive; if it is not, then Gen’s usage of ἐν τῷ + infinitive may possibly have
been because the Hebrew word appeared to him as an infinitive and not a noun. The word ‫ ֶחְמָלה‬is
only used 2x in the Tanakh (Genesis 19:16; Isaiah 63:9).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

96

reference to the angels' speech in 19:15, which demonstrates the Lord's intention to spare Lot
and his family. Therefore, the phrase has been grammaticalized using a textual constituent
marked for identifiability.
4.5.2. Genesis 34:15
‫־ְבּ ֖ז ֹאת ֵנ ֣אוֹת ָלֶ֑כם ִ֚אם ִתְּה ֣יוּ ָכ ֔ ֹמנוּ ְלִהֹ֥מּל ָלֶ֖כם ָכּל־ָז ָ ֽכר׃‬0‫ַא‬
LXX: ἐν τούτῳ ὁμοιωθησόμεθα ὑμῖν καὶ κατοικήσομεν ἐν ὑμῖν, ἐὰν γένησθε ὡς
ἡμεῖς καὶ ὑμεῖς ἐν τῷ περιτμηθῆναι ὑμῶν πᾶν ἀρσενικόν,
NETS: In this we will become like you and settle among you, if you, even you,
become like us when every male of yours has been circumcised,
Context: After the incident in which Iakob's daughter Dina was defiled by Sychem,
Hemmor sought to arrange a marriage between them. Iakob's sons – particularly Symeon and
Leui – agree to give Dina in marriage, settle among the Chorrites, and intermarry with the
inhabitants of the land on the condition that the Chorrite (Hivite) men be circumcised. This
text is taken from Symeon and Leui's speech to Hemmor and Sychem.
SS and IS: Structurally, ἐν τῷ περιτμηθῆναι ὑμῶν πᾶν ἀρσενικόν276 is an adjunct that
communicates the occasion in time (temporal cause) for its predicator γένησθε (become). In
other words, when the circumcision occurs for every male, it would cause the Chorrites to
become like the Hebrews (see Figure 2.4.1.6). The information encoded by ἐν τῷ
περιτμηθῆναι ὑμῶν πᾶν ἀρσενικόν is anticipated by Symeon and Leui's words in the
preceding verse, "We will not be able to carry out this matter, to give our sister to a man who
has a foreskin, for it is a disgrace to us" (NETS). This information should be viewed as active
in the discourse register (as a part of the schema activated by refusing to give their sister to a
man with a foreskin) and presuppositional. Therefore, it has been grammaticalized using a
textual constituent marked for identifiability.
Similar to the text in §4.4.6, on the surface, the syntactic construction appears to be
newly asserted information. However, it must not be viewed in that manner. As it was stated
above, Lambrecht deals with this type of issue. In clarifying precisely what the focus of an
utterance is, he states that, "The focus is that portion of a proposition which cannot be taken

276

The Greek expression here could be harmonized to the Hebrew of the parallel statement in 34:22
(which uses ‫ ְבּ‬+ Infinitive). The Hebrew of the current verse is an epexegetical lamed plus infinitive,
that may be understood in terms of the English gerund “by being circumcised.”

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

97

for granted at the time of speech."277 It is perhaps most plausible to view the information
communicated by ἐν τῷ περιτέμνεσθαι as information that ought to be taken for granted since
Hemmor and Sychem had just heard that Dina would not be given to an uncircumcised man.
The new state of information, then, is that the Hebrews will become like the Chorrites and
settle among them if the Chorrites become like them. This new state of information is
cognitively anchored to what is presuppositional, i.e., the topic of circumcision that was
implicitly activated in the discourse register in 34:14. However, who would be circumcised
so that the Hebrews and the Chorrites would become one people was not taken for granted.
Therefore, like the discussion of the text in §4.4.6, this text does not answer the question,
"What must happen for the Hebrews to become like the Chorrites." It answers the question,
"Who must be circumcised for the Hebrews to become like the Chorrites." The assertion that
brought a new state of information is that it was not just Sychem in taking Dina as a wife, but
it is every male among them. It is a subtle distinction, but it fits best with the discourse
pragmatics of how ἐν τῷ + infinitive is consistently used. Lambrecht refers to this type of
focus-structure as argument-focus structure, which has as its core the identification of a
referent.278
4.6. Summary
Structurally, ἐν τῷ + infinitive is most typically an adjunct in LXX Genesis. However,
the construction is a complement in two passages (19:33 and 19:35). Semantically, though, it
always indicates occasion in time for its predicator. More specific semantic nuances are
dependent on the lexemes used in different contexts. Sometimes the nuance is causal,
sometimes limitative (providing temporal scope). At times, the contexts are reporting events,
while at other times, they are conditional. Pragmatically, though, the ἐν τῷ + infinitive
construction always communicates presuppositional information that is marked for
identifiability with the article. This usage is consistent with that in the compositional Greek
texts discussed in chapter three.

277
278

Lambrecht, Information Structure and Sentence Form, 207.
Ibid., 222ff.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

98

CHAPTER 5: ANALYSIS OF OTHER GREEK RENDERINGS
"[H]is linguistic skills are demonstrated through his semantic differentiation and his ability to
use a variety of Greek terms or expressions depending on contextual demands."279
– Mark W. Scarlatta
5.1. Introduction
In the introduction of this thesis, it was stated that Septuagint Greek should be analyzed
in relation to contemporary PC Greek and the underlying Hebrew source text. This task
requires investigation of Greek compositional corpora and the LXX’s Hebrew source text,
particularly regarding how each Hebrew construction was rendered using various Greek
lexemes and phrases in ways that made sense to the translator. Mark Scarlatta remarks,
"Although the translator of Genesis closely adhered to the Hebrew text, his linguistic skills
are demonstrated through his semantic differentiation and his ability to use a variety of Greek
terms or expressions depending on contextual demands."280 The goal in chapter three was to
explain how ἐν τῷ + infinitive was used in PC compositional corpora contemporaneous with
the LXX. The comparison of those texts with LXX Genesis demonstrates that Gen used ἐν
τῷ + infinitive with the same discourse pragmatic function and structural semantics. He used
a natural Greek syntactic construction in a manner consistent with its conventional usage.
The renderings in LXX Genesis routinely pay attention to what is taking place in the
narrative beyond the level of the sentence, even tracking a conversation over several chapters
to give a rendering that best communicates the information status of a certain constituent that
is still active in the discourse register from the perspective of the conversation's interlocutors.
This is perhaps one indication that Gen considered the broader discourse context when
rendering the Hebrew text.
There were several other renderings that Gen used for the Hebrew ‫ ְבּ‬+ infinitive
construction. It is beyond the scope of this chapter to analyze every one of those texts.
However, having some understanding of the way those other Greek renderings function in
the Greek linguistic system can also further reinforce what the textual data has shown about

279

Mark W. Scarlatta, "Genesis," in T&T Clark Companion to the Septuagint, ed. James Aitken
(London: Bloomsbury T&T Clark, 2015), 13-28, here 16-17.
280
Ibid., 16-17.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

99

ἐν τῷ + infinitive. This opens the window wider into the translation technique used in LXX
Genesis. The reason why one equivalent was used to the exclusion of others, as mentioned
earlier, is certainly multidimensional and involves several factors.281 It is impossible to
completely know the rationale behind every rendering. However, in this thesis, that issue is
addressed from the perspective of IS analysis. From that perspective, choosing one rendering
and thus not choosing another would have been determined by discourse pragmatic concerns.
It would not have been an arbitrary decision. Even though ἐν τῷ + infinitive exhibits close
formal correspondence with ‫ ְבּ‬+ infinitive, there would have been compelling reasons in
certain contexts to choose particular renderings because of the discourse context. Therefore,
knowing how the other translation equivalents function will also further clarify the
differences between ἐν τῷ + infinitive and ‫ ְבּ‬+ infinitive. The Hebrew construction was more
flexible and less marked from a discourse pragmatic perspective.
Simon Dik has stated, "Whenever there is some overt difference between two
constructions X and Y, start out on the assumption that this difference has some kind of
functionality in the linguistic system."282 Cross-linguistically, grammatical structure provides
various ways to communicate the same ideational content. However, structures vary
depending on discourse pragmatic needs. In languages like AG, in which syntax constitutes
one means of encoding IS, different syntactic constructions often instantiate different
pragmatic functions. This was true of the renderings that were used to translate the Hebrew ‫ְבּ‬
+ infinitive construction, which occurs 45x in Genesis. The frequencies of the various
renderings used to translate the Hebrew construction are depicted in the table below.

281

Dhont, "Septuagint Translation Technique and Jewish Hellenistic Exegesis," 24.
Simon C. Dik, The Theory of Functional Grammar, Part 1: The Structure of the Clause, 2nd ed.,
ed. Kees Hengeveld (Berlin: Mouton de Gruyter, 1997), 18.
282

100

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

Table 5.1: Frequency of Renderings for ‫ ְבּ‬+ infinitive in LXX Genesis
0

2

4

6

8

10

12

14

16

18

20

finite verb
noun
ἐν τῷ + noun
παρα τὸ + infinitive
adverb
participle
ἐν τῷ + infinitive
temporal conjunct. + finite verb

Within AG as a linguistic system, there were choices about the utilization of linguistic
forms based on how a language producer wanted to construe the pragmatic salience of a
constituent. Pragmatic salience is unrelated to inherent salience but instead depends on what
the language producer desires to accentuate or not. As noted in the introduction to this thesis,
accentuation is distinct from focus, although there is some relationship. An utterance's focus,
by default, possesses pragmatic salience and is therefore typically grammaticalized using
certain linguistic forms that are more marked for accentuation. However, topical constituents
can also be construed with greater or lesser degrees of pragmatic salience according to the
way grammatical forms function within the linguistic system. This means that certain forms
marked for accentuation can be utilized to communicate the same ideational content as other
forms that are unmarked for accentuation. Katja Hetterle has argued that adverbial clauses
"in a language can show different degrees of downgrading," so that there is "a continuum
approach to subordination."283 She argues that adverbial phrases and clauses created using
participles and infinitives, for example, are more downgraded than constructions using finite
verbs.284 The more downgraded a construction, the less accentuation or pragmatic salience it
would possess. This is another area in which syntax and IS intersect. Different choices within
the linguistic system are found within what Hetterle labels a "downgrading hierarchy."285
283

Katja Hetterle, Adverbial Clauses in Cross-Linguistic Perspective (Berlin: De Gruyter Mouton,
2015), 155.
284
Ibid., 155. She views nominalized forms has having the highest degree of downgrading (170).
Therefore, they would have the least amount of pragmatic salience unless located within an
utterance's focus.
285
Ibid., 147-196. She discusses the downgrading hierarchy in the entire chapter.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

101

This hierarchy can be viewed as conceptualizing what linguistic forms are used for the most
pragmatically salient information and what forms are used for the least pragmatically salient
information. Information that is active in the discourse register is typically grammaticalized
with no accentuation because it is freshly lit up in a listener's consciousness and therefore
requires less salient linguistic marking to be cognitively recalled. However, a language
producer can choose to grammaticalize an active constituent using an accentuated form if
there is a desire to give that constituent greater pragmatic salience. This often happens for
states of affairs that possess inherent salience, even though the information is already active
in the discourse register. Often, these states of affairs are communicated at the beginning of a
discourse unit in presentational sentences or at the end of a discourse unit in a summary
statement. So, the inherent salience and position in the discourse make it appropriate to use a
linguistic form that is marked for elevated pragmatic salience.
For Gen, choosing to use one rendering as opposed to others within the downgrading
hierarchy was a choice to present the constituent with greater or lesser degrees of pragmatic
salience. It is difficult to arrange the renderings in Figure 5.1 in a strict hierarchy since not
all the forms can function structurally in the same manner. The following, though, is an
attempt to construct a downgrading hierarchy of finite verbs, non-finite verbs, and nominals
that were used as renderings for ‫ ְבּ‬+ infinitive, starting with those that might be used for the
most pragmatically salient information and then proceeding to those that might be used for
the least pragmatically salient information: 1) finite verb, 2) temporal conjunction + finite
verb, 3) participle, 4a) ἐν τῷ + infinitive,286 4b) παρὰ τὸ + infinitive, 4c) adverb, 5) ἐν τῷ +
noun, 6) noun. Representative examples of each translation equivalent will be discussed
below.

286

Using Hetterle's insight – cited above – that there is "a continuum approach to subordination," I
regard ἐν τῷ + infinitive to be more downgraded than a participle. I base this conclusion on the data
in chapter three where five texts with ἐν τῷ + infinitive subordinated to a participle (Polybius, The
Histories 2.29.3, 2.32.10, 2.79.10, 4.64.7, P.Koeln 8.327 text 2). The same, therefore, would be true
of every preposition + article + infinitive such as παρὰ τὸ + infinitive. I also conclude – and this is
less relevant for this chapter – that ἐν τῷ + infinitive is more downgraded than a simple infinitive
since the textual data also show that this construction can be subordinated to an infinitive (Polybius,
The Histories 3.89.6; I. Prose 30).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

102

5.2. Discourse Pragmatic Function of Other Greek Renderings
5.2.1. Finite Verb
(1) Genesis 42:15
‫ְבּ ֖ז ֹאת ִתָּבֵּ֑חנוּ ֵ֤חי ַפ ְרֹע ֙ה ִאם־ ֵתּ ְצ ֣אוּ ִמ ֔ ֶזּה ִ֧כּי ִאם־ְבּ֛בוֹא ֲאִחיֶ֥כם ַה ָקֹּ֖טן ֵ ֽה ָנּה׃‬
LXX: ἐν τούτῳ φανεῖσθε· νὴ τὴν ὑγίειαν Φαραώ, οὐ μὴ ἐξέλθητε ἐντεῦθεν, ἐὰν μὴ ὁ
ἀδελφὸς ὑμῶν ὁ νεώτερος ἔλθῃ ὧδε.
NETS: In this you shall manifest yourselves: by the health of Pharao, you shall not
depart from this place unless your younger brother comes here!
This is the only place in LXX Genesis where a finite verb is used to render ‫ ְבּ‬+ infinitive.
In the context, Ioseph was speaking to his brothers when they first came to buy grain in
Egypt. They had informed him of their father's health and their younger brother Beniamin.
Ioseph then accused them of being spies as a pretense for demanding that Beniamin be
brought to Egypt. The information grammaticalized by ἔλθῃ possesses a different pragmatic
status than what was seen previously in texts with ἐν τῷ + infinitive. The verb communicates
newly asserted information, and therefore, it was fitting to use a linguistic form that was
marked for accentuation. Runge has stated, "The finite action is the most prominent one."287
Therefore, newly asserted information is often communicated using a finite verb. Newly
asserted information can also be grammaticalized using non-finite forms, however, the
information possesses more accentuation when a finite verb is used. As newly asserted
information in a conditional clause, the finite verb is a part of the clause's focus, what
Lambrecht calls predicate-focus structure.288
5.2.2. Temporal Conjunction + Finite Verb
As Figure 5.1 indicates, a temporal conjunction + finite verb was used 19x as a rendering
for ‫ ְבּ‬+ infinitive in LXX Genesis. The translator employed either ὅτε + finite verb or ἡνίκα +
finite verb with no apparent distinction in choice between the two temporal conjunctions.
Therefore, this syntactic construction was the most frequent rendering, being used 1x more
than ἐν τῷ + infinitive. There are clear similarities in the structural semantics of a temporal
conjunction + finite verb and ἐν τῷ + infinitive, and there are also clear differences in the
way the two syntactic constructions function pragmatically. Like the example above
287

Runge, Discourse Grammar of the Greek New Testament, 245.
Lambrecht, Information Structure and Sentence Form, 222. Lambrecht states that the predicatefocus structure is the default focus-structure.
288

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

103

involving a finite verb, the choice of a temporal conjunction + finite verb also signals
elevated pragmatic salience, and thus accentuation. From the perspective of IS, there are
some plausible reasons for going with this option to signal elevated pragmatic salience. One
reason is that the construction is appropriate to use when grammaticalizing newly asserted
information. Its use in passages like Gen. 42:15 discussed above is fitting because the newly
asserted information is grammaticalized utilizing a form appropriate to its discourse
pragmatic status. There are four places in LXX Genesis (Gen. 30:42; 36:24; 42:21; 45:1) in
which a temporal conjunction + finite verb is employed, ostensibly for this reason. Consider
the following passage, taken from Esau's genealogy in Genesis 36, as a good example of how
the construction grammaticalizes newly asserted information:
(2) Genesis 36:24
‫ְוֵ֥א ֶלּה ְב ֵֽני־ ִצְב ֖ﬠוֹן ְוַא ָ֣יּה ַוֲﬠ ָ֑נה ֣הוּא ֲﬠָ֗נה ֲא ֨ ֶשׁר ָמ ָ֤צא ֶאת־ַהֵיִּמ ֙ם ַבּ ִמְּד ָ֔בּר ִבּ ְרֹעת ֥וֹ ֶאת־ַהֲח ֹמ ִ֖רים ְל ִצְב֥ﬠוֹן ָא ִֽביו׃‬
LXX: καὶ οὗτοι υἱοὶ Σεβεγών· Ἀιὲ καὶ Ὠνάν· οὗτός ἐστιν ὁ Ὠνᾶς, ὃς εὗρεν τὸν Ἰαμὶν
ἐν τῇ ἐρήμῳ, ὅτε ἔνεμεν τὰ ὑποζύγια Σεβεγὼν τοῦ πατρὸς αὐτοῦ.
NETS: And these are the sons of Sebegon: Aie and Onan; this is the Onas who found
Iamin in the wilderness, when he was pasturing the draft animals of his father
Sebegon.
Most of the sons mentioned in the genealogy have no role in the unfolding narrative in
Genesis. The names and relevant details about them are only mentioned here. Therefore, the
information communicated about them is generally newly asserted. The syntactic
construction ὅτε ἔνεμεν renders ‫ ְבּ‬+ infinitive (‫ )ִבּ ְרֹעת ֥וֹ‬and is likewise newly asserted. It gives
important information about the text's focal constituent—that is, it provides the occasion in
time for its predicator εὗρεν (the same structural semantics as for ἐν τῷ + infinitive). It has
been rendered, though, using a syntactic construction that makes its elevated pragmatic
salience apparent as newly asserted information.
A second reason for choosing a temporal conjunction + finite verb to signal elevated
pragmatic salience concerns the state of affairs being communicated. As mentioned above,
information that is active in the discourse register is typically grammaticalized with no
accentuation because it is freshly lit up in a listener's consciousness. Therefore, the
information requires less salient linguistic marking to be cognitively recalled. A language
producer, though, can choose to encode an active constituent using an accentuated form if
there is a desire to give that constituent greater pragmatic salience. To use the metalanguage

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

104

of IS analysis, this has less to do with the speaker's perception of the listener's cognitive state
(such as presupposition and assertion) and more to do with what the speaker wants the
listener to perceive as important. As mentioned previously, this is not only done with focus
constituents but with topic constituents, as well.289 This often happens for states of affairs
that possess inherent salience – and therefore greater importance within the overall flow of a
discourse – even though the information may already be active in the discourse register. As it
will be seen below, these states of affairs are routinely communicated at discourse
boundaries, either at the beginning of a discourse unit in a presentational sentence or at the
end of a discourse unit in a summary statement. In the employment of a temporal conjunction
+ finite verb in LXX Genesis, the states of affairs occupy these positions within discourse
units because they provide inherently salient temporal information about the entire discourse
unit.
Table 5.2.2: Temporal conjunction + Finite Verb Rendering Salient States of Affairs
Reference
Gen. 2:4
Gen. 12:4
Gen. 16:16
Gen. 17:24
Gen. 17:25
Gen. 21:5
Gen. 25:20
Gen. 25:26
Gen. 33:18
Gen. 34:25
Gen. 35:9

289

State of Affairs
(When + ...)
creation originated
movement from
Mesopotamia
Hagar bore Ismael
Abraam circumcised
himself
Abraam circumcised
Ismael
Isaak was born
Isaak took Rebekka
Rebekka bore sons
movement from
Mesopotamia
Chorrites were in
circumcision pain
arrival from
Mesopotamia

Predicator + Arguments
"This is the book of the
origin of heaven and
earth"
"Abram was seventy-five
years"
"Abram was eighty-six
years"
"Abraam was ninety-nine
years"
"his son Ismael was
thirteen years"
"Abraam was a hundred
years"
"Isaak was forty years"
"Isaak was sixty years"
"Iakob came to Salem"
“Symeon and Leui took
each one his dagger and
entered into the city”
"God appeared to Iakob"

Lambrecht, Information Structure and Sentence Form, 325.

Position in
Discourse Unit
beginning
beginning
end
end
end
middle
beginning
end
beginning
beginning
beginning

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

Gen. 35:22
Gen. 38:5
Gen. 41:46
Gen. 48:7

Israel dwelt in the
land
Ioudas's wife bore
sons
Ioseph stood before
Pharao
movement from
Mesopotamia

"Rouben went and lay
with Balla"
"she was in Chasbi"

beginning

"Ioseph was thirty years"

beginning

"your mother Rachel
died"

end

105

end

Additionally, when the information in these states of affairs is communicated using a
temporal conjunction + finite verb as a rendering for ‫ ְבּ‬+ infinitive, the Genesis texts indicate
that the information does nothing to move the narrative forward within the internal sequence
of events. Levinsohn notes that this type of information creates an "action discontinuity" in
narrative.290 This is not unusual for information that is presuppositional, as the previous
analysis of ἐν τῷ + infinitive has demonstrated. In a sense, the information functions as a sort
of temporal topic announcement, what perhaps might be labelled the discourse unit's
temporal frame topic,291 locating other events (some or all) in the discourse unit within its
temporal frame of reference. The data seems to indicate, then, that there is a distinction in the
usage of a temporal conjunction + finite verb and ἐν τῷ + infinitive. Whereas ἐν τῷ +
infinitive communicates less pragmatically salient information that provides the occasion in
time for its predicator as one event in the internal sequence of a discourse unit, a temporal
conjunction + finite verb (grammaticalizing an inherently salient state of affairs)
communicates the occasion in time for a sequence of events in a discourse unit. Therefore,
the information plays a much bigger role pragmatically due to its importance within the flow
of the discourse. A close examination of the texts in LXX Genesis will confirm this
generalization about the discourse function of a temporal conjunction + finite verb compared
to the ἐν τῷ + infinitive construction. Consider the following textual example:
(3) Genesis 33:18
‫ַוָיּב ֹ ֩א ַיֲﬠֹ֨קב ָשׁ ֵ֜לם ִ֣ﬠיר ְשֶׁ֗כם ֲא ֶשׁ ֙ר ְבֶּ֣אֶרץ ְכּ ַ֔נַﬠן ְבֹּבֹ֖או ִמ ַפּ ַ֣דּן ֲא ָ֑רם ַו ִ֖יַּחן ֶאת־ ְפּ ֵ֥ני ָה ִֽﬠיר׃‬
290

Levinsohn, Discourse Features of New Testament Greek, 188. He writes, "...action discontinuities
are to be discerned when a sentence describes an event that fails to move the narrative forward to the
next action in sequence. This happens when simultaneous events or restatements are involved."
291
The term "frame topic" is taken from Daniel Büring, "(Contrastive) Topic," in The Oxford
Handbook of Information Structure, eds. Caroline Féry and Shinichiro Ishihara (Oxford: Oxford
University, 2016), 64-85, here 84.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

106

LXX: καὶ ἦλθεν Ἰακὼβ εἰς Σαλὴμ πόλιν Σικίμων, ἥ ἐστιν ἐν γῇ Χανάαν, ὅτε ἦλθεν
ἐκ τῆς Μεσοποταμίας Συρίας, καὶ παρενέβαλεν κατὰ πρόσωπον τῆς πόλεως.
NETS: And Iakob came to Salem, a city of Sikima, which is in the land of Chanaan,
when he came from Mesopotamia of Syria, and he encamped facing the city.
This text is taken from a short discourse unit in Genesis 33:18-20. Although ὅτε ἦλθεν
("when he came") structurally modifies the predicator ἦλθεν at the beginning of 33:18,
semantically all the other predicators in the discourse unit also happen within its temporal
scope. The entire unit states, "And Iakob came to Salem, a city of Sikima, which is in the
land of Chanaan, when he came [ὅτε ἦλθεν] from Mesopotamia of Syria, and he encamped
facing the city. And from Hemmora, Sychem’s father, he acquired for one hundred lambs the
portion of the field, there where he had set up his tent, and there he set up an altar and
invoked the God of Israel" (33:18-20). So, the sequence of events in the discourse unit –
Iakob came to Salem, encamped, acquired the portion of the field, set up an altar, and
invoked God – all occur within the temporal scope of ὅτε ἦλθεν ἐκ τῆς Μεσοποταμίας
("when he came from Mesopotamia"). It is appropriate that states of affairs that are more
inherently salient, such as Iakob's movement from Mesopotamia, would be encoded using a
syntactic structure that is more pragmatically salient, accentuating its pragmatic status to fit
with its significance in the narrative.
5.2.3. Participle
LXX Genesis contains three texts in which a participle is used as a rendering for ‫ ְבּ‬+
infinitive (Gen. 27:5; 30:38; 50:17). These texts do not use the participle as a circumstantial
frame on the left periphery, a common occurrence in compositional Greek. Gen could have
employed participles in this manner when translating ‫ ְבּ‬+ infinitive so long as fidelity to the
Hebrew word order allowed for that, but Gen instead chose a temporal conjunction + finite
verb or ἐν τῷ + infinitive to create a frame on the left periphery. All three texts that use a
participle employ it in the post-nuclear position, following the Hebrew Vorlage's word order.
While this does not affect semantics, there is a pragmatic difference between pre-nuclear and
post-nuclear participle positions. Levinsohn argues that the pre-nuclear anarthrous participle
always grammaticalizes background information. However, this is not necessarily true for the

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

107

post-nuclear anarthrous participle,292 especially when these participles illuminate an
additional aspect of the nuclear state of affairs or communicate newly asserted information
connected to the nuclear state of affairs. Pragmatically, then, Levinsohn argues that there is
continuity of situation between the nuclear state of affairs and the state of affairs
grammaticalized by the participle.293
It needs to be reiterated that the same ideational content that could be communicated
with ἐν τῷ + infinitive can also be communicated using a participle. Apart from any
pragmatic considerations, ἐν τῷ + infinitive could be replaced with a participle in every text
in chapters two and three without any effect on semantics. However, ἐν τῷ + infinitive is
syntactically marked as an identifiable textual constituent. Anarthrous participles, in contrast,
are not syntactically marked as identifiable textual constituents, therefore participles can
either communicate presuppositional information or newly asserted information.
Pragmatically, they are more flexible. However, none of the passages in LXX Genesis in
which a participle is used as a rendering for ‫ ְבּ‬+ infinitive communicate newly asserted
information.
Furthermore, this thesis argues that ἐν τῷ + infinitive makes explicit a temporal
relationship with its predicator (whether considered temporal scope or temporal cause),
whereas an anarthrous participle only grammaticalizes circumstantial information about its
predicator. Any conclusion concerning a participle's semantic relationship with its predicator
as indicating attendant circumstance, cause, concession, condition, manner, means, purpose,
or time must be concluded from the context.294 Participles, then, are also more flexible
semantically compared to the ἐν τῷ + infinitive construction. Every semantic relationship
that can be grammaticalized using various Greek prepositions + article + infinitive can be
grammaticalized using a participle. Preposition + article + infinitive grammaticalizes a more
specific semantic relationship that is determined by the semantics of the preposition and the
lexical semantics of the infinitive and its predicator. But participles are more semantically
dependent on context. Therefore, the semantic relationship between participle and predicator
292

Levinsohn, Discourse Features of New Testament Greek, 181.
Ibid., 13, fn 10.
294
Robert W. Funk, A Beginning-Intermediate Grammar of Hellenistic Greek. Part II: Syntax
(Atlanta: Scholars Press, 1973), 669.
293

108

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

is determined by context and the lexical semantics of participle and predicator. Albert
Rijksbaron says concerning participles in adjunctive phrases, "The interpretation of such
participle constructions is determined by the context and by the semantic characteristics of
the states of affairs involved. Some of these various semantic functions may be made explicit
by means of adverbs occurring either with the participle or with the main verb."295 Consider
the following passages in which a participle is used to translate the Hebrew ‫ ְבּ‬+ infinitive in
LXX Genesis:
(4) Genesis 50:17
‫ַו ֵ֥יְּב ְךּ יוֵֹ֖סף ְבַּדְבּ ָ֥רם ֵא ָ ֽליו׃‬
LXX: καὶ ἔκλαυσεν Ἰωσὴφ λαλούντων αὐτῶν πρὸς αὐτόν.
NETS: And Ioseph wept as they were speaking to him.
The participle in this verse is the head of a genitive absolute construction. It contains
presuppositional information that is already active in the discourse register, as Ioseph's
brothers' speech has been reported in the preceding verses. As mentioned above, the
construction is unmarked syntactically for identifiability. So, the fact that λαλούντων
grammaticalizes an identifiable textual constituent is only apparent from the context. The
NETS translation inserts "as" to plausibly render the semantic relationship296 between the
predicator ἔκλαυσεν and λαλούντων. This indicates temporal scope. In other words, it is
within the temporal boundary of their speech to him that he wept. Ioseph's weeping
ostensibly continued after they stopped speaking. It might be rendered, "And Ioseph wept
because they were speaking to him." But that would be less accurate since Ioseph wept not
because they were speaking to him, but because of their speech's content. Regardless,
λαλούντων αὐτῶν πρὸς αὐτόν gives circumstantial information about Ioseph’s weeping.
Since it involves a participle, there is continuity between the nuclear state of affairs, Ioseph’s
weeping, and the genitive absolute's state of affairs. This continuity would not have been
present had the translator used ἐν τῷ + infinitive.297

295

Albert Rijksbaron, The Syntax and Semantics of the Verb in Classical Greek: An Introduction, 3rd
ed. (Chicago: University of Chicago, 2006), 122.
296
Genitive absolute constructions are structurally independent of the nuclear clause, but semantically
dependent.
297
See Levinsohn, Discourse Features of New Testament Greek, 188-189.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

109

5.2.4. Παρὰ τό + Infinitive
LXX Genesis has one text that employs παρὰ τό + infinitive as a rendering for ‫ ְבּ‬+
infinitive (Gen. 29:20). The Greek construction is very similar to that of ἐν τῷ + infinitive
from an IS perspective. Both constructions grammaticalize presuppositional information that
is marked syntactically for identifiability. Such information is already active in the discourse
register and belongs to an utterance's topical domain. However, it grammaticalizes a specific
relationship with its predicator that is different than that associated with ἐν τῷ + infinitive.
Generally, παρὰ τό + infinitive indicates a cause or motive that is associated, at least in the
LXX, with abstract states of affairs (cf. Deut 7:8; 9:28; 2 Rgns 10:3).298 So there is some
overlap in this regard with the ἐν τῷ + infinitive construction, but παρὰ τό + infinitive does
not grammaticalize any sort of temporal nuance in its semantic relationship with its
predicator; it is very close in this respect to the causal function that is encoded with διὰ τό +
infinitive. The following example also illustrates how the Hebrew ‫ ְבּ‬+ infinitive construction
can be used to indicate cause.
(5) Genesis 29:20
‫ַוַיֲּﬠ ֧ ֹבד ַיֲﬠ ֹ֛קב ְבָּרֵ֖חל ֶ֣שַׁבע ָשׁ ִ֑נים ַו ִיְּה ֤יוּ ְבֵﬠיָני ֙ו ְכָּי ִ֣מים ֲאָח ִ֔דים ְבַּאֲהָבת ֖וֹ ֹא ָ ֽתהּ׃‬
LXX: καὶ ἐδούλευσεν Ἰακὼβ περὶ Ῥαχὴλ ἔτη ἑπτά, καὶ ἦσαν ἐναντίον αὐτοῦ ὡς
ἡμέραι ὀλίγαι παρὰ τὸ ἀγαπᾶν αὐτὸν αὐτήν.
NETS: And Iakob was subject seven years for Rachel, and they were in his sight like
a few days because he loved her.
Genesis 29:18 states, "Now Iakob loved Rachel, and he said, 'I will be subject to you
seven years for your younger daughter Rachel.'” The construction παρὰ τὸ ἀγαπᾶν αὐτὸν
αὐτήν, then, refers to the clause "Iakob loved Rachel" anaphorically. Therefore, the syntactic
construction grammaticalizes presuppositional information using an identifiable textual
constituent. As argued above, it creates a causal semantic relationship with its predicator.
Thus, the construction explains what caused the seven years of labor to seem like a few days
in Iakob's sight. Παρὰ τὸ ἀγαπᾶν αὐτὸν αὐτήν does not encode any sort of event, activity, or
accomplishment that can function to communicate the temporal cause or scope of a

298

This is consistent with one example of παρὰ τό + accusative according to LSJ, "more generally of
the margin by which an event occurs, i.e. of the necessary and sufficient cause or motive" (Henry
George Liddell, Robert Scott, and Henry Stuart Jones, A Greek-English Lexicon [Oxford: Clarendon
Press, 1996], 1304).

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

110

predicator in the narrative like ἐν τῷ + infinitive can. It communicates an abstract emotional
state that is causal.
5.2.5. Adverb
There is only one passage in LXX Genesis that employs an adverb as a rendering for ‫ ְבּ‬+
infinitive. Adverbs are typically not used in the LXX to render that Hebrew construction, but
due to the context of the verse, it is appropriate. Generally, the Hebrew ‫ ְבּ‬+ infinitive
construction provides adverbial information, so an adverb can be appropriately used as a
functional equivalent.
(6) Genesis 35:17
ֹ ‫ַו ְיִהי ְבַהְק‬
‫ ֵבּן׃‬0‫שָׁתהּ ְבִּלְד ָתּהּ ַותּ ֹאֶמר ָלהּ ַהְמַי ֶלֶּדת ַאל־ ִתּי ְרִאי ִכּי־ ַגם־ֶזה ָל‬
LXX: ἐγένετο δὲ ἐν τῷ σκληρῶς αὐτὴν τίκτειν εἶπεν αὐτῇ ἡ μαῖα Θάρσει, καὶ γὰρ
οὗτός σοί ἐστιν υἱός.
NETS: And it came about while she was giving birth with difficulty that the midwife
said to her, “Take courage, for this one also is a son for you.”
In this verse that was discussed in chapter four, ἐν τῷ + infinitive is employed to render
the second of two consecutive ‫ ְבּ‬+ infinitive constructions (‫)ְבַה ְקֹשָׁתהּ ְבִּלְד ָתּהּ‬. The second one is
rendered using ἐν τῷ + infinitive, and the first one — the intransitive Hiphil — is rendered
using the adverb.
5.2.6. ’Εν τῷ + Νoun
There is a single instance in Genesis in which ἐν τῷ + noun is the counterpart to ‫ ְבּ‬+
infinitive. There is no semantic or pragmatic difference between ἐν τῷ + infinitive and ἐν τῷ
+ noun, especially when the noun itself semantically predicates a verbal process. Consider
the following text:
(7) Gen. 35:16
‫ַו ִיְּסע ֙וּ ִמֵ֣בּית ֵ֔אל ַֽו ְיִהי־֥ﬠוֹד ִכְּבַרת־ָהָ֖אֶרץ ָל ֣בוֹא ֶאְפ ָ֑רָתה ַו ֵ֥תֶּלד ָרֵ֖חל ַו ְתּ ַ֥קשׁ ְבִּלְד ָ ֽתּהּ׃‬
LXX: ἐγένετο δὲ ἡνίκα ἤγγισεν χαβράθα εἰς γῆν ἐλθεῖν Ἐφράθα, ἔτεκεν Ῥαχὴλ καὶ
ἐδυστόκησεν ἐν τῷ τοκετῷ.
NETS: Now it came about when he drew near Chabratha to go to the land of
Ephratha, that Rachel gave birth, and she experienced severe birth pangs in the birth.
The construction ἐν τῷ τοκετῷ communicates a verbal process with the noun τοκετός
(birth). This is evident from the noun's lexical semantics and from the fact that it possesses
the same denotation as the finite verb in the preceding nuclear clause ἔτεκεν ("gave birth).
The following verse, Genesis 35:17 (cf. §5.2.5; also discussed in chapter four), uses ἐν τῷ +

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

111

infinitive in describing the same event. Like ἐν τῷ + τίκτειν, ἐν τῷ τοκετῷ refers
anaphorically to presuppositional information. Its text-internal antecedent is ἔτεκεν,
information that was newly asserted in the previous clause using a finite verb. Since the
information is freshly lit up in the reader's consciousness, it is then encoded using a textual
constituent marked for identifiability. Why, then, would the translator use ἐν τῷ + noun
instead of ἐν τῷ + infinitive? While one cannot be certain, the fact that there are three
Hebrew ‫ ְבּ‬+ infinitive constructions in proximity to one another in Genesis 35:16-17 (one at
the end of 35:16 and two near the beginning of 35:17) may have something to do with it.
Only ‫ ַו ְיִהי‬at the beginning of 35:17 separates the three constructions. Therefore, due to that
proximity, it may not be surprising that the translator, for the sake of Greek style, would
choose to render one construction using ἐν τῷ + noun, one with ἐν τῷ + infinitive, and one
with an adverb.299 He has therefore translated the text's meaning accurately with a variety of
Greek renderings.
5.2.7. Νoun
Finally, there is a single passage in LXX Genesis in which a noun appears as a
counterpart to ‫ ְבּ‬+ infinitive.
(8) Genesis 21:16
‫ ַו ֨ ֵתּ ֶשׁב ָ֜להּ ִמ ֗ ֶנּ ֶגד ַה ְרֵח֙ק ִכְּמ ַטֲח ֵ֣וי ֔ ֶק ֶשׁת ִ֣כּי ָ ֽאְמ ָ ֔רה ַאל־ֶא ְרֶ֖אה ְבּ֣מוֹת ַה ָ֑יֶּלד ַו ֵ֣תּ ֶשׁב ִמ ֔ ֶנּ ֶגד ַו ִתּ ָ֥שּׂא ֶאת־ֹקָ֖להּ‬0֩‫ַו ֵתֶּל‬
‫ַו ֵ ֽתְּב ְךּ׃‬
LXX: ἀπελθοῦσα δὲ ἐκάθητο ἀπέναντι αὐτοῦ μακρόθεν ὡσεὶ τόξου βολήν, εἶπεν γάρ
Οὐ μὴ ἴδω τὸν θάνατον τοῦ παιδίου μου. καὶ ἐκάθισεν ἀπέναντι αὐτοῦ, ἀναβοῆσαν
δὲ τὸ παιδίον ἔκλαυσεν.
NETS: And after departing she sat down opposite him a good way off, about a
bowshot, for she said, "I will not look upon the death of my child." And she was
seated opposite him, and the child cried out and wept.
One English translation renders Hagar's words in the Hebrew text, "Do not let me see
when the child dies" (LSB). This translation perhaps captures the temporal nuance encoded
with ‫ ְבּ‬+ infinitive. However, Gen decided to use a rendering that is structurally less complex
than ἐν τῷ + infinitive, though he still appropriately conveyed the meaning of the text. There

299

See Bentein, Verbal Periphrasis in Ancient Greek: Have– and Be– Constructions, 17. Bentein
discusses how periphrastic constructions were sometimes used for style variation. If varatio can
explain the text-linguistic make-up of compositional texts, it can perhaps explain some of the
renderings chosen by the LXX translators, as well.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

112

may have been some structural reasons for not using ἐν τῷ + infinitive since generally this
construction is employed as an adjunct, though there are some scattered examples of its
usage as a complement.300 This may be an example, then, in which the translator chose a
particular rendering because it was the simplest solution. It could also be the case that Gen
read the Hebrew word as a noun, which is quite possible, and read the ‫ ְבּ‬prefix as a marker of
the object.
5.3. Summary
In the discussion regarding the rendering of the Hebrew ‫ ְבּ‬+ infinitive construction, this
chapter has data in support of the quotation at the beginning of the chapter that the translator
used "a variety of Greek terms or expressions depending on contextual demands."301
Discourse pragmatics certainly played a role in determining what renderings were chosen.
Most obvious, perhaps, is the distinction between the choice of a finite verb – whether it be a
finite verb by itself (1x) or temporal conjunction + finite verb (19x) – and that of ἐν τῷ +
infinitive. The textual data have demonstrated, especially concerning these constructions, that
the translator appears to have kept the discourse context in mind when considering what
terms and expressions to utilize. Even though ἐν τῷ + infinitive provided the most formally
quantitative rendering of the grammatical forms of the Hebrew, it did not become a
stereotype due to its IS status within the AG linguistic system.

300

In the twelve occurences of ἐν τῷ + infinitive in Polybius discussed in chapter three, two are in the
complement position. Of the twenty-three cases in LXX Genesis, only two are in the complement
position. The papyri and inscriptions attest no examples.
301
Scarlatta, "Genesis," 17.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

113

CHAPTER 6: CONCLUSION
A WINDOW INTO TRANSLATION TECHNIQUE
“If the translator varies his rendering according to the context, then his attention to the
context is a form of consistency.”302
– Theo A. W. van der Louw
6.1. Summary of Findings
To provide a small window into the translation technique used in LXX Genesis, I have
attempted to examine the way a single syntactic construction was used in various PC Greek
corpora. For this investigation, two compositional corpora (PC literary texts and PC
documentary sources) were compared to LXX Genesis. There are many differences in these
texts in terms of sentence structure. Polybius and authors of epigraphic texts typically employ
more complex sentence structures and levels of subordination, while papyri and LXX
Genesis contain simpler sentence structures with a higher frequency of finite verbs. However,
the structural semantics and discourse pragmatic function of ἐν τῷ + infinitive were
consistent throughout these corpora despite the variation in syntactic complexity.
Although the primary goal of this thesis has been to investigate the discourse pragmatic
function of ἐν τῷ + infinitive, IS analysis cannot be neatly separated from structural
semantics.303 Structural semantics provides the basis for understanding the meaning of
linguistic structures, while IS builds upon this foundation in the exploration of how those
structures are used to convey information effectively in different discourse
contexts. Therefore, the structural semantics of the ἐν τῷ + infinitive construction have been
prominent in the linguistic analysis of this thesis. As the textual data have shown, this
construction is most frequently an adjunct. Yet both Polybius and LXX Genesis provide
examples in which the construction is a complement. However, the semantic relationship
with a predicator – whether adjunct or complement – has consistently been described as
occasion in time with certain more specific nuances. It has been argued that these nuances—
temporal scope and temporal cause—are contingent on the lexical semantics of both ἐν τῷ +

302

Theo A. W. van der Louw, “The Evolution of the Greek Genesis Translator,” in Die Septuaginta –
Geschichte, Wirkung, Relevanz, eds. Martin Meiser, Michaela Geiger, Siegfried Kreuzer, and Marcus
Sigismund (Tübingen: Mohr Siebeck, 2018), 146-57, here 149.
303
Levinsohn, Discourse Features of New Testament Greek, viii.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

114

infinitive and its predicator. This provides a clearer perspective than that of some
grammarians, as has been argued in chapter two. It has been noted that they regard the ἐν τῷ
+ infinitive construction itself to grammaticalize in every instance a particular semantic
relationship with its predicator in certain corpora of AG. The argument is that CG employs ἐν
τῷ + infinitive to indicate cause, and the LXX uses it distinctively to indicate temporality,
borrowing from the structural semantics of Hebrew. However, in this thesis, it has been
argued that this construction only encodes meaning based on the lexical semantics of the
infinitive and its predicator, constrained by the syntax of the prepositional phrase.
Pragmatically, the textual data have shown that ἐν τῷ + infinitive is prototypically used to
grammaticalize presuppositional information in discourse. This is information that a speaker
intends to be taken for granted because it represents common ground assumptions shared by
the speaker and the listener. In the AG linguistic system, presuppositional information is
marked as identifiable using the article. Therefore, the entire syntactic construction denotes
information that ought to be identifiable to a listener or reader. Furthermore, the information
communicated using ἐν τῷ + infinitive belongs to the topic domain in an utterance. In other
words, it encodes information that indicates what an utterance is about. Generally, the
information denoted by ἐν τῷ + infinitive is not accentuated; it is information that possesses
less pragmatic salience compared to that of other grammatical forms.
This lack of pragmatic salience is apparent in chapter five, where ἐν τῷ + infinitive has
been compared to other renderings in LXX Genesis of the Hebrew ‫ ְבּ‬+ infinitive construction.
Syntactically, the ἐν τῷ + infinitive counterpart is a “downgraded” form that routinely
communicates information that is fresh within the discourse register. Therefore, it does not
need to be accentuated. Additionally, it is only used to provide occasion in time for a single
state of affairs within a narrative sequence of events. In contrast, finite verbs are used to
communicate more pragmatically salient information either because the information is in the
process of being activated in the discourse register or because the information denotes a state
of affairs that provides the occasion in time for multiple events within a narrative sequence.
The textual data, then, support the argument that Gen was consistent in his renderings, using
syntactic constructions according to the discourse pragmatic value that best fit the context.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

115

As van der Louw asserts, “If the translator varies his rendering according to the context, then
his attention to the context is a form of consistency.”304
6.2. Translation Issues
6.2.1. The Issue of Register
It was observed in chapter two that scholars and students of AG are far removed from
the socio-cultural realities of the Hellenistic world. Therefore, it is difficult for them to make
appropriate observations about register since they lack the sociolinguistic intuitions that
native speakers possess. However, they do possess the textual data left behind by those native
speakers. Using these data, then, they can proceed by identifying lexical and syntactic
constituents that are restricted to a particular register or that are employed more frequently
than others in a particular register. The complex nature of this task is the reason why "[w]e
are only in the early stages of understanding register in the PC period."305 In chapter two, it
was noted as well that LXX Genesis occupies a middle-level linguistic register and contains
some literary elements. This conclusion was drawn based on careful studies by many
scholars of extant texts from the Graeco-Ptolemaic era in Egypt. One task of linguists
working in LXX studies is to identify what belongs to a middle-level linguistic register and
what is constitutive of literary Greek. Though much of the lexical stock and many syntactic
structures in any given language will have wide utility, certain words and structures are more
register-specific.
The question at hand, then, is whether ἐν τῷ + infinitive is a syntactic construction that
was used across registers in AG or is it a literary construction. Answering this question
definitively is very difficult, but the attempt to do so might lead to further insight into the
translation technical profile of LXX translators in general, and Gen more specifically. The
fact that it occurs only occasionally in documentary sources (which tend to be low to middlelevel registers) and more frequently in literary texts might constitute support for the
conclusion that it belongs to a literary register. However, one of the documentary sources
cited in this thesis is a personal letter from a pig farmer. Based on the assumption that pig

304
305

van der Louw, “The Evolution of the Greek Genesis Translator,” 149.
Aitken and Dhont, "The Septuagint with the History of Greek: An Introduction," 438.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

116

farmers would not have had access to the same sort of education and schoolroom texts that
were available to LXX translators and the educated elite in Ptolemaic Egypt, it is hard to
imagine that a pig farmer could or would use syntax that was characteristic of a literary
register. Therefore, the tentative conclusion of this thesis is that ἐν τῷ + infinitive is a
syntactic construction that was utilized across registers within the AG linguistic system. It is
possible that the publishing of additional documentary sources will provide more attestations
of this syntactic construction in these sources and give a more complete picture of AG. But
this construction was perhaps not widely used because it was more specific semantically,
indicating occasion in time for its predicator, whereas a construction involving an adverbial
participle, though less specific semantically, can be used to indicate various semantic
relationships with a predicator depending on the context. The wide semantic utility of
adverbial participles and the low frequency of the ἐν τῷ + infinitive construction are
indicative of Zipf’s meaning frequency law, which states that cross-linguistically,
constituents with more meaning possibility tend to be used more frequently than those with a
more limited semantic range.306 In this regard, the adverbial participle is a workhorse able to
grammaticalize the various semantic values that can also be grammaticalized by several
different prepositions + [article] + infinitive.
6.2.2. The Issue of Frequency
Another issue that is related to Gen’s translation technique is the issue of frequency. In
the introduction of this thesis, it was stated that some scholars have argued that LXX Greek
exhibits certain syntactic structures with greater frequency than is the case in compositional
texts.307 This type of influence, called positive interference, refers to conventional syntax that
is evinced more commonly because "it happens to resemble a feature in the source
language."308 Would, then, the common occurrence of ἐν τῷ + infinitive in LXX Genesis be
an example of “positive interference?”

306

Ramon Ferrer-i-Cancho and Michael S. Vitevitch, “The Origins of Zipf’s Meaning-Frequency
Law,” Journal of the Association for Information Science and Technology 69, issue 11 (2018): 136979, here 1369.
307
Evans, "The Nature of Septuagint Greek: Language and Lexicography," 94.
308
Aitken and Dhont, "The Septuagint within the History of Greek," 441.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

117

While its usage in LXX Genesis is more frequent than in the majority of compositional
Greek texts, this issue needs to be considered carefully. From the beginning of the PC era
(ca. 330 BCE) until the Common Era, ἐν τῷ + infinitive was used 397 times by 62 authors in
extant compositional texts included in the TLG database. This corpus, excluding the LXX,
contains 12,694,837 words. Therefore, this syntactic construction occurs 0.03 times per 1,000
words in the corpus as a whole. This is far less frequent than 0.71 times per 1,000 words in
LXX Genesis. So, comparing LXX Genesis with the extant literary texts from the PC era, it
would appear that the frequency of its usage in the LXX is an example of positive
interference. However, some PC authors used it much more frequently than others, such as
the Pythagorean philosopher Theages. He employed the construction more than 10 times per
1,000 words, which is far more frequent than is the case in LXX Genesis. Additionally, there
are fourteen other authors of compositional texts (see Appendix 1) who used the construction
with a greater frequency than the translator of LXX Genesis. Proponents of the argument for
frequency of usage, which is often employed as an argument for Hebraic interference in LXX
studies, must take into consideration the fact that even in compositional texts, there is a great
amount of variation from one author to the next. Conventional Greek does not exhibit any
consistency in the frequency with which certain syntactic constructions are used. Frequency,
then, is an issue of idiolect.
6.3. Areas for Further Research
The conclusions in this thesis about the structural semantics and discourse pragmatic
function of ἐν τῷ + infinitive need to be scrutinized in the light of a much larger database of
AG texts. Ideally, this would mean investigating more thoroughly not only PC Greek literary
texts but also CG literary texts. There are approximately 300 attestations of this construction
in CG literary texts, the study of which would provide further insight into its structural
semantics and discourse pragmatic function. This could provide evidence to contradict the
assertion that it was only used causally during that era of the Greek language. The IS
distinction between ἐν τῷ + infinitive and constructions using finite verbs in compositional
Greek texts and in other books of the LXX also needs to be further explored. Moreover, other
books in the LXX need to be examined to see how the respective translators of those books

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

utilized ἐν τῷ + infinitive within the AG linguistic system as one of several possible
renderings of ‫ ב‬+ the infinitive construct. It is possible that other translators, such as the
translator of the Psalter, used it as a stereotyped rendering without considering its
information structural status. But only an examination of textual data can demonstrably
validate or invalidate that hypothesis.

118

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

119

WORKS CITED
Abbott, T. K. Essays, Chiefly on the Original Texts of the Old and New Testaments.
London: Longmans, Green & Co., 1891.
Aejmelaeus, Anneli. Parataxis in the Septuagint: A Study of the Renderings of the
Hebrew Coordinate Clauses in the Greek Pentateuch. Dissertationes Humanarum
Litterarum 31. Helsinki: Suomalainen Tiedeakatemia, 1982.
Aitken, James K., and Marieke Dhont. “The Septuagint with the History of Greek: An
Introduction.” Journal for the Study of Judaism 54 (2023): 432-49.
Aristotle. Nicomachean Ethics. Translated by H. Rackham. Loeb Classical Library 73.
Cambridge, MA: Harvard University, 1926.
Bailey, N. A. “Thetic Constructions in Koine Greek: With Special Attention to Clauses
with εἰμί 'be', γίνομαι 'occur', ἔρχομαι 'come', ἰδού/ἴδε 'behold', and Complement
Clauses of ὁράω 'see'.” PhD diss., Vrije Universiteit Amsterdam, 2009.
Barton, John. The Word: How We Translate the Bible – And Why It Matters. New York:
Basic Books, 2023.
Bentein, Klaas. Verbal Periphrasis in Ancient Greek: Have– and Be– Constructions.
Oxford: Oxford University, 2016.
————. “Going Nominal: The Ancient Greek Articular Infinitive Between Syntax and
Context.” In Subordination and Insubordination in Post-Classical Greek: From
Syntax to Context, edited by Klaas Bentein, Eleonora Cattafi, and Ezra la Roi, 1–41.
Berlin: De Gruyter, 2025.
Bertrand, Nicolas. “Information Structure and Greek.” In Encyclopedia of Ancient Greek
Language and Linguistics, vol. 2, edited by Georgios K. Giannakis. Leiden: Brill,
2014.
Blass, Friedrich, Albert Debrunner, and Robert W. Funk. A Greek Grammar of the New
Testament and Other Early Christian Literature. Chicago: The University of Chicago
Press, 1961.
Bons, Eberhard. “The Septuagint and Greek Style.” In The T&T Clark Handbook of
Septuagint Research, edited by William A. Ross and W. Edward Glenny. London:
T&T Clark, 2021.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

120

Boyd-Taylor, Cameron. “In a Mirror, Dimly – Reading the Septuagint as a Document of
Its Times.” In Septuagint Research: Issues and Challenges in the Study of the Greek
Jewish Scriptures, edited by Wolfgang Kraus and R. Glenn Wooden, 15-31. Atlanta:
SBL Press, 2006.
————. “Toward an Analysis of Translation Norms: A Sighting Shot.” In BIOSCS 39
(2006): 27-46.
Brock, Sebastian. “Aspects of Translation Technique in Antiquity.” Greek, Roman, and
Byzantine Studies 20, no. 1 (2004): 69-87.
Brunetti, Lisa, Kordula De Kuthy, and Arndt Riester. "The Information-Structural Status
of Adjuncts: A Question-under-Discussion-Based Approach," Discours 28 (2021): 348.
Büring, Daniel. "(Contrastive) Topic." In The Oxford Handbook of Information Structure,
edited by Caroline Féry and Shinichiro Ishihara, 64-85. Oxford: Oxford University,
2016.
Burk, Dennis Ray. “A Linguistic Analysis of the Articular Infinitive in New Testament
Greek.” PhD diss., The Southern Baptist Theological Seminary, 2004.
Caragounis, Chrys C. The Development of Greek and the New Testament: Morphology,
Syntax, Phonology, and Textual Transmission. Grand Rapids: Baker Academic, 2006.
Chomsky, Noam. Syntactic Structures. 2nd ed. Berlin: Mouton de Gruyter, 2002.
Civilleri, Germana Olga. “Abstract Nouns.” In Encyclopedia of Ancient Greek Language
and Linguistics, vol. 1, edited by Georgios K. Giannakis, 4-7. Leiden: Brill, 2014.
Conybeare, F. C., and St. George Stock. Grammar of Septuagint Greek. Boston: Ginn,
1905.
Cox, Claude. “Tying It All Together: The Use of Particles in Old Greek Job.” BIOSCS 38
(2005): 41–54.
Dhont, Marieke. “Septuagint Translation Technique and Jewish Hellenistic Exegesis.” In
T&T Clark Handbook of Septuagint Research, edited by William A. Ross and W.
Edward Glenny, 21-35. London: T&T Clark, 2021.
Deissmann, Adolph. “Hellenistic Greek with Special Consideration of the Greek Bible.”
In The Language of the New Testament: Classic Essays. Edited by Stanley Porter,
Journal for the Study of the New Testament Supplement Series 60. Sheffield:
Sheffield Academic, 1991.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

121

————. Bibelstudien. Marburg: Elwert, 1895.
————. Neue Bibelstudien. Marburg: Elwert, 1897.
————. Licht vom Osten. Tübingen: Mohr, 1908.
Dik, Helma J. M. “Word Order in Ancient Greek: A Pragmatic Account of Word Order
Variation in Herodotus.” PhD diss., University of Amsterdam, 1995.
————. Word Order in Greek Tragic Dialogue. Oxford: Oxford University Press,
2007.
Dik, Simon C. The Theory of Functional Grammar, Part 1: The Structure of the Clause.
2nd ed. Edited by Kees Hengeveld. Berlin: Mouton de Gruyter, 1997.
Dooley, Robert A. Functional Approaches to Grammar: Implications for SIL Training.
Manuscript.
Evans, Trevor. “The Nature of Septuagint Greek: Language and Lexicography.” In The
Oxford Handbook of the Septuagint, edited by Alison Salveson and Timothy Michael
Law, 90-104. Oxford: Oxford University, 2021.
————. “The Grammarian Cannot Wait: Thackeray, Muraoka, and the Analysis of
Septuagint Syntax.” Journal for the Study of Judaism 54 (2023): 558-581.
Ferrer-i-Cancho, Ramon, and Michael S. Vitevitch. “The Origins of Zipf’s MeaningFrequency Law.” Journal of the Association for Information Science and Technology
69, no. 11 (2018): 1369-79.
Fillmore, Charles J. “Frame Semantics.” In Linguistics in the Morning Calm, edited by
the Linguistics Society of Korea, 111-137. Seoul: Hanshin Publishing Co., 1982.
Fresch, Christopher J. Discourse Markers in Early Koine Greek: Cognitive-Functional
Analysis and LXX Translation Technique. Atlanta: SBL Press, 2023.
Funk, Robert W. A Beginning-Intermediate Grammar of Hellenistic Greek. Part II:
Syntax. Atlanta: Scholars Press, 1973.
Gee, James Paul. Introducing Discourse Analysis: From Grammar to Society. London:
Routledge, 2018.
George, Coulter H. Expressions of Time in Ancient Greek. Cambridge: Cambridge
University Press, 2014.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

122

Halliday, M. A. K. Halliday’s Introduction to Functional Grammar. 4th ed. Revised by
Christian M. I. M. Matthiessen. Abingdon: Routledge, 2014.
Hatch, Edwin. Essays in Biblical Greek. Oxford: Clarendon Press, 1889.
Hetterle, Katja. Adverbial Clauses in Cross-Linguistic Perspective. Berlin: De Gruyter
Mouton, 2015.
Hiebert, Robert. "Translation Technique in the Septuagint of Genesis and Its Implications
for the NETS Version." BIOSCS 33 (2000): 76-93.
————. "Genesis: To the Reader." In A New English Translation of the Septuagint,
edited by Albert Pietersma and Benjamin G. Wright, 1-6. Oxford: Oxford University
Press, 2007.
————. "Linguistic Interference in Septuagint Genesis." Journal of Septuagint and
Cognate Studies 55 (2022): 55-73.
Hippocrates. Epidemics 2, 4-7. Edited and translated by Wesley D. Smith. Loeb Classical
Library 477. Cambridge, MA: Harvard University Press, 1994.
Horrocks, Geoffrey. Greek: A History of the Language and Its Speakers. 2nd ed.
Chichester, UK: Wiley Blackwell, 2010.
————. “Phases of the Greek Language.” In Die Sprache der Septuaginta/ The
Language of the Septuagint, vol. 3 of Handbuch zur Septuaginta, edited by Eberhard
Bons and Jan Joosten, 71-88. Gütersloh: Gütersloher Verlagshaus, 2016.
Johannessohn, Martin. Der Gebrauch der Präpositionen in der Septuaginta. Berlin:
Weidmannsche Buchhandlung, 1925.
Kennedy, H. A. A. Sources of New Testament Greek: The Influence of the Septuagint on
the Vocabulary of the New Testament. Edinburgh: T&T Clark, 1895.
Kiparsky, Paul. "Greek Anaphora in Cross-Linguistic Perspective." Journal of Greek
Linguistics 12 (2012): 84–117.
Lambrecht, Knud. Information Structure and Sentence Form: Topic, Focus, and the
Mental Representations of Discourse Referents. Cambridge: Cambridge University
Press, 1994.
Lee, John A. L. A Lexical Study of the Septuagint Version of the Pentateuch. Septuagint
and Cognate Studies 14. Chico, CA: Scholars Press, 1983.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

123

————. A History of New Testament Lexicography. Studies in Biblical Greek. Edited
by D. A. Carson. New York: Peter Lang, 2003.
————. The Greek of the Pentateuch: Grinfield Lectures on the Septuagint 2011-2012.
Oxford: Oxford University, 2018.
————. “Back to the Question of Greek Idiom.” In The Legacy of Soisalon-Soininen:
Towards a Syntax of Septuagint Greek, edited by Tuukka Kauhanen and Hanna
Vanonen, 13-25. Göttingen: Vandenhoeck & Ruprecht, 2020.
Le Moigne, Philippe. “Le Livre d’Ésaïe dans la Septante: Ecdotique, Stylistique,
Linguistique.” PhD diss., L’École pratique des hautes études, 2001.
Levinsohn, Stephen H. Discourse Features of New Testament Greek: A Coursebook on
the Information Structure of the New Testament Greek. 2nd ed. Dallas: SIL
International, 2000.
————. “Towards a Unified Linguistic Description of οὗτος and ἐκεῖνος.” In The
Linguist as Pedagogue: Trends in the Teaching and Linguistic Analysis of the Greek
New Testament, edited by Stanley E. Porter and Matthew Brook O’Donnell, 204-16.
Sheffield: Sheffield Phoenix Press, 2009.
Levinsohn, Stephen H., and Mark Dubis. “The Use of the Greek Article in 1 Peter: A
Case Study.” In The Article in Post-Classical Greek, edited by Daniel King, 101-25.
Dallas: SIL International, 2019.
Liddell, Henry George, Robert Scott, and Henry Stuart Jones. A Greek-English Lexicon.
Oxford: Clarendon Press, 1996.
Lust, Johan, Erik Eynikel, and Katrin Hauspie. A Greek-English Lexicon of the
Septuagint. Stuttgart: Deutsche Bibelgesellschaft, 2003.
Mandilaras, Basil G. The Verb in the Greek Non-literary Papyri. Athens: Ministry of
Culture and Sciences, 1973.
Manning, J. G. The Last Pharaohs: Egypt Under the Ptolemies, 305-30 BC. Princeton:
Princeton University Press, 2010.
Matić, Dejan. “Topic, Focus, and Discourse Structure: Ancient Greek Word Order.”
Studies in Language 27, no. 3 (2003): 573-633.
Modrzejewski, Joseph Mélèze. “The Septuagint as Nomos: How the Torah Became a
'Civic Law' for the Jews of Egypt.” In Critical Studies in Ancient Law, Comparative
Law and Legal History Studies in Honour of Alan Watson, edited by John Cairns and
Olivia Robinson, 183-200. Oxford: Hart Publishing, 2001.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

124

Moeschler, Jacques. “Causality, Lexicon, and Discourse Meaning.” Rivista di Linguistica
15.2 (2003): 343-69.
Moulton, James Hope. A Grammar of New Testament Greek, Vol. 1: Prolegomena, 2nd
edition. Edinburgh: T&T Clark, 1906.
Muraoka, Takamitsu. A Syntax of Septuagint Greek. Leuven: Peeters, 2016.
Pietersma, Albert. “The Society of Biblical Literature Commentary on the Septuagint:
Basic Principles.” In The SBL Commentary on the Septuagint: An Introduction, edited
by Dirk Büchner, 1-16. Atlanta: SBL Press, 2017.
Polak, Frank. “Context Sensitive Translation and Parataxis in Biblical Narrative.” In
Emanuel: Studies in Hebrew Bible, Septuagint and Dead Sea Scrolls in Honor of
Emanuel Tov, edited by Shalom M. Paul, VTSup 94, 525-539. Leiden: Brill, 2003.
Polybius. The Histories, Volume I: Books 1-2. Translated by W. R. Paton. Loeb Classical
Library 128. Cambridge, MA: Harvard University Press, 2010.
————. The Histories, Volume II: Books 3-4. Translated by W. R. Paton. Loeb
Classical Library 137. Cambridge, MA: Harvard University Press, 2010.
————. The Histories, Volume III: Books 5-8. Translated by W. R. Paton. Loeb
Classical Library 138. Cambridge, MA: Harvard University Press, 2011.
————. The Histories, Volume V: Books 16-27. Translated W. R. Paton. Loeb
Classical Library 160. Cambridge, MA: Harvard University Press, 2012.
Porter, Stanley. Verbal Aspect in the Greek of the New Testament, with Reference to
Tense and Mood, Studies in Biblical Greek. Edited by D. A. Carson. New York: Peter
Lang Publishing, 1989.
————. “Introduction.” In The Language of the New Testament: Classic Essays, edited
by Stanley Porter, Journal for the Study of the New Testament Supplement Series 60,
11-38. Sheffield: Sheffield Academic, 1991.
————. “History of Scholarship on the Language of the Septuagint.” In Die Sprache
der Septuaginta/The Language of the Septuagint, vol. 3 of Handbuch zur Septuaginta,
edited by Eberhard Bons and Jan Joosten, 15-38. Gütersloh: Gütersloher Verlagshaus,
2016.
Porter, S. E., and M. B. O'Donnell. “Building and Examining Linguistic Phenomena in a
Corpus of Representative Papyri.” In The Language of the Papyri, edited by T. V.
Evans and D. D. Obbink, 287-311. Oxford: Oxford University, 2010.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

125

Rijksbaron, Albert. The Syntax and Semantics of the Verb in Classical Greek: An
Introduction. 3rd Edition. Chicago: University of Chicago, 2006.
Rhodes, Richard A. “Greek Prepositions: A Cognitive Linguistic View.” In Postclassical
Greek Prepositions and Conceptual Metaphor, edited by William A. Ross and Steven
E. Runge, 11-36. Berlin: Walter de Gruyter, 2022.
Robertson, A. T. A Grammar of the Greek New Testament in the Light of Historical
Research. Logos Bible Software, 2006.
Ross, William. Post-classical Greek and Septuagint Lexicography. Atlanta: SBL Press,
2022.
Runge, Steven E. Discourse Grammar of the Greek New Testament: A Practical
Introduction for Teaching and Exegesis. Bellingham, WA: Lexham Press, 2010.
————. “Towards a Unified Understanding of the Greek Article from a Diachronic,
Cognitive Perspective.” In The Article in Post-Classical Greek, edited by Daniel
King, 127-52. Dallas: SIL International, 2019.
Scarlatta, Mark W. “Genesis.” In T&T Clark Companion to the Septuagint, edited by
James Aitken, 13-28. London: Bloomsbury T&T Clark, 2015.
Silva, Moisés. Biblical Words and Their Meaning: An Introduction to Lexical Semantics.
Revised and Expanded Edition. Grand Rapids, MI: Zondervan, 1994.
Soisalon-Soininen, Ilmari. Die Infinitive in der Septuaginta. Helsinki: Suomalainen
Tiedeakatemia, 1965.
Sollamo, Raija. “Some ‘Improper’ Prepositions, such as ἐνώπιον, ἐναντίον, ἔναντι, etc.,
in the Septuagint and Early PC Greek.” Vetus Testamentum 25, fasc. 4 (October
1975): 773-82.
Thackeray, Henry. A Grammar of the Old Testament in Greek: According to the
Septuagint. Cambridge: Cambridge University Press, 1909.
Tov, Emanuel. “Greek Words and Hebrew Meanings.” In Melbourne Symposium on
Septuagint Lexicography, edited by Takamitsu Muraoka, 109-28. Atlanta: Scholars
Press, 1990.
Tronci, Liana. “Contact-Induced Change and Language-Internal Factors: The Καὶ
Ἐγένετο Type as a Case-Study.” In Papers On Ancient Greek Linguistics:
Proceedings of the Ninth International Colloquium on Ancient Greek Linguistics,

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

126

edited by Martti Leiwo, Marja Vierros, and Sonja Dahlgren, 177-204. Helsinki:
Societas Scientiarum Fennica, 2020.
Turner, Nigel. A Grammar of New Testament Greek: Syntax, vol. 3. Edinburgh: T. & T.
Clark, 1963.
Van der Louw, Theo A. W. “The Evolution of the Greek Genesis Translator.” In Die
Septuaginta – Geschichte, Wirkung, Relevanz. Edited by Martin Meiser, Michaela
Geiger, Siegfried Kreuzer, and Marcus Sigismund, 146-57. Tübingen: Mohr Siebeck,
2018.
Van Emde Boas, Evert, Albert Rijksbaron, Luuk Huitink, and Mathieu De Bakker.
Cambridge Grammar of Classical Greek. Cambridge: Cambridge University Press,
2019.
Walbank, F. W. A Historical Commentary on Polybius, vol. 1. Oxford: Clarendon Press,
1957.
Wallace, Daniel. Greek Grammar Beyond the Basics: An Exegetical Syntax of the New
Testament. Grand Rapids: Zondervan, 1996.
Wevers, John William. Notes on the Greek Text of Genesis. Society of Biblical Literature
Septuagint and Cognate Studies Series 35. Atlanta: Scholars Press, 1993.
Zerwick, Maximilian. Biblical Greek: Illustrated by Examples. Translated by Joseph
Smith S. J. Rome: Editrice Pontificio Istituto Biblico, 1963.

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

127

APPENDIX 1: 'Εν Τῷ + Infinitive Frequency in Post-Classical Texts
Num. Author/Text
Attestations Total Words Frequency per
1,000 words
309
PC Literature
397
12,694,837
0.03x
Papyri310
9311
~812,429
~0.01x
1
Dinarchus
1
19,019
0.05x
2
Theophrastus
21
217,058
0.09x
3
Epicurus
6
27,815
0.21x
4
Menander
7
80,882
0.08x
5
Cleanthes
1
9,746
0.10x
6
Clearchus
2
10,068
0.20x
7
Duris
2
5,108
0.39x
8
Hecataeus
5
13,418
0.37x
9
Persaeus Phil.
1
1,865
0.54x
10
Demetrius Phalereus
1
4,457
0.22x
11
Timaeus Hist.
1
25,466
0.04x
12
Damoxenus Comic.
1
1,626
0.62x
312
13
Charondas
1
860
1.16x
14
Dioscurides
1
1,431
0.69x
15
Diotogenes
1
1,645
0.61x
16
Erasistratus
2
19,583
0.10x
17
Lycon
2
910
2.20x
18
Chrysippus
47
192,890
0.24x
19
Euryphamus
3
722
4.16x
20
Metopus
5
1,428
3.50x
21
Straton
2
7,618
0.26x
22
Hieronymus
1
1,019
0.98x
23
Bryson
1
216
4.63x
24
Callicratidas
1
1,250
0.80x
25
Theages
12
1,129
10.63x
26
Polybius
24
316,866
0.08x
309

This takes into consideration compositional literary texts BCE and excludes the LXX. Statistics
were taken from Thesaurus Linguae Graecae.
310
This takes into consideration papyri BCE. The total word count is only an approximation as there
are no precise statistics available. There is only the total word count for all papyri from 8th century
BCE to 8th century CE (4,513,494; https://www.trismegistos.org/words/about.php). On papyri.info,
18% of total uses of the conjunction και is attested BCE; the same percentage is true for the article ὁ
and the verb εἰμί. So, the total word count for papyri BCE in this table is based on 18% of the total
word count in all papyri. Obviously, this can only give an approximation.
311
This total number adds one additional papyrus that is not discussed in this thesis because it is from
a literary register before the PC era.
312
Compositional texts that contain the same or higher frequency than LXX Genesis are highlighted.
There are 14 authors/texts in the compositional PC corpus before the CE that contain a higher
frequency than LXX Genesis. LXX and NT frequencies are included at the end.

128

DISCOURSE PRAGMATIC FUNCTION OF ’ΕΝ ΤΩΙ + INFINITIVE

27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66

313

Aristophanes
Heraclides
Philo Mechanicus
Demetrius
Anonymus Photii
Timaeus Phil.
Aristocles Paradox.
Agatharchides Geogr.
Ezechiel
Apollodorus Gramm.
Antipater
Demetrius Lacon
Diogenes Phil.
Anonymus Epicureus
Heracleodorus
Comanus
Dionysius Scytobrachion
Posidonius
Antiochus Phil.
Ptolemaeus Gramm.
Testamenta XII Patriarcharum
Diogenis Sinopensis Epistulae
Diodorus Siculus
Dionysius Halicarnassensis
Arius Didymus
Tryphon I Gramm.
Philodemus
Philoxenus Gramm.
Socrates
Philo Judaeus
Strabo
Aristonicus
Dorotheus
Pseudo-Archytas
Antiochus Astrol.
Vitae Adam et Evae
LXX
LXX Genesis
NT
Luke/Acts

4
1
2
2
3
1
1
2
1
1
2
1
2
1
1
3
1
5
1
2
7
4
36
31
13
1
35
4
1
53
3
5
1
9
1
4
534313
23
52
38

47,279
2,730
18,955
5,025
1,993
4,659
790
18,995
1,632
12,636
4,273
3,248
10,655
2,394
4,524
2,409
11,229
139,273
1,060
9,088
20,255
8,863
464,305
415,573
28,655
13,807
164,519
37,165
734
449,267
324,136
117,050
28,658
12,749
9,094
6,573
623,782
32,567
137,938
37,879

0.08x
0.37x
0.11x
0.40x
1.51x
0.21x
1.27x
0.11x
0.61x
0.08x
0.47x
0.31x
0.19x
0.42x
0.22x
1.25x
0.09x
0.04x
0.94x
0.22x
0.35x
0.45x
0.08x
0.07x
0.45x
0.07x
0.21x
0.11x
1.36x
0.12x
0.01x
0.04x
0.03x
0.71x
0.11x
0.61x
0.86x
0.71x
0.38x
1.003x

The total attestations in the LXX (and total word count) are based on A. Rahlfs, Septuaginta, 9th
ed. (Stuttgart: Württemberg Bible Society, 1935).