Jump to ContentJump to Main Navigation
Verbal Syntax in the Greek PentateuchNatural Greek Usage and Hebrew Interference$

T. V. Evans

Print publication date: 2001

Print ISBN-13: 9780198270102

Published to Oxford Scholarship Online: September 2011

DOI: 10.1093/acprof:oso/9780198270102.001.0001

Show Summary Details
Page of

PRINTED FROM OXFORD SCHOLARSHIP ONLINE (www.oxfordscholarship.com). (c) Copyright Oxford University Press, 2017. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a monograph in OSO for personal use (for details see http://www.oxfordscholarship.com/page/privacy-policy). Subscriber: null; date: 24 February 2017

(p.265) Appendix 1 Notes On Tables 1, 8, 9, and 11

(p.265) Appendix 1 Notes On Tables 1, 8, 9, and 11

Verbal Syntax in the Greek Pentateuch
Oxford University Press

1. Notes on Table 1

  1. (a) Table 1 includes all 18,953 synthetic verbal forms in the Greek Pentateuch according to Wevers's text. His Göttingen editions actually contain 18,954 forms, but corrections affecting the verbal classification proposed by Wevers for his critical editions at Lev. 5: 7; 21: 1, Num. 11: 33, 22: 11, and Deut. 30: 13 have been incorporated here.1 At Lev. 21: 1 Wevers would now omit the present participle λέγων, which appears in his edition.

    The figures were generated initially by morphological analysis of the CATSS databases. The resulting CATSS counts have been manually corrected throughout to accord with Wevers's text (instead of Rahlfs's edition) and to remove tagging errors contained, in the databases (cf. § 1.6 and for details of my changes to the CATSS figures see App. 2).

  2. (b) This table does not take account of periphrastic tense forms. Figures for these are given in Table 139.4.13). See Chapter 9 for discussion of the complex problem of verbal periphrasis.

  3. (c) The counts involve a limited degree of subjectivity since certain Greek forms are ambiguous. It is not possible to allocate all forms to particular categories in a purely objective manner. Problematic decisions mainly concern, identification of forms as imperfect or aorist indicative, as aorist or present subjunctive (especially in liquid and nasal stems), and as aorist subjunctive or future indicative. My classifications of doubtful cases are based on contextual probabilities. The number of such cases is very small, however, and different conclusions would have minimal, impact on the statistics. It should be noted in this connection that some of the CATSS errors listed in App. 2, § 1 may reflect other scholars' reasoned decisions based on context, not merely the results of automatic tagging.

    With regard to the imperfect or aorist indicative type, -έχεεν forms from χέω compounds have been, interpreted as aorist indicatives, not imperfects, contrary to the CATSS analysis. The morphology is ambiguous, but context strongly suggests that the forms in question ought all to be read as (p.266) aorists. Out of 15 examples the only cases which seem readily open, to the alternative interpretation occur in Gen, 38: 9 and 39: 21. Ambiguous forms are also produced, in the third person singular by liquid, and, nasal stems. Here too the aorist indicative has been read in the 6 instances according to contextual probabilities. For details of all examples see App. 2, § 1.

    Debatable aorist subjunctive or future indicative instances relate mainly to first person singular deliberative questions. McWhorter makes a persuasive case that ambiguous forms in such questions should usually be taken as aorist subjunctive, not future indicative.2 This is my interpretation (requiring some changes to the CATSS analysis; see App. 2, § 1) of most Pentateuchal examples, e.g. ποιήσω in Gen. 27: 37; 39: 9; Exod. 17: 4, Wevers, however, takes the Gen, 39: 9 instance as future indicative.3 Apparently in agreement is Voitila, since he recognizes only two instances of deliberative subjunctives in direct questions within the chapters he treats, and cites the two in Gen, 44: 16 (λαλήσωμεν and δικαιωθῶμεν).4 These writers, are presumably influenced by the following future ἁμαρτήσομαι, but this should not dictate classification of ποιήσω, even if the two verbs are taken as closely coordinated. Note that the Gen 44: 16 examples follow an indicative (τί ἀντεροῦμεν τῷ κυρίῳ ἢ τί λαλήσωμεν ἢ τί δικαιωθῶμεν;).5

    Incidentally, Mayser recognizes instances of the deliberative subjunctive in. the Ptolemaic papyri, only in indirect questions6 (for Pentateuchal examples in indirect questions note Exod. 8: 9 τάξαι πρός με, πότε εὔξω-μαι …, 10: 26 ἡμεῖς δε οὐκ οἴδαμεν, τί λατρεύσωμεν …). But on the one hand it is not clear to what extent ambiguous forms in direct questions are an issue in the papyri and how Mayser treats them, while on the other the often formulate character of his material perhaps makes the absence there of deliberative direct questions less notable, The deliberative subjunctive in direct questions is also witnessed in NT usage.7

  4. (d) A more objective decision between future indicative and aorist subjunctive is possible for ἐξάξω in Gen, 19: 8. Wevers identifies this as a sigmatic aorist exhortative subjunctive from ἐξάγω, clearly influenced by (p.267) the ΜΤ cohortative אוציאה‎.8 However, I take the form as future indicative, in agreement with the CATSS tag. The sigmatic aorist of ἄγω and its compounds becomes common enough in. the Koine,9 but is infrequent in the LXX.10 I recognize no convincing examples in the Greek Pentateuch. The Gen. 19: 8 instance makes excellent sense as a future indicative, so Wevers's interpretation seems both unnecessary and unlikely

  5. (e) The verb ἥκω has a hybrid conjugation in post-Classical Greek, its present indicative adopting perfect endings in the plural (see § 6.2.5 n. 42). Both the singular and plural forms are classified here as present indicatives, as tagged in the CATSS files. Taylor's classification of the plural forms as perfects is certainly reasonable on formal grounds,11 but tends to obscure the unity of the paradigm.

  6. (f) Not included in my counts are three instances of οφελον (in Exod. 16: 3 translating מי יהן‎; in Num. 14: 2 and 20: 3 translating לו‎), which is tagged in the CATSS files as an aorist participle. The form, was originally verbal, but becomes fossilized in post-Classical Greek as an optative particle.

    The two interpretations of the word's precise origin to be found in the literature are worth noting here. It is commonly taken as developing from an unaugmented first singular aorist indicative of ὀφείλω,12 but Wackernagel's interpretation, that the form, is historically a neuter participle with, ἐστίν understood, is certainly possible and has its supporters.13 Taylor seems wrong to parse the form as a participle in its LXX occurrences,14 where it clearly functions as a particle; however, the similar CATSS tagging may well reflect mechanical, error, rather than, intention. Whatever the origin of the form, it has been, completely removed from, the verbal system.

  7. (g) The participle ἐχόμενος in its usage translating Hebrew prepositions and prepositional phrases is included in the counts, in agreement with the CATSS tagging. Though, treated, by some writers as a quasi-preposition, it retains its character as a verbal adjective in the Pentateuchal examples (see § 4.4.2 n. 40).

(p.268) 2. Note on Table 8

The discrepancy between, my total figure of 80 Pentateuchal optatives and Turner's of 70 (56 volitives; 5 potentials; 9 comparatives), which I glean from his general lists of LXX citations (see § 7.3 n. 21), seems to be due to more than use of different texts and methods regarding variants. Since the process Turner employed to generate his statistics is unclear, I shall riot attempt to reconcile our two counts, but note certain, apparent errors in his lists, If he used Rahlfs's edition, he missed numerous examples; I note Gen, 1.6: 5; 28: 3 (bis; Turner cites only 1 of the 3 occurrences in this verse), 4; 48: 20; Exod. 5: 21 (bis); Num. 5: 21, 22 (Turner cites only 1 of 2 occurrences in. this verse); 11: 12; Deut. 1: 11 (bis); 28: 8, 9, all of which are read by Rahlfs. Conversely Turner cites an example from Gen. 19: 9, but whether he refers to the A reading κριναι (both Rahlfs and Wevers read κρίνειν), which Brooke and McLean accent κρῖναι, i.e. as an infinitive,15 or less probably συντρῖψαι, these forms are both to be taken as aorist infinitives. I can see no textual warrant at all for his Num., 22: 7 citation.16

3. Notes on Table 9

  1. (a) The figures for historical prose writers in Table 9 are adapted from Miller, who does not claim for them ‘Rigid accuracy’; ‘They are the result of a single careful count’.17

  2. (b) The less complete NT figures are from Fanning, excluding ‘the imperfects of εἰμί and the occurrences of ἔφη’.18 Fanning's figures are based on the counts of Hawkins and the percentages of Schlachter19 (with percentages for Mark corrected according to Fanning's own count).

  3. (c) My supplementary samples are selected, almost at random, but I have deliberately chosen mainly narrative passages from Lysias and Demosthenes. The statistics for the different authors in these samples must be treated with reserve. They are drawn from samples of approximately ten pages of text each. For Vita Aesopi G this represents nearly a quarter of the (p.269) whole work, but for Chariton less than, a fifteenth, and for the orators a tiny fraction of their corpora. Further, to test these particular works editions of dissimilar formats have had to be used, so that the ten pages of each do not amount to anything more than roughly equivalent portions of Greek text. Nevertheless, their evidence, for what it is worth, accords with general trends. Chariton and the rhetorical narrative passages are similar to other literary prose documents, the vernacular language of Vita Aesopi G to the NT books. This is as we might expect.

  4. (d) From the Chariton sample 2 imperfect and 3 aorist indicative forms in Homeric quotations have been omitted.

4. Note on Table 11

The figures of Table 11, unlike those for all my Pentateuchal word, counts, are based on Rahlfs's 1935 edition of the LXX. They have been generated purely by morphological analysis of the CATSS databases without manual modification, and so must be assumed to reproduce some morphological tagging errors. Experience of CATSS tagging relevant to the imperfect and aorist indicatives in the Pentateuchal books (on which see App. 1, § 1(c)) suggests, however, that in most cases errors would not be sufficiently numerous to distort the statistics significantly.


(1) Wevers, Leviticus Notes, 57, 331; id., Numbers Notes, 180–1, 366; id., Deuteronomy Notes, 484.

(2) A. W. McWhorter, ‘A Study of the So-called Deliberative Type of Question(τί ποιήσω;) as Found in Aeschylus, Sophocles, and Euripides’’, TAPhA 41 (1910),157–67, esp. 165–7 on ambiguous forms.

(3) Wevers, Genesis Notes, 654.

(4) A. Voitila, ‘La technique de traduction du Yiqtol (l'imparfait hébreu) dans l'Histoire du Joseph grecque (Gen 37, 39–50)’, in C, E. Cox (ed.), VII Congress of the International Organization for Septuagint and Cognate Studies, Leaven 1989 (Atlanta, Ga.: Scholars Press, 1.991), 223–37 at 227–8.

(5) Cf. the Eur. Ion 758 example in § 2.6.3 and McWhorter ‘Deliberative’, 159–60, for statistics on subjunctive/future combinations in deliberative questions in Aeschylus, Sophocles, and Euripides.

(6) Mayser, Grammatik, ii/1. 235–6; cf. Mandilaras, Verb, §§ 571–3, esp § 572, observing a similar situation, in the later papyri.

(7) BDR, Grammatik, § 366; Turner, Syntax, 98–9; and for the independent deliberative subjunctive in Modern Greek see Thumb, Handbook, 126.

(8) Wevers, Genesis Notes, 269.

(9) Gignac, Morphology, 290–1, with 290 n. 6; Mayser, Grammatik, i/2. 144.

(10) Wevers, Exodus Text, 229–30; Thackeray, Grammar, 233 (he notes ἐπάξω in Exod. 33: 5, regarding it as probably future, but Wevers, Exodus Text, 229–30, reads ἐπαγάγω there); Helbing, Grammatik, 90.

(11) Taylor, Parsing Guide, 212.

(12) Helbing, Grammatik, 73–4; Moulton, Prolegomena, 200–1 and 201 n. 1; Thackeray, Grammar, 199,280; LSJ, s.v. ὀφείλω, II.3.c.; Robertson, Grammar, 923, 1003–4; MH, Grammar, 191, 252; SD, Syntax, 345–6.

(13) J. Wackernagel, Sprachliche Untersuchungen zu Homer (Göttingen: Vandenhoeck & Ruprecht, 1916), 199–200; cf. BDR, Gmmmatik, § 67.2. and n. 4; BAGD, s.v ὄφελον; Wevers, Exodus Notes, 243; McKay, New Syntax, § 10.3.1. n. 1.

(14) Taylor, Parsing Guide, 326.

(15) A. E. Brooke, N. McLean, and (from vol. ii/1 (1927) onwards) H. St. J. Thackeray, The Old Testament in Greek (3 vols. in 9 pts.; Cambridge: Cambridge University Press, 1906–40), i/1. 4.5.

(16) Turner, Syntax, 132 n. 1.

(17) C. W. E. Miller, ‘The Imperfect and the Aorist in Greek’, AJPh 16 (1895), 139–85 at 142 n. 1. Cf. the figures of L. Schlachter, ‘Statistische Untersuchungen überden Gebrauch. der Tempora und Modi bei einzelnen griechischen Schriftsteilern’, IF 22 (1907), 202–42 at 229.

(18) Fanning, Verbal Aspect, 254.

(19) Sir J. C. Hawkins, Horae Synopticae: Contributions to the Study of the Synoptic Problem, 2nd edn. (Oxford: Clarendon Press, 1909), 51; Schlachter, ‘Statistische Untersuchungen’, 229, Note that MH, Grammar, 457, count 165 imperfects in John, incorporating Burney's correction to Hawkins's count, and. that this is the figure cited by Porter, Verbal Aspect, 136.