¹ A version of this paper was presented at the AULLA XXIII Congress at the University of Melbourne on 8 February 1985.

² I.e. infinitive or participle.

³ An example of contextual influence, admittedly in terms of English meaning for convenience, is the use of to mean ‘come’ or ‘go’. For an example of inflexional influence see my discussion of verbs of knowing on 297–309 in ‘The Perfect and Other Aspects in New Testament Greek’, Nov. Test. 23 (1981), 289–329 (hereafter referred to as ‘McK. NTPerf.’)

⁴ As in LSJ s.v.

⁵ Cf. McKay, K.L. Greek Grammar for Students: A concise grammar of classical Attic with special reference to aspect in the verb (Canberra, A.N.U. Dept. of Classics (1974/77), §23.1.5;Google Scholar ‘On the Perfect and Other Aspects in the Greek Non-Literary Papyri’, BICS 27 (1980), 23–49 (see 25–7 and nn.16, 17); and McK. NTPerf. 296 f. The first two are referred to hereafter as ‘McK. Gram.’ and ‘McK. Pap.’. respectively.

⁶ Traditionally called ‘present’. I here use the terminology adopted in McK. Gram.

⁷ E.g. Humbert, J., Syntaxe Grecque² (Paris 1954), 177 §298;Google Scholar Post, L.A., ‘Dramatic Uses of the Greek Imperative’, AJP 59 (1938), 31;Google Scholar Turner, N. in Moulton, J.H., Howard, W.F. and Turner, N., A Grammar of New Testament Greek 3 (London: 1963), 75.Google Scholar

⁸ I use ‘command’ to cover any function of the imperative: instruction, order, prayer, entreaty, advice, concession, etc.

⁹ McK. Gram. §28; Louw, J.P., ‘OnGreekProhibitions’ Acta Classica 2 (1959), 43–57;Google Scholar Bakker, W.F., The Greek Imperative: An investigation into the aspectual differences between the present and aorist imperatives in Greek prayer from Homer up to the present day (Amsterdam 1966).Google Scholar These are hereafter referred to simply as ‘Louw’ and ‘Bakker’ respectively.

¹⁰ An article of mine, ‘Aspect in Imperatival Constructions in New Testament Greek’ has recently been published in Nov. Test. 27 (1985), 201–226.

¹¹ Perhaps Euripides, since Aeschylus has already spoken in favour of the test. I do not think here is simply formulaic; but even if it is, the aspect preserved in the formula is the one most commonly appropriate. So in although the middle form and the accent suggest that this has become more a particle than an imperative, the aorist is the aspect normally to be expected in such circumstances.

¹² Cf. Xen. Hell. 5.1.18 and Pl. Prot. 314b in both of which the going is clearly to be completed before the dining or hearing can take place, but the commencement of the going is immediately required in order to make the complete achievement of the other activity possible. These examples are quoted in McK. Gram. §§28.2.1 and 28.1 (154 f.). The formulaic use of ἄϒε and developed from the normal choice of the imperfective in such circumstances.

¹³ The substitution of subjunctive for imperative in ancient prohibitions does not affect the aspectual question.

¹⁴ See also 52 ff. below. This use of the imperfective is no more purely formulaic than the common use of the imperfect tense to introduce speeches in narrative (see McK. Gram. §§23.3 and A 6, 7) or the linking of aorist and imperfect in narrative as in Hom. Il 1.33 (see 48 f. below).

¹⁵ Stanford, W.B., (ed.), Aristophanes: The Frogs² (London 1980), 190.Google Scholar

¹⁶ Bakker (44) in the main gives a reasonable account of this passage, but he weakens it by suggesting that the poets do not need although not specified as such, it is as much a signal as the cry () for letting go.

¹⁷ Uhlig, G. Uhlig, Grammatici Graeci (Teubner, ) 2.2.358 (Bekker p. 253).Google Scholar

¹⁸ Not ‘completed’, as proposed by LSJ, s.v.

¹⁹ Although it could mean ‘extending’ or ‘duration’.

²⁰ It was for this reason that I rejected ‘durative’ and ‘punctiliar’ as replacements for ‘present’ and ‘aorist’.

²¹ I would, however, suggest that in Luke 10.4 is not so much’do not try to carry’ (55) as ‘do not carry (be carrying)’, either habitually or, more likely, simply against the background of the prescribed journey and throughout it.

²² Cf. Bakker 38–40. He discusses and rejects Louw’s argument, but as he fails to perceive the weakness of translating as ‘terminating point’ he does so on the grounds that ‘the aorist does not express the of an action, but denotes a simple occurrence, an absolute fact that does not have any connection with another verbal notion’ (39).

²³ Seiler, H., L’aspect et le temps dans le verbe néo-grec (Paris 1952). See Bakker 21 ff.Google Scholar

²⁴ As I have already stated, I make no attempt to assess the validity of Seller’s theory in relation to modern Greek.

²⁵ V.1. but the verbs which follow are jussive infinitives: this does not affect the aspectual question.

²⁶ Note that ò represents but has the advantage of being unambiguously aorist.

²⁷ In 1413 he also uses the perfect ‘let her (Thetis) be aware’, which may imply that she must ‘get to know’ (by itself ), but puts emphasis on the need for her to ‘have the knowledge’, and may be all the more concessive or defiant for that.

²⁸ In Eur. Bacch. 1120 f. (Bakker 51) the aorist is used more urgently in an effort to ward off imminent death. In this context it would be reasonable for Pentheus to draw attention to the threat being made to him by using the imperfective ‘stop trying to kill’, as Croesus’ son did in Hdt. 1.85.4, but by using the aorist he simply urges her not to kill, and its abruptness adds to the urgency of the appeal.

²⁹ McK. Gram. §A 12 (p. 218), which in fact I wrote before seeing Bakker.

³⁰ See McK. Gram. §§A 12, 13 for an illustration of different aspects with the same politeness effect.

³¹ Cf. Bakker 47.

³² Of course one could not in such circumstances rule out an inceptive nuance, but it would be subordinated to the complexive. Probably an ancient Greek would not think of the possible distinction.

³³ If she had said it would have had the same kind of abrupt urgency as in Eur. Bacch. 1120 f. (n.28 above), but the imperfective has a deeper psychological appeal which makes it preferable to her.

³⁴ The same applies to Bakker’s similar pair of passages on p. 51, Eur. Bacch. 1120, where the son appeals to the mother with and Eur. Hec. 287, where the appeal to an outsider uses the aorist, but any distinction of degrees of immediacy of the feeling of pity is fanciful.

³⁵ It is difficult to translate ( idiomatically and preserve the echoes of both here and in 1048.

³⁶ See 43 f. above.

³⁷ In 44 n.41 he refers to the clerk’s preparedness as ‘in certain circumstances’ sufficient to allow for the omission of an aorist imperative.

³⁸ According to Miller, C.W.E., ‘The Limitation of the Imperative in the Attic Orators’, AJP 13 (1892), 399–436 at 405,Google Scholar occurs 134 times and 48.

³⁹ Cf. nn.11 and 14 above: also McK. Pap. 37–42.

⁴⁰ In Greek terms it is more significantly a matter of recognising the written letters, and so is the means of producing the speaking. On the other hand, if it was normal in the ancient world to read aloud, the same sort of overlapping as occurs in English is not surprising.

⁴¹ His only use of in such a context is in 10.16.

⁴² Cf. 13.35, 50; also with slight variation, 1.30, 31; and 9.8 Dem. 40.33 … 40.35 …

⁴³ Lys. 23.4, 8, 11, 14, 15. The two exceptions in 23 are at 3 and 10.

⁴⁴ According to Miller in AJP 13 (1892), 425, there are seven perfect imperatives in the orators. I have not traced them all, but some have a different effect, e.g. Dem. 40.30

⁴⁵ This is analogous to the use of the pluperfect in Thuc. 4.47.1, quoted in McK. Gram. §23.4.1, and the perfect tense in Ar. Lys. 859, quoted in McK. Gram. §24.5.1 and in K.L. McKay, ‘The Use of the Ancient Greek Perfect Down to the Second Century A.D.’, BICS 12 (1965), 1–21 at 13 f.

⁴⁶ Cf. in Eur. LA. 1413, referred to in n.27 above. IG 1³.14 (no. 40 in R. Meiggs and Lewis, D.M. A Selection of Greek Historical Inscriptions to the End of the Fifth Century B.C. [Oxford 1969]),Google Scholar an inscription of mid-fifth century Athens, has at lines 29 ff. as the punishment for murder, and applying to the next (uncertain) situation mentioned. As in the sense ‘I am in exile’ (and perhaps more generally) is a stative verb, there is a certain parallelism between these perfect and imperfective imperatives.

⁴⁷ Cf. McK. Gram. §23.4.3.

⁴⁸ ‘Don’t begin to fear’ would be

⁴⁹ In narrative with past reference similarly it is likely that the aorist tense is unmarked for action verbs and the imperfect tense for stative verbs. In general timeless statements, however, the present tense seems to be the unmarked one, and the aorist marked. The perfect is always marked. Any application of markedness theory to ancient Greek seems to need some reference to context rather than a generalised account such as that in Ruipérez, M.S. Estructura del sistema de aspectos y tiempos del verbo griego antiguo (Salamanca 1954).Google Scholar

⁵⁰ If the imperfective had been used it would imply either habitual use, a significant parallelism between using and not fearing, or a beginning to use.

⁵¹ In classical Greek the jussive future is not much found with a negative, and when it is the negative μή or In N.T. Greek oύ is used, in spite of the tendency in later Greek for μή to spread at the expense of oύ. My recent article in Novum Testamentum (see n.10 above) deals with this and other imperative substitutes. In the application of aspectual principles the Hellenistic Koine does not differ from all forms of classical Greek to any observable extent.

⁵² Lewis, D.M. ‘Notes on the Decree of Themistocles’, CQ 11 (1961), 64 n.l.CrossRefGoogle Scholar I quote from the text of the decree printed in Meiggs, R. and Lewis, D.M. A Selection of Greek Historical Inscriptions (see n.46 above), 48 ff.,Google Scholar and for translation details I use the notes of Jameson, M.H. ‘Revised Text of the Decree of Themistokles from Troizen’, Hesperia 31 (1962), 313 ff.CrossRefGoogle Scholar For another epigraphic example see n.46.

⁵³ The example (CIA 4.2.54b. 11 ff.) quoted in Meisterhans, K. and Schwyzer, E. Grammatik der attischen Inschriften (Berlin 1900), 244 Google Scholar paragraph 16, is similar, but not quite the same, introducing an activity by other agents which is parallel to the specific instruction The imperfective could have been used for both if further emphasis on the parallelism had been desired, but this would have sacrificed some of the completeness effect given by the aorist.

⁵⁴ For cf. the comments on in Ar. Frogs 1378 and in Lys. 7.10, on 43 and 54 above, and nn. 11, 12; and for cf. the discussion of PI. Rep. 327b on 50 above. I have not listed the conjectured in line 12, but its explanation would be similar.