RECEIVED JUL 2

Novum Testamentara XXVII, 3 (1985)

ASPECT IN IMPERATIVAL CONSTRUCTIONS

IN NEW TESTAMENT GREEK

by

K. L.t McKAY
The Australian National University, Canberra

Introduction
Although there is nowadays general recognition of some of the ef-
fects of aspectual usage in New Testament (and indeed in all an-
cient) Greek, there appears still to be a great deal of confusion
about the value of aspectual distinctions in the imperative. This is
not altogether surprising, for there are some parallel passages in the
Gospels in which different aspects are used for apparently the same
meaning, and other passages in which rules which seem valid
elsewhere appear to be neglected. The purpose of the present article
is to re-examine the relevant material in the light of the aspectual
approaches I have developed in previous works1 in order to
demonstrate that in NT Greek aspectual usage in the imperative
and its equivalents is essentially the same as that in all areas of ver-
bal syntax from Homer to many centuries after the NT documents
were written.2 It is not my purpose to analyse the work of other
1
Especially the following, which hereafter are quoted by the abbreviations
shown in brackets: "The Use of the Ancient Greek Perfect down to the Second
Century AD", Bulletin of the Institute of Classical Studies 12 (1965), 1-21 (McK.
Perf.), "Syntax in Exegesis", TyndaU Bulletin 23 (1972), 39-57 (McK. Exeg.), Greek
GrammarfarStudents: A concise grammar of Classical Attic with special reference to aspect
in the verb, Classics A.N.U., Canberra 1974 and 1977, especially sections 23 ff.
(McK. Gram.), "On the Perfect and Other Aspects in the Greek Non-Literary
Papyri", B.I.C.S. 27 (1980), 23-49 (McK. Pap.), "On the Perfect and Other
Aspects in New Testament Greek", Nov. Test. 23 (1981), 289-329 (McK. NT
Perf.). See also my forthcoming article in Antichthon 20 (1986).
2
But not necessarily in modern Greek. There seems to be no doubt that the
aspectual distinctions of modern Greek are descended from and are in some ways
similar to the aorist-imperfective part of the ancient Greek verbal system, but the
modern Greek perfect is very different from that of ancient Greek, and the other
changes in the structure of the language make it necessary to prove rather than
assume continuity of value. W. F. Bakker in his Utrecht doctoral thesis The Greek
Imperative, Hakkert, Amsterdam 1966, tries to explain the ancient Greek aspects
in terms of a theory based on modern Greek, and while much of his argument is
202 Κ. L. MCKAY

grammarians, nor to establish a system by induction or by deduc­

tion from observation of the data, but to test against this body of
data a system which I have already expounded elsewhere and found
to offer a satisfactory solution to the problem of ancient Greek
aspect. Inevitably, however, there will be some comparison with
and comment on the work of others, and some discussion of basic
principles.3 Among "imperatival constructions" I include the im­
perative (2nd and 3rd person), the subjunctive which regularly
takes the place of the imperative in aorist prohibitions and occa­
sionally elsewhere, the subjunctive used in exhortation (with some
reference to deliberative questions, which are closely related to
these), future statements and some subjunctive equivalents used as
commands, infinitives used both as direct substitutes for the im­
perative and to represent the imperative in indirect commands, and
participles representing imperatives. I do not include other
imperative-equivalents not found in the New Testament, such as
potential statements and some types of question.4

Aspect and Context

Aspectual usage is very closely dependent on context. The sub­

jective choice of the speaker or writer is in some respects part of the
context and in other respects one of the factors moulding the con­
text, so in the search for an objective explanation the observer must
be careful not to expect to find a system so mechanically organized
that all its operations are entirely predictable, especially as in
reading ancient texts we have to deduce from the text itself much
of the context which the writer would have taken for granted in
5
making his choice. As I have pointed out elsewhere, in many con-
illuminating, he is forced to support his theory by distorting the contextual settings
of a large proportion of the examples he quotes, both classical and koine. Bakker's
bibliography and introductory chapters are a useful starting point for a fuller
survey of theories about imperative usage. His analysis (on p. 45) of Plato, Republic
327b may be compared with mine in McK. Gram. Appendix A § 12.
3
I am at present engaged in the preparation of a syntax of the verb in NT
Greek, and this article is intended to deal with some of the problems uncovered
in that preparation.
4
The absence of such equivalents in such a limited body of text is not necessari­
ly significant. It simply means that the NT material does not include contexts
favourable to them. Ac 13:10 ού παύσχι διαστροφών ...; won'tyou stop perverting ...?
could be equivalent to an imperative (stop pervertingl) if its context were different.
5
McK. Perf. 4 f., McK. Pap. 24 f., 31 f., McK. NT Perf. 290-5, K. L. McKay,
"Repeated Action, the Potential and Reality in Ancient Greek", Antichthon 15
(1981), 36 f.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 203

texts the choice of aspect is relatively unimportant or is ultimately

decided by a subjective judgment of what is appropriate, but this
does not mean that aspectual differences are meaningless: we must
judge the force of the different aspects from those contexts in which
the choice is significant, and in considering the remainder merely
satisfy ourselves that the choice is not inappropriate. In dealing
with a limited body of material such as the New Testament it is also
important to bear in mind that the range of contextuad possibilities
may be limited, and the rarity or non-occurrence of a feature which
is relatively common in other types of literature (e.g. Attic comedy)
is not necessarily proof that that feature had disappeared from the
language or was unknown to N T writers.
It seems axiomatic that the aspectual system found in the im-
perative and in the jussive uses of the subjunctive may be expected
to be essentially the same as that found in the indicative in historical
or descriptive narrative, in the infinitive as found in all types of
writing, and so on, although there may be some special effects or
limitations which produce distinctive, but not radically different,
realizations in any of these aireas: and so I have found it to be. It
may immediately be noted that in imperative usage in the N T the
perfect aspect is, although quite regular, almost non-existent, 6 so
that our concern is almost entirely with the imperfective7 and aorist
aspects. One probable implication of this is that N T writers who
were native speakers of Aramaic are even less likely to be found
misusing the Greek aspectual system in this area, for their own
language had a two-aspect system similar to, but not necessarily
identical with, the aorist/imperfective segment of Greek. At least
any Aramaic influence on their aspectual usage will be less percepti-
ble to us, but in view of their ability to deal with the Greek perfect
in relation to the aorist, a distinction not represented in Aramaic, 8
we need not be too concerned at the possibility of missing some
nuance.
The difference between the aorist and imperfective aspects is that
the former represents an activity9 as a total10 action, in its entirety
6
On the perfect imperatives which do occur in NT see McK. NT Perf. 324 f.
and 303: see also p. 223 (n. 57) below.
7
I use the terms imperfective, aorist, perfect and future as explained in McK. Gram.,
McK. Exeg., McK. Pap. and Mck. NT Perf. See also n. 11 below.
8
McK. NT Perf. 294 f., 314-22.
9
For my use of activity, action and process see McK. Gram. 23.1.1, note 2.
10
Whole or complete are suitable descriptions, but completed is not, especially in
relation to the imperative and the subjunctive.
 204 Κ. L. MCKAY

without dwelling on its internal details; while the latter represents

an activity as a process going on, with the focus on its progress or
development. Both relate to the context, the aorist usually in­
dicating that the activity, being whole in itself, is in some way
separate from other activities in the context, so that a series of
aorists most commonly (but not necessarily) signals a succession of
actions; and the imperfective usually indicating some kind of
parallel, initiatory or continuative relationship with other activities
in the context. Duration and repetition are not restricted to the im­
perfective: a long drawn out activity or a series of repeated activities
may be represented as a totality in relation to the context and so
be expressed by the aorist.11 Conversely too a momentary act may
in relation to a context be represented as a process, and so express­
ed by the imperfective. It is the relationship of the activity to the
relevant context which determines the aspect, not any innate
characteristic of the activity.
I have found it useful to distinguish two main lexical verb-types:
action verbs and stative verbs. As I have pointed out elsewhere,12
it is not always easy to distinguish these, and some verbs seem to
belong in part to both types; but in the main the distinction is clear
enough, provided that one judges in terms of ancient Greek usage
and not in terms of apparent lexical equivalents in other languages.
Although some stative verbs seem to be found mainly in the im­
perfective, it is clear that the vast majority of the verbs of both types
could be used in all aspects according to the aspectual requirements
II
It is because of such factors that I decided not to adopt the terms durative and
punctiliar, which are favoured by some scholars. I wanted to retain as much of the
traditional terminology as was suitable for my purpose. Present has too many tem­
poral overtones, so I retained it only for the present tense (indicative), and chose
for the aspect imperfective, which was already in common use among linguists to
describe approximately what is needed. Perfective, on the other hand, which
balances it in Slavonic, Semitic and some other aspect systems, was not suitable,
for it might have been applied to either the aorist or the perfect, so I retained both
these terms in preference. There seemed to be less problems in using them for both
an aspect and a tense within that aspect than in the case of the imperfective aspect
and the present tense. For the other tense of the imperfective aspect I retained im­
perfect.
12
Cf. McK. Gram. § 23.1.5, McK. Pap. 25-7 and nn. 16, 17, McK. NT Perf.
296 f. These types are similar in some respects to the transformative and non-
transformative types proposed by M. S. Ruipérez in Estructura del sistema de aspectos
y tiempos del verbo griego antiguo, Salamanca 1954, §§ 90 f. In § 117 he adds a further
refinement of instantaneous and durative types, but I doubt if this really helps in
the study of aspect, and in any case it seems less relevant to my distinction between
action and stative verbs, which I find more useful.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 205

of the context, the types of verb being only one of the contextual
factors.13
The number of forms and basic functions of the verb system of
any language is limited, but their interrelationship in various con­
texts can produce a wider variety of effects, and in a language as
expressive as ancient Greek it is reasonable to expect a greater
rather than only minor exploitation of such possibilities. It is conve­
nient to catalogue some of the more common effects in terms of
their possible translation into other languages, but this does not
mean that such realizations in particular contexts or even types of
context were significant to the original speakers and their hearers.
It may help to illustrate this point from narrative, mainly in­
dicative, usage before we proceed to the imperative. In Ac 2: 43-3:
5 there is a preponderance of imperfects over aorists in the sum­
ming up of the developing position of the church and the setting of
the scene for a significant event in that development. The first έγίν-
ετο in 43 may be continuative (was developing) or, in view of the
singular πάση ψυχή, iterative (came upon one after another); but the sec­
ond έγίνετο is clearly iterative (were done one after another). In 44 είχον
(stative verb) is general background (had, were holding throughout
the period, irrespective of the increase, or decrease, of numbers),
whereas in 45 έπίπρασκον and διεμέριζον (action verbs) are iterative
or distributive (individually sold, etc.), and είχεν is shown by καθότι
ácv τις to be iterative/distributive also, expressing the state each one
was in on the various occasions of the distributions. In 46 μετελάμ-
βανον is iterative, and so is προσετίθει in 47. The participles in these
verses have similar aspectual significance, προσκαρτερουντες being
general description while καθ'ήμέραν shows it is also iterative,
κλώντές is iterative, αΐνοΰντες iterative or general description
(depending largely on whether the verb was felt to be action or
stative), ϊχοντες descriptive, and σωζόμενους distributive, but with a
nuance of parallelism with προσετ(θει (the adding and being saved
run in line with each other). In 3:1 the background description

1S
Defective verbs are not evidence to the contrary. The fact that V?*p- is found
only in the imperfective and V$ve"pt- in the aorist (and perfect) does not necessarily
imply that there was no aorist of the one and no imperfective of the other: we have
no evidence of their earlier development, but there can be no doubt that φέρειν and
ένεγκειν were for centuries regarded as complementary to each other, as different
aspects of the same verbal meaning. After >/έπ had been supplanted, in both verb
and noun, by yfkty- the predominant aorist corresponding to λέγειν continued to
be ebceTv, although λέξοκ was also commonly used as its equal-value alternative.
206 Κ. L. MCKAY

moves from the general to the specific, and the scene is set by the
ongoing process of άνέβαινον, which could be taken as either incep­
tive (began to go up) or simply as a background picture (were going up).
Equally part of the background is έβαστάζετο in 2, which is parallel
to άνέβαινον (as they were going, a man was being carried), while the fur­
ther descriptive background in έτίθουν is shown by χαθ'ήμέραν to be
iterative (as therefore also is the infinitive αίτείν). In 3 the aorist
ίδών denotes an event of seeing (catching sight of), and the following
ήρώτα is inceptive/continuative (began/proceeded to ask). After more
aorists in 4 έπεΐχεν is clearly continuative (continued to pay attention).
These few comments illustrate how the basic value of an aspect can
have various realizations as the context, both general and im­
mediate, develops, but it is not always possible to be precise in
describing the realizations, of which the ancients may have been
much less conscious: they had one aspect to cover a range of the
realizations we see as distinct from one another.

Aspect in the Imperative

One of the problems in assessing aspectual usage in commands,
entreaties, prayers, exhortations and the like is that there are usual­
ly less contextual indications, so that the imperative (or subjunc­
tive) verb form is often the main evidence of the speaker/writer's
intention. In real life the tone of voice or a gesture by the speaker
or some movement by another person may be an important contex­
tual element which helps the hearers to appreciate the significance
of the aspect chosen, but unless such details are specifically record­
ed we do not have their assistance in understanding the text. In fail­
ing to record them the writer may simply have had the whole
picture so clearly in his own mind that he overlooked the possibility
of leaving the reader doubtful, he may have been confident that his
readers, being familiar with the current situation and the customs
of his time, would be able to compensate for his omissions, or he
may have judged that the clumsiness resulting from too full a
description might unduly diminish the effect of his presentation.
Even in real life the choice of an aspect may seem surprising in
terms of the context up to the point of utterance, especially if it is
intended as preparation for what follows.
In the imperative the essential difference between the aorist and
the imperfective is that the former urges an activity as whole action
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 207

and the latter urges it as ongoing process. The most common

realizations of the imperfective are likely to be continue doing, begin
to do, try to do, do habitually (or from time to time), do at the same time
(parallel with), and combinations of these. The aorist is to be ex­
pected for a single definite action (do) when none of the foregoing
nuances are to be applied, and can be used for a series of repeated
activities when there are contextual indications of iteration and
some reason to regard the series as a whole, and generally when
there is emphasis on the need for a complete response rather than
merely an attempt. In a stative verb the aorist urges either com·
plexively a limited period of an activity, as in Lu 24: 29 μεΐνον, stay
overnight, or inceptively an entry into an activity, as in Jas 4: 9
πενθήσατε, become sorrowfiil; but in many realizations the difference
between action and stative verbs in the imperative is not obvious.
Time is even less important in the imperative than in other moods:
logically it cannot be past or present, and it makes no difference to
the aspect whether immediate or distant future, or actual or general
time is implied.

Positive Imperative: Second Person

There is, I believe, no serious dispute that in a positive
command14 for a specific complete action to be performed the aorist
imperative is normally found. The following are a few of the large
number that could be quoted:15 Mt 4:3 είπε, 4:6 βάλε, 5:29 ?ξελε ...
καί βάλε (differing from the preceding example only in that the con­
text is contingent rather than actual), 6:26 έμβλέψατε, 8:21
έπίτρεψόν, 9:27 έλέησον, 12:13 ϊκτεινόν, 14:15 άπόλυσον, 21:21 δρθητι
καί βλήθητι, Lu 16:7 δέξαι ... καί γράψον, Jn 4:31 φάγε, Ro 16:3
άσπάσασθε, give my greetings to,16 I Cor 11:13 κρίνατε, make a judgment,
14
I use this word for convenience to cover not only command, but also advice,
entreaty, prayer and any other categories for which the imperative is suitable.
15
Most of the illustrative examples which follow were selected more or less at
random. A large proportion come from Mt because there is a wider variety of rele­
vant contexts in the Gospels, and Mt not only stands first in the text but also con­
tains nearly 20% of all the relevant examples in the NT. Lu contains slightly less,
Mk and Jn about half as many each, Ac slightly less than Jn, and the rest
significantly less (the order of the next few being I Cor, Ro, Rev, Heb, Jas, Eph).
As I have indicated elsewhere, I regard statistical details as much less important
than the examination of data in relation to individual context.
16
The imperfective άσπάζου in III Jn 15 is specifically distributive (cf. κατ*
δνομα). In the papyri the imperfective is almost always used for composite
greetings (e.g. P.Oxy. 300.6, P.Fay. 123.24), but the aorist is common for both in-
208 Κ. L. MCKAY

Col 4:10 δέξασθε, receive (it is irrelevant that έάν ϊλθη allows for the
possibility that the action will not take place), II Tim 4:21
σπούδασον, Phlm 17 προσλαβοδ, Jas 2.18 δεΐξόν, Rev 1:11 γράψον,
6:16 πέσετε ... καί κρύψατε.
Also specific complete actions, but with more general settings,
are Eph 6:11 ένδύσασθε, Col 3:5 νεκρώσατε. There may be less of
these, but they are just as properly aorist, even when the verb used
appears to be stative, as in I Pet 5:2 ποιμάνατε, be shepherds, I Jn 5:21
φυλάξατε, be guards, Rev 22:9 προσκύνησον, be a worshipper.*1
The aorist could also be used for repeated actions when the con­
text makes it clear that iteration or distribution was intended, with
each action in the series a complete one. In Mt 6:17 Αλειψαί ... καί
... νίψαι the preceding participle νηστεύων echoes the general clause
δταν δέ νηστεύητε of verse 16. Similarly in Mt 10:11 εξετάσατε and
μείνατε18 are shown to be iterative/distributive by the clause εις ήν
δ'&ν πόλιν ... είσέλθητε. In II Tim 2:2 παράθου is shown to be similar
only by the plurality of πιστοΐς άνθρώποις and the general tenor of
the context, and in 3 συνκακοπάθησον can be presumed to have the
same force, as there is no indication of a specific sharing of suffering.
The five aorist imperatives in II Tim 4:2 are iterative because of
the general context, reinforced by the distributive force of the two
adverbs εύκαίρως άκαίρως which accompany έπ(στηθι. If προσκύνησον
in Rev 22:9 is not to be taken as referring to a general attitude of
worship but to a ritual act of worship it must also be iterative (direct
your acts of worship to God), for the angel continues to give instruc­
tions in a way that suggests that John is still paying attention to
him.
The imperfective is more commonly found in contexts of general
exhortation urging a continuing attitude or activity or a series of ac­
tivities. Sometimes with action verbs the distinction between a con­
tinuing effort and a series of activities is not clear, but with stative
verbs a continuing attitude is usually signalled. Some examples are
dividual and composite greetings (cf. I Cor 16:20 άσπάσασθε αλλήλους). See also
the reference to II Cor 13:11, 12 in McK. Exeg. 50.
17
The translations given here and throughout are not intended so much to be
idiomatic as to illustrate the literal value of the Greek (which is, of course, not
always certain). I take το ποίμνιον as an internal accusative (in respect ofyour office
as shepherds) and ¿αυτά as external (in relation to yourselves): in seeking to understand
the Greek as Greek it is necessary to avoid the rigidity of categorization of the cases
so commonly found in grammars. For Rev 22:9 see also the next paragraph.
18
These are action and stative verbs respectively. For parallel passages see also
p. 212 below.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 209

Mt 5:12 χαίρετε καί άγαλλιδσθε, be, and go on being, glad, even exul­
19
tant, 5:44 άγαπδτε ... καί προσεύχεσθε, love... and pray (attitude and
20
series of activities), 6:1 προσέχετε, keep paying attention, 7:7 αιτείτε
... ζητείτε ... κρούετε, keep on asking, seeking, knocking (is the seeking,
like the other two, a series, or is it a continuing effort?), 10:6
πορεύεσθε, keep directing your footsteps, 10:8 θεραπεύετε ... εγείρετε ...,
21
heal... raise ...on occasions (of sickness, death ...), Lu 10:5 λέγετε,
22
say in each place, 10:8 έσθίετε, eat at each mealtime, 13:24 άγωνίζεσθε,
exert a continuing effort (or keep on making efforts), Jn 8:39 ποιείτε, do
habitually, Ac 13:40 βλέπετε, keep watching, Ro 12:2 μεταμορφουσθε,
go on being transformed, 12:14 ευλογείτε, habitually speak well of, 13:3
ποίει, 14:22 Ζχε. Sometimes an adverb or other modification is very
closely associated with the aspectual force of the verb, as in Mt 6:33
ζητείτε δέ πρώτον, let your principal aim be, and continue to be, directed
towards.23
A continuing attitude or activity is also normally expressed by
the imperfective when applied to a particular situation. Some ex­
amples are Mt 9:2 θάρσει, be cheerful (as this is a stative verb it makes
no difference whether the mam addressed was already showing signs
of cheerfulness or not), 2 4 9:30 όρατε, be watchful, keep alert,25 Jn 14:11
19
Lu 6:23 χάρητε ... καί σκιρτήσατε is not really parallel because it limits the re­
joicing to £v ¿xetvty TQ ήμερα.
20
On the other hand in Mt 6:6 πρόσευξαι is aorist to emphasize that the whole
act of prayer should be in private. Note that with the change to aorist comes also
the change from plural to singular.
21
For the following δότε see p. 214 below.
22
Cf. the reference to Mt 10:11 above. It is possible that the imperfective here
could have the force of proceed to say (inceptive), as the iterative/distributive force
is made explicit by the preceding clause, but such a possibility is not important
enough to affect the thrust of the sentence.
23
Such a neuter accusative form must have developed from, and may still have
been felt to be, an internal accusative. It is to be noted that this command is a
general background to the teaching on prayer which precedes it, and so constitutes
an activity parallel to its context, another normal function of the imperfective. Cf.
also Lu 10:5 (quoted above), where πρώτον is purely adverbial, showing the order
of events to be followed on each occasion.
24
The aorist θάρσησον, by making the activity a total event, would usually imply
the entry into cheerfulness (and so its absence at the time of speaking), but it could
be complexive, referring to a total act of cheerfulness, which would imply some
definition of the period concerned. In most contexts these factors are less relevant
than an open-ended attitude of cheerfulness.
25
The aspectual value of οράτε, δρα, βλέπετε, βλέπε, βλεπέτω is the same
whether they are accompanied by another imperative or stand alone. In Rev 22:9
δρα μή, of course, the negative is not attached to δρα but to an unexpressed
προσκυνεί (or possibly προσκύνησης). When a specific act of looking is signalled we
find βλέψον, as in Ac 3:4, ΐδετε as in Mt 28:6, ?δε as in Mt 25:20. For fòoó and
Ιδε used like a particle see p. 221 below.
210 Κ. L. MCKAY

πκττεύετέ, believe, have faith in, Ac 2:34 (Ps 110:1) χάθου, be seated (an
initial action, take a seat, is implied, but the essence of the invitation
is the continued position on the seat), 21:28 βοηθεΐτε, be our helpers,
be on our side, 21:36 αίρε, get on with carrying off, (referring to the inter­
rupted attempt at violent removal).
The commanding of an activity in process in a context which
shows that the activity has not already commenced or is unlikely to
be fully executed, may in effect be inceptive or conati ve, a com­
mand to begin or to attempt the activity. In some contexts these
realizations of the imperfective are as clearly distinguishable as we
could desire, but in some there seems to be no clear means of
distinguishing them from one another or from the continuation
realization. Of course it is unlikely that speakers of ancient Greek
saw any need to distinguish clearly between them. If the notion of
beginning or attempting was important there were verbs with these
meanings available fpr use, but such verbs were not used when
there was less emphasis on beginning or attempting than on the ac­
tivity which was to be begun or attempted and an imperfective form
would in the context give sufficient indication of the intended
nuance. Some examples are Mt 9:5 ίγειρε καί περιπατεί, try to rise and
walk (or ... and start walking),26 9:24 αναχωρείτε, start moving away, be
on your way, Jn 4:21 πίστευε, try/begin to believe,27 Ac 9:15 πορεύου, start
onyour way,2& Ac 13:15 λέγετε, go ahead and speak, IITim3:l γίνωσκε,
be receptive to this information,29 Rev 16:1 υπάγετε καί έκχέετε, be onyour

26
One might have expected έγείρου, which is v.l. to έγειρε in many texts;
έγείρεσθε is found at Mt 26:46, etc. The fact that the intransitive use of the active
seems to be confined to έγειρε is irrelevant to its aspectual value, even if it had
become somewhat formulaic. It may have something of the force of time to begetting
up\ In Mt 17:7 where a more brusque command is appropriate the aorist έγέρθητε
is found, and when a dead man is addressed in Lu 7:14, έγέρθητι.
27
Or simply, as in 14:11, quoted in the preceding paragraph, have faith in,
believe-, as the verb is stative a conative/mildly inceptive realization is virtually in­
distinguishable from continuative.
28
Or continue with thejourney you hesitated to undertake. It is noteworthy that the im­
perfective seems to have been preferred for verbs of motion (cf. Mt 9:24 above),
except where there is reason to emphasize completeness: in Mt 8:9 πορεύθητι, go
(on a specified errand), in contrast with which Ιρχου, come, implies be availableforfurther
instructions.
29
On the significance of γίνωσκε see McK. NT Perf. 303 and McK. Pap. 26.
Cf. also Mk 13:29, where γινώσχετε signifies recognition as the various clues ap­
pear, in contrast with the call to immediate attention in μάθετε in the previous
verse (where γινώσχετε is indicative).
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 211

way and start pouring out (there are seven bowls to be poured in suc­
cession), 22:20 ?ρχου.30
As has been noted already, there are many contexts in which the
choice of aspect is either purely subjective or depends on contextual
factors beyond our knowledge. I now turn to a few parallel passages
in the synoptic Gospels in which aspectual differences are found.31
First, in the Lord's prayer Mt 6:11 has δος for the daily bread peti­
tion while Lu 11:3 has δίδου (v.l. δος), but the former adds σήμερον,
making this a specific request for the day, while the latter adds το
καθ* ήμέραν,32 making it a request for repeated giving, for which the
imperfective is quite appropriate, although the aorist could have
been used to signal the completeness of each act of giving re­
quested. Aorist and imperfective are equally appropriate in the
general commands of Mt 5:42 τω αίτοΰντί σε δός and Lu 6:30 παντί
αιτοΰντί σε δίδου, as αιτούν™ in both represents ος δν aítf¡
(anyone/everyone who asks), so that the distributive notion is clear and
the aorist can be used to emphasize the completeness of each act,
while the imperfective is simply iterative/distributive.33
In Mk 5:36 πίστευε is found parallel with πίστευσον (v.l. πίστευε)
in Lu 8:50. The former could imply continue with the faith you showed
in coming to me before we were interrupted, while the latter is probably
best taken as complexive, have faith to the end of this episode, although
it might suggest the need to return to a position of faith (begin again)
after the frustration of delay. The complexive idea is probably sup­
ported by Luke's addition of καί σωθήσεται, as this helps to suggest
the end desired.
30
"Ερχου may here be rather continuative, as it echoes ίρχομαι with its preg­
nant meaning (/ intend to come, so I am in effect on my way), a common enough nuance
of this verb, not least in Rev.
51
It is remarkable that there are so few. I deal only with those in which aspects
differ, ignoring those in which other elements differ but the aspect is the same, as
in Mt 18:9 «ξελε ... καί βάλε and Mk 9:47 εκβαλε, or Mt 27:49 ¿φες and Mk 15:36
Αφετε. Two more parallel passages are referred to later under "Future" and one
under "Participles".
32
Some of the witnesses which substitute 8ός also have σήμερον. Ν. Turner, in
the third volume of J . H. Moulton, W. F. Howard and N. Turner, A Grammar
of New Testament Greek, Edinburgh 1906-63 (hereafter referred to as Turner Gram.),
p. 77, notes the importance of the adverbial attachments, but overstates the force
of δίδου by translating continue to give. His point about the normal preference for
the aorist in prayers (also noted by others) is relevant only as a statistical observa­
tion, not as a rule for the choice of aspect.
33
Turner Gram. 76 misinterprets Mt 5:42 as referring to a definite occasion.
212 Κ. L. MCKAY

In Mt 10:11,14 μείνατε ... εκτινάξατε are parallel with μένετε ...

εκτινάξατε in Mk 6:10, 11 and μένετε ... αποτινάσσετε in Lu 9:4,5.
The context is one of iteration, so that the imperfective is an accep­
table possibility, and in the case of μένετε, a stative verb, the idea
could be continue to stay. That the aorist μείνατε is appropriate for a
34
complete stay at each place has been shown above, and the same
applies to εκτινάξατε. It may be significant that Matthew has a fuller
account with more elaboration of the distributive intention (εις ήν
δ'&ν πόλιν ή κώμην είσέλθητε), so it is not so surprising that he turned
completely to aorist imperatives; and that Luke not only chose the
imperfective, which overshadows the completeness of each activity,
but also the prefix άπο- in his reference to shaking off rather than
the slightly more expressive έκ- favoured by Matthew and Mark.
Another aspectual variation is found in Mt 15:10 άκούετε καί
συνίετε and Mk 7:14 ακούσατε μου πάντες καί σόνετε (ν.11 άκούετέ μ.
π. κ. συνίετε and άκούετε κ, συνίετε). Both convey the same message,
but with different levels of urgency: the imperfective (be attentive and
try to understand) is rather more general, and the aorist, especially
with the addition of μου πάντες, is more briskly specific (listen to me,
all ofyou, and take this in).
The remaining group of parallels to be noticed here is Mk 15:13
σταύρωσον αυτόν and Lu 23:21 σταυρού, σταυρού αυτόν (v.l. σταύρωσον
σταύρωσον α.). Again both are appropriate. As Luke makes more of
Pilate's reluctance to order the execution his imperfective better
suits the dramatic situation (go on with what you should be doing and
stop obstructing us),35 and its effect is heightened by the repetition.
The aorist would have been appropriate enough, even if in this con­
text it seemed stylistically weaker: in the context of Mark's nar­
rative there is less to choose between the two aspects. In the more
indirect approach of Mt 27:22, where the passive σταυρωθήτω is
found, there seems to be no reason to signal anything but the com­
plete action. As the narratives proceed Mt 27:23 and Mk 15:14
repeat the words used in the preceding verses, but Luke varies his
construction with indirect speech in 23:23 αιτούμενοι αυτόν σταυρ-
ωθηναι (v.l. σταυρώσαι), and with this relaxation of the dramatic
presentation he too reverts to the aorist.
When two or more commands are linked closely together they
are sometimes of the same aspect, and sometimes of different
34
On p. 208.
35
Note also the similar force of alpe in Lu 23:18.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 213

aspects, but the variations are not usually without recognizable

significance. An aorist followed by an imperfective commonly im­
plies a close connection between the initial complete action and the
activity which proceeds from it, but mostly the pattern is not simply
an indication of this kind of link between the two. In Mt 2:13
παράλαβε ... καί φεΰγε ... καί ΐσθι, take ... and flee ... and remain, an
initial complete action is commanded, and then a more prolonged
one whose urgency will be relaxed after its commencement, and an
indefinite stay. The fleeing is, however, very significantly linked
with the taking, and the two amount to take (the child) away. The
same pattern is found without the context of urgency in Mt 2:20
παράλαβε ... καί πορεύου, where be onyour way is still at least as ap­
propriate as a command to make a complete journey. In Mt 5:24
άφες commands the interruption of the activity of offering, υπάγε
the transitional journey, διαλλάγηθι the complete reconciliation,
and πρόσφερε the resumption of the procedure of offering. In Mt 9:6
ίγειρε, άρον ... καί δπαγε36 the same pattern (take up ... and proceed)
is preceded by another imperfective (try to get up). This triple pattern
also occurs in Jn 5:8 ϊγειρε, άρον ... καί περιπατεί, where the last im­
perfective is more significantly begin to walk, or habitually walk.37
The imperfective is found preceding the aorist where the first
mentioned activity is transitional to a complete action, and this is
quite common when the transitional activity is one of motion, as in
Mt 25:9 πορεύεσθε ... καί αγοράσατε, go ... and buy, and Jn 1:46 ίρχου
καί ϊδε, come and have a look: in each of these the movement could
have been urged as a complete action, but the emphasis has been
placed on its beginning, while the subsequent activity could hardly
be urged as anything but a complete action in the context. Slighdy
more complex is Mk 14:13 υπάγετε ... καί απαντήσει ύμϊν άνθρωπος
... ακολουθήσατε αύτω, go ... and, as you go, a man will meet you ...
follow him.SB Here, of course, the act of following is a specific one
which in the context needs to be treated as a whole: more often in
the Gospels the notion of following appears as a continuing process,
as in Lu 18:22 πώλησον καί διάδος ... καί ακολουθεί μοι, sell... make
a distribution ...be my follower. An instructive illustration of the use

36
Cf. the preceding verse, Εγειρε καί περιπατεί.
37
Cf. Jn 4:35 επάρατε ... καί θεασασθε, raise ... and take a look, where the two ac­
tivities are seen as complete, irrespective of their logical connection.
58
Cf. Ac 9:6 άνάστηθι καί εΐοελθε ... xal λαληθήσεταί..., where the entering gains
strength as a complete action after which the telling can take place.
214 Κ. L. MCKAY

of these aspects in verbs of motion is to be found in Ac 8:26 and

29: άνάστηθι καί πορεύου, get up and start going, and πρόσελθε καί
κολλήθητι, approach and join (both specific and complete actions).
In Mt 3:3 (Is 40:3) ετοιμάσατε gives the overall command to be
carried out in its entirety, and then ποιείτε, echoing it with further
detail, is imperfective to show the parallelism: there must be a pro­
cess of making straight within the total preparation. 39 The principle
is similar, but not identical, in Mt 10:8, where the
iterative/distributive imperfectives of θεραπεύετε ... έκβάλλετε are
summed up in the aorist δότε (δωρεάν being, of course, emphatic).

Positive Imperative: Third Person

There seems to be no reason why the aspectual usage of the third
person positive imperative should not be the same as that of the se­
cond person, and an examination of the NT text proves this to be
the case, although the spread of examples is different, and the
change from the direct approach of the second person to the more
detached impersonal approach of the third person may be a signifi­
cant contextual factor in the choice of aspect.40 An example of
essentially the same command being presented both ways is in Mt
27:40 κατάβηθι and 27:42 καταβάτω. Also closely related and re­
questing a specific definite complete action aire the verbs of Lu 7:7
είπε ... καί ¿αθήτω δ παίς μου, speak the word and let my boy be healed.
Some other examples of specific aorist actions are Lu 22:36
άράτω ... πωληαάτω ... άγορασάτω, he must take ...he must sell... and
buy, Mt 8:13 γενηθήτω, let it happen. In Mt 26:42 γενηθήτω το θέλημα
σου, your will be done, also the reference is to a specific imminent
event, whereas the same clause in Mt 6:10 is more general, but
both request complete fulfilment.41 With both these contrast Lu
22:42 μή το θέλημα μου άλλα το σον γινέσθω, where the comparison
of the two wills leads to greater emphasis on the process (in the com­
ing event let your will be the one being carried out).*2
39
This does not mean that ποιήσατε could not have been used. Note that in Jn
1:23, where a condensed version is given, the verb is εύθύνατε.
40
Cf. the comments above (p. 212) on Lu 23:21 and 23.
41
The first three petitions of the Lord's Prayer in Mt 6:9, 10 are general but
complete. Έλθάτω does not mean continue to come, as Turner Gram, suggests:
whatever the wider context of the coming of the Kingdom in the Gospel, this
prayer is for its complete coming, which is properly expressed by the aorist.
42
In Mt the parallel to this is 26:39 ούχ ως έγώ θέλω αλλ' ως σό, with εστίν
understood (it is not a matter of what I want ... ). Of course γενέσθω could have been
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 215

The aorist is found with iterative/distributive force, signalling

completeness on each occasion, in Mt 5:31 δότω, 10:13 έλθάτω ...
έπιστραφήτω; and in a general complexive sense with a stative verb
in Mt 5:16 λαμψάτω. In Mt 16:24 άπαρνησάσθω ... καί άράτω ... καί
άκολουθείτω, he must deny ... take up ... proceed to follow, aorist and im­
perfective aspects are combined appropriately. It should be noted
that the context is particular enough, only the whoever is wanting to
... which defines the subject of the imperatives introducing a
distributive note; but the metaphoric force of άράτω and (less
noticeably) άπαρνησάσθω helps to give them a rather more general
application, or at least involves the possibility of a series of related
decisions. There is an interesting distinction in Lu 3:11 μεταδότω
... ομοίως ποιείτω, he should share/give one away ...he should do likewise:
although the same sharing seems to be involved in both verbs, the
giving away of a cloak must be a once for all action, whereas the
sharing of one's food is likely to be a continuing possibility, and the
aspects point to this difference.
The imperfective of a stative verb is used iteratively in Mt 5:37
Ιστω, always let (your word) be; in 15:4 (Ex 21:17, v.l. τελευτήσει)
τελευτάτω the imperfective must refer to due process for a capital
offence, and not to summary execution; 4 3 in 19:12 χωρείτω, let him
be receptive/go along with, the fact that receive is a suitable translation
into English idiom should not obscure the way in which the Greek
expression must have developed; in the parenthetic νοείτω of 24:15,
whether we conceive the appeal to the reader as specific to each one
or general to all, the understanding required is a continuing activi­
ty. In I Cor 7:2,3 έχέτω ... έχέτω ... άποδιδότω, each must continue
to have ... keep on giving, the stative and action verbs respectively in­
dicate continuing state and iteration, both appropriately in the im­
perfective. In Jesus' frequent appeal for attention, as in Lu 8:8 ó
?χων ώτα ... άκουέτω, anyone who can hear should listen, the imperfec­
tive suits the nuance of continuing attentiveness, but in Lu 16:29
άκουσάτωσαν meets the appeal for a message with a brusque sugges­
tion where the attention should be turned, the implication of this
aorist balanced against Ιχουσι being that the opportunity they have
is not being used. In Rev 2:7 άκουσάτω also a specific message is
being presented to those who should be able to understand, and an

used (and is a v.l., with rearrangement of clauses): it would simply give the same
prayer a different emphasis.
43
The same aspect with a different verb is used in Lev 20:9 θανατούσθω.
216 Κ. L. MCKAY

appeal to take it in rather than to be attentive to its implications is

appropriate. In I Cor 16:22 ήτω the imperfective stative verb im­
plies being in a continuing state of rejection; /but in the context of
16:14 πάντα ... γινέσθω, let everything be done, the effect of the action
verb in the imperfective is iterative. In I Cor 1:31 (Jer 9:24) ó
καυχώμενος ... καυχάσθω there is not only continuity but a
parallelism with the participle, so instructing what should be the
content of any legitimate boasting (let the boaster's boast be . . . ) . In Mt
24:16 φευγέτωσαν, they must make for ..., the imperfective is incep-
tive/conative. Whatever the contextual realizations, the aspectual
use of the third person positive imperative in the New Testament
is shown by this sample of examples to conform to that of the second
person positive imperative, and in the remaining sections second
and third persons will be treated together.

Prohibitions
Negative commands, or prohibitions, in the New Testament
follow the same pattern as in classical Greek, and the aspectual
usage in them remains essentially unchanged.44 In the aorist the
subjunctive replaces the imperative regularly in the second person
and occasionally in the thinj person, while the imperative is used
in the imperfective and mosdy in the third person aorist. Within
this system there appears to be no difference in meaning between
imperative and subjunctive. The effect of the aorist is to prohibit
or advise against an activity as a totality, whether that activity is
shown by the context to be a single action, momentary or extended,
or a series of actions, and whether it has already begun or not. The
imperfective prohibits or advises against an activity as a process,
and according to context can imply do not begin/try to ..., do not con­
tinue to ..., do not habitually ... As with the positive use of the im-

44
J. P. Louw in ' O n Greek Prohibitions", Acta Classica 2 (1959), 43-57, gives
a useful account of prohibitions, showing that aspectual values in classical and NT
Greek are essentially the same. Starting from Apollonius Dyscolus, who ap­
preciated the structure of ancient Greek even if his explanations are not as
thorough as we would like, Louw weakens his exposition with undue emphasis on
duration as a translation of παράτασις and terminating point for συντελείωσις. The
former is better taken in his alternative (but rarely mentioned) rendering, in its oc­
currence, or, in my terms, as an ongoing process; and the latter is surely better taken
as something like completeness (treating as a whole). This misunderstanding of
αυντελείωσις is the basis of the criticism of Louw (and A. Poutsma) by W. F. Bak-
ker in The Greek Imperative, p. 39.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 217

perative there are contexts in which there is not much difference

between the effects of these two aspects, and others in which the dif­
ference is clear and significant.45
The aorist is appropriate whenever a firm and definite prohibition,
whether specific or general, is intended, mostly when the activity
has not begun, as in Mk 9:25 ίξελθε ... χαί μηχέτι βίσέλθης, come out
and do not return, Mt5:17 μήνομίσητε, don't get the idea, 6: 4μήγνώτω,
it must not find out at all, 10:5 μή άπέλθητε, don't go off,46 10:9 μή
κτήσησβε, don't obtain, 24:17 μή καταβάτω, he must not come down;*7
but it is sometimes used for an activity that is already in train, as
in Ac 16:28 μηδέν πράξης ... κακόν, do no harm.
The imperfective is used for negative, as for positive, attitudes
and habitual activities, as in Ro 12:14 ευλογείτε χαί μή χαταρασθε,
invoke blessings, not curses, 13:8 μηδενί μηδέν οφείλετε, don't have any
debts at all (the double negative adds an emphasis of completeness
which might have been differently achieved with an aorist), I Tim
5:22 μηδενί επιτεθεί, do not lay ... on anyone, Heb 13:2 μή έπι-
λανθάνεσθε, always remember/do not forget. It is also used to urge cessa­
tion of an activity already begun or prepared for, as in Rev 5:5 μή
κλαίε, don't cry/stop crying (cf. 4focXottov),Ro 14:15 μή ... άπόλλυε,
don't ruin (in the context of the hypothetical ουκέτι κατά άγάπην
περιπατείς), Ι Cor 14:20 μή παιδία γίνεσθε, don't go on being/trying to
be children.48 In a different context it may not imply continuity, but
beginning or attempting, as probably in Mt 6:16 μή γίνεσθε ...
σκυθρωποί, don't assume a gloomy air (but the iterative force may be
the dominant one), 9:30 μηδείς γινωσκέτω, ket nobody recognize/begin
finding out.49
In Mt 17:7 μή φοβεισθε clearly means don't go on being afraid (cf.
6 έφοβήθησαν); in 10:31 it need not have that effect and may as easi­
ly be don't start being fearful, depending on Jesus' perception of what
45
In McK. Exeg. 50 I have commented on the difference between Mt 5:36 and
Jas 5:12, not appreciated by Turner Gram. 75. See also McK. Gram. App. A
§§ 12, 13 for μή £λλως ποίει and μή άλλως ποίησης in different contexts.
46
Note the balancing positive imperfective πορεύεσθε, but let your journey be ... in
verse 6.
47
The v.l. χαταβαινέτω, while less likely in the context, is not impossible (he must
not try to go down).
48
So also, most probably, Jn 20:17 μή μου Απτού, which I discussed in "Some
Linguistic Points in Marxsen's Resurrection Theory", Expository Times 84 (1973),
331.
49
Cf. Mt 6:4 μή γνώτω (quoted above), where the possibility of knowledge is
more under one's own control. See n. 29 above.
218 Κ. L. MCKAY

their feelings were; in 10:26 μή ... φοβήθητε is don't getfrightened,or,

if he thought they were already afraid, it is a firm stop being afraid',
1:20 μή φοβηθής is clearly put away your fear, for while Joseph was
showing concern for Mary he had firmly decided (19 έβουλήθη) to
divorce her, and here he is being instructed to reverse this decision.
In Mt 6:25 μή μεριμνάτε, don't be anxious, is a general warning
against the natural tendency to anxiety or its habitual exercise,
similar to μή θησαυρίζετε in verse 19. The change to μή ...
μεριμνήσητε in verse 31 in essence treats the anxiety either complex-
ively as a totality, or inceptively (don't get anxious), μεριμναν being
a stative verb, and the effect of the sharper focus, whichever way
it is understood, is a more emphatic prohibition in the light of the
teaching that has followed since μή μεριμνάτε. In Lu 12:11 μή
μεριμνήσητε (v. 11. (προ)μεριμνατε) is more specifically inceptive, urg­
ing a rejection of anxiety on any occasion when things start to go
wrong (είσφέρωσιν refers to the process of arrest, begun but not com­
plete); while μή μεριμνάτε in 12:22 is the same as in Mt 6:25.
A comparison of two similar discourses which have some signifi­
cant differences can be useful. In Mt 10:26 μή οΰν φοβηθήτε αυτούς
is resumptive, summing up the lesson of the preceding section, and
appropriately making this a total prohibition. There follows in 27
είπατε ... κηρύξατε (v.l. κηρύσσετε) the command to speak boldly: the
context is a general one, but there are to be no half measures, so
the aorist is appropriate. Then in 28 the attitude which should ac­
company this speaking is expressed by the imperfective μή φοβεΐσθε
and a parallel positive φοβεΐσθε (v. 11. φοβηθήτε ... φοβήθητε). Finally
in 31 μή οΰν φοβεΐσθε (v.l. φοβηθήτε) comes in the midst of a
reassurance. In Lu 12:2 there is no prohibition comparable to that
in Mt 10:26, and the reference to proclamation is passive and more
general in effect. The instructions to the disciples begin at verse 4
with a vigorous μή φοβηθήτε and a fuller statement of its object than
there is in Mt 10:28, followed up in 5 with υποδείξω δέ ύμΐν τ£να
φοβηθήτε· φοβήθητε ... τούτον φοβήθητε. There is here no reason for
filling in background attitudes, and whether one takes them as com­
plexive or inceptive the firmness of the aorists is very compatible
with the rhetorical insistence on the point. At verse 7, however, the
gender reassurance seen in Mt 10:31 is found, with μή φοβεΐσθε. In
a similar encouragement in Lu 12:32 the imperfective, μή φοβοΰ, is
again found.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 219

In Jn 7:24, after criticizing the Jewish leaders for inconsistency,

Jesus proceeds with μή κρίνετε, don't go on judging/stop judging, and
then concludes with a firm κρίνατε. Although the context makes it
clearly general in application and the emphasis is on the type of
judgment (δικαίαν as against κατ'δψιν), its contrasting effect makes
the aorist preferable to a matching κρίνετε, which could mean much
the same but might be open to the weaker interpretation try to judge.

Future
Instead of the imperative the future indicative is sometimes
found expressing commands in the New Testament. As the future,
in classical as well as koine Greek, expresses intention or volition
more than simple futurity, and in some of its uses overlaps with the
subjunctive of other aspects, this is not a surprising feature. It is to
be noted that although the future expressing the equivalent of a
command occurs in classical Greek,50 and seems a natural feature
of Greek, its NT occurrences are mainly in quotations from the Old
Testament and Hebrew tradition. In a few of these, such as Mt
21:13 (Is 56:7) κληθήσεται, will be called, it could be argued that the
future is basically prophetic rather than prescriptive, but in the con­
text of a divine message the distinction can only be theoretical.51 In
the summary of commandments in Mt 19:18,19 ού φονεύσεις, ού
μοιχεύσεις, etc., stand alongside the imperative τίμα; and in Lu
18:20 and Jas 2:11 the form preferred is μή μοιχεύσης, μή φονεόσης,
etc. In Mt 5:43 Jesus quotes αγαπήσεις and then proceeds in 44 with
αγαπάτε. In Lu 13:9 έκκόψεις probably has the same kind of force
as εκκοψον in 7 (although advice rather than instruction), but it
might be something like you will be able to cut. Κληθήση in Jn 1:42
(the only example of the usage in this Gospel) is as much prophetic
as mildly prescriptive. In Lu 22:10 έρεϊτε seems to be clearly parallel
with ακολουθήσατε: follow ... say ..., and in 17:4 αφήσεις repeats with
emphasis the meaning of δφες in the preceding sentence. The use
of δψεσθε in Mt 27:24 and Ac 18:15, both spoken by Roman of-

50
McK. Gram. § 28.3.1, W. W. Goodwin, Syntax of the Moods and Tenses of the
Greek Verb §§ 69, 70, R. Kühner u. B. Gerth, Ausführliche Grammatik der griechischen
Sprache § 387.6. In classical Greek the future does not appear to be used much with
a negative in this sense, and when it is, μή is sometimes found instead of ού.
51
The angel's words in Lu 1 are similar.
220 Κ. L. MCKAY

ficials, might be a translation of uideritis,52 but in Mt 27:4 δψχι is

spoken by Jewish leaders.
The use of ού μή with a future indicative or aorist subjunctive for
an emphatic statement of negative futurity sometimes occurs in
contexts which give them a prescriptive force, as in Mt 15:6 ού μή
τιμήσει, will not be obliged to honour, 21:19 ού μηχέτι ... γένηται (v.l.
γένοιτο), there will bellet there be no more. In most NT contexts where
the future is found with ού μή its form is very similar to the aorist
subjunctive, which is found in some witnesses, but in ancient Greek
generally there seems to be a complete overlap of function between
the aorist subjunctive and the future in this construction, which is
as likely to be a statement of simple futurity as any other use of the
future indicative. The extra emphasis provided by ού μή is par­
ticularly suited to the same total exclusion of an activity as is sig­
nalled by a negatived aorist, so it is not surprising that only the
aorist subjunctive is found as an alternative to the future indicative.
In Col 2:8 βλέπετε μή τις ... Ισται and Heb 3:12 βλέπετε ... μήποτε
?σται... the μή clauses are probably dependent, like that in Mt 7:6
(where some witnesses have subjunctive instead of future).

Exhortation
Before proceeding to other alternatives to the imperative, and
their subordinate forms, I turn briefly to the closely related area of
first person subjunctive exhortations. Most of these are in the
plural, because most commonly they occur as an extension of an
imperatival intention,53 and the inclusion of the speaker in the ac­
tivity urged on the hearers can in some circumstances be merely a
polite formality. What concerns us here is that the aspectual usage
in first person exhortations is the same as in the other imperatival
usages noticed above. In Mt 21:38 άποχτείνωμεν ... χαί σχώμεν, let
us kill ... and get control of ..., in spite of the ambiguity of form of
the first verb, the aorist is just what we would expect in a reaction
to a new opportunity to settìe a problem finally. In Mk 4:35
διέλθωμεν, let us cross over, the aorist is again natural, although it
would only take an emphasis in the speaker's mind on the details
of preparing for the crossing, instead of its contemplation as a
52
This use of the Latin future perfect (note the ambiguity of form with the
perfect subjunctive) is probably a relic of its earlier use as an aoristic future.
53
Note the switch to second person in Ro 13:14 ¿νδύσασθε after the first person
verbs of the preceding verses mentioned below.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 221

whole, to make διερχώμεθα appropriate. The suggestion in Jn 19:24

μή σχίσωμεν αυτόν, άλλα λάχωμεν, let's not tear it but draw lots, is a
decisive one, irrespective of whether a suggestion to tear the robe
had been made or not. In Ro 13:12,13 άποθώμεθα ... ένδυσώμεθα ...
περιπατήσωμεν, let us put away ... put on ... walk!behave, the context
is set by verse 11 so that even the behaviour is to be looked at as
a complexive whole. In Heb 4:11 σπουδάσωμεν, let us hasten!be
serious, the aorist may be complexive or inceptive, as σπουδάζειν, like
περιπατεΐν (at least in its metaphorical sense), is a stative verb.
In Mt 26:46 άγωμεν, let us be on our way, the imperfective is used
in the same way as it is in the imperative with verbs of motion
where the emphasis is on the beginning of a journey. There may
be something of the same in Heb 4:16 προσερχώμεθα, let us approach,
but this may be iterative, or descriptive (let our approach be one μετά
παρρησίας), parallel with χρατώμεν, keep holding on, in verse 14: in
Greek terms all of these are possible, and we can only judge from
the context whether any of them is more significant to us. In Ro
14:13 μηκέτι ... κρίνω μεν, let us stop judging, the ambiguous verb
form is shown by its setting to be imperfective, and contrasts nicely
with the aorist κρίνατε which follows. It is not surprising that in the
general exhortations of the Epistles there are many imperfectives,
as in Ro 14:19 διώκω μεν, let us (habitually) go after, I Cor 5:8 έορτάζω-
μεν, let us feast, 10:8 μηδέ πορνεύωμεν, nor indulge in sexual licence (the
emphasis here appears to be less on discontinuing than on not tak­
ing up the habit, but the verb form suits both).
In Mt 27:49 ϊδωμεν is preceded by the imperative δφες, which
may be vividly singular to indicate individuals in the crowd talking
to their neighbours, or may have become a somewhat standardized
expression, like ίδού, which having originally been a normal aorist
imperative, had become a particle.54 Where δφες occurs with a first
person singular subjunctive it seems to be a little more specifically
allow me, as in Mt 7:4 δφες έχβάλω, let me cast out, because mostly
a speaker's desire for his own activity tends to be expressed overtly
in circumstances where he wishes to gain the approval of another,
or in some other way to draw another's attention to it.55 Whatever

54
The use of the active singular t8t in a plural context, as in Mt 26:65 and Mk
3:34, illustrâtes the way in which this development apparently took place, but here
too the aspect is eminently suitable.
55
Hence also 8topo, as in Ac 7:34 (Ex 3:10), and the classical fife, etc.
222 Κ. L. MCKAY

the circumstances, however, the choice of aspects is found to fit the

same pattern as in the imperative.
That is true also in the interrogative construction related to ex­
hortation, the deliberative question. This mostly occurs in the first
person, but as the first person singular aorist subjunctive of many
verbs has the same form as the future indicative, which has a closely
similar meaning, there are some uncertainties. In Mk 12:14 δώμεν
ή μή δώμεν; are we to give or noti the emphasis is entirely on the identi­
ty of the activity, so it is viewed as a whole.56 Similarly the various
groups simply seek a specific answer in Lu 3:10, 12, 14 τί
ποιήσωμεν; what must we do?, but in Jn 6:28 τί ποιώμεν ίνα έργαζώμεθα
... ; by what behaviour can we be doing ... ? the parallelism of the activity
with that in the subordinate clause puts a greater emphasis on its
continuity. In Ro 6:1 έπιμένωμεν ...; are we to remain ...? the con­
tinuation of the existing state is signalled, while in 6:15
άμαρτήσωμεν; (v.l. -ομεν) it is the deliberate act of sin. In I Cor
11,22 τί είπω; what can I say? the aorist suggests the writer's exhaus­
tion of the possibilities, whereas in Heb 11:32 τί ?τι λέγω; how am
I to continue? the imperfective suggests the abundance of examples
to choose from. Occasionally a third person subjunctive is found,
corresponding to the third person imperative. Such are Lu 23:31
τί γένηται; what is to happen? and Ro 10:14 πώς ... έπϋκαλέσωνται ...;
how are they to call ...?, both with aorists to suit their contexts.

Infinitive
One alternative to the use of an imperative is the use of a first
person verb of commanding, urging, advising, etc., with a depen­
dent infinitive, as in Mt 5:34 λέγω ύμίν μή όμόσαι δλως, / tell you not
to swear at all, Ro 12:1 παρακαλώ ... ύμας ... παραστήσαι ..., / urge
you ...to offer... In these the aspect of the infinitive is the same as
the imperative or subjunctive would be if λέγω, παρακαλώ were in
parataxis and a direct command were given (μή όμόσητε,
παραστήσατε). The same principle applies whenever an infinitive of
indirect command is found with a verb of commanding, etc., in
narrative or in any other type of context. Some examples are Mt
8:18 έκέλευσεν άπελθείν, he gave orders for going away (he said άπέλθατε,
or, more likely, άπέλθωμεν), 14:9 έκέλευσεν δοθηναι, he ordered that it
be given (δοθήτω), Tit 2:1, 2 λάλει α πρέπει ... πρεσβότας νηφάλιους
56
Note also δούναι (14) and άπόδοτε (17).
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 223

είναι, say/teach what befits ... namely, that the old men be temperate
(πρεσβυται ... ίστωσαν or 2στε), I Tim 6:17 τοις πλουσίοις ... παράγ­
γελλε μή ύψηλοφρονεΐν, μηδέ ήλπικέναι ..., instruct therichnot to be ar­
rogant, nor to hold great hopes ... (μή ύψηλοφρονεΐτε, μηδέ ήλπικότες
?στε).57
Occasionally in ancient Greek the infinitive is found replacing an
imperative in a direct command,58 and there seems to be an exam­
ple of this usage in Ro 12:15 χαίρειν ... κλαίειν, rejoice... weep, where
the patterns is set by ευλογείτε and μή καταράσθε immediately
preceding.59 Other alleged examples are probably best explained in
other ways: in Lu 9:3 ϊχειν may be a consecutive or purpose in­
finitive (so as/in order to have); in Ac 23:24 παραστήσαι may be depen­
dent on εΐπεν in the preceding verse, the writer having changed
from direct to indirect speech after the list of items intervening after
ετοιμάσατε; and in Phil iii 16 στοιχείν is almost certainly dependent
on φρονώμεν (15), being foreshadowed by τοΰτο before the construc­
tion is complicated by the addition of related points. Of course,
whatever the explanation of each, the aspect of the infinitive is the
same as it would be if a direct imperative had been used.

"INA Clauses
Instead of an infinitive a Vva (or 8πως) clause is sometimes found
depending on a verb of commanding, etc., and the aspect of the
subjunctive in such a clause is that appropriate to the imperative it
represents. Such are Mt 8:34 παρεκάλεσαν δπως μεταβη, they begged
him to go away (μετάβηθι or -βα), I Gor 1:10 παρακαλώ hi υμάς ... Ινα
το αυτό λέγητε, / urge you to be in agreement. Sometimes such a clause
is used independently, as in Eph 5:33 ή δέ γυνή ϊνα φοβήται ..., the
wife must respect ..., which balances ίίκαστος ... άγαπάτω; and
possibly in I Tim 1:3, where tva παραγγείλης could be argued to be

57
This indirect evidence of the perfect imperative can be added to the direct
examples referred to in McK. NT Perf. 324 f.
58
In classical Greek the subject of the imperatival infinitive, when its case is
expressed or implied, is regularly in the nominative if second person, and ac­
cusative if third person (which is less common). This fact is sometimes overlooked
by commentators. The explanation may be a suppressed Βος/8ότε, which would
produce such a case variation according to person, but if so it is likely that the sup­
pression became so normal that it was entirely forgotten.
59
When χαίρειν is used as an opening greeting in a letter, as in Ac 15:23, it must
be understood as depending on an unexpressed εΰχομαι or equivalent verb.
224 Κ. L. MCKAY

dependent on an unexpressed παρακαλώ implied by the preceding

παρεχάλεσα. But the aspectual significance is clear enough.

Participles
Ancient Greek had a great tendency to vary all kinds of construc­
tion by the use of participles, which could be substituted for almost
any kind of subordinate clause and also in many circumstances
where one might have expected parataxis rather than hypot axis. In
theory one would expect a proposition A and/but Β to be refor­
mulated as when! if I because (/etc.) A, Β on the way to substituting a
participle for A, but the participle became so common that in prac­
tice there was often no conscious working through an intermediate
stage when a speaker chose to use a participle. Nowhere is this truer
than when participles replace imperatives and their equivalents,
and this possibility is nowhere exploited more fully than in the New
Testament. It may be true that native speakers of Aramaic had a
tendency to push the limits of participial usage beyond those of nor­
mal Greek, but this would be encouraged by the flexibility of any
such limits, and in judging such matters we need to remember that
the NT Episdes, in some of which this exploitation is most
marked,60 have no exact parallel in purely Greek literature. Our
present concern, however, is with aspect, and it is sufficient to note
that when a participle replaces an imperative or jussive subjunctive
it preserves the aspect the verb would have had in that mood.61
As a participle registers only aspect and voice, any other
significance it may have must be inferred from its context, and in
some contexts there is a certain amount of ambiguity. For example,
in Lu 5:24 ϊγειρε καί δρας (v.l. άρον ... καί)62 ... πορεύου, get up, pick
up ... and go on your way, the participle would make good sense as

60
It ought perhaps to be noted that there appear to be no participles linked with
direct imperatival expressions in Jn, Ro, Gal, Phil, Phlm, I Jn, II Jn, III Jn, while
in II Cor and Rev there are only the rather un-Greek periphrases μή γίνισθβ
έτεροζυγοδντες (II Cor 6:14) and γ{νου γρήγορων (Rev 3:2). This absence is less a
matter of language than of style and subject matter.
61
Apart from the circumstances mentioned in n. 64 below, the only situations
in which a participle appears to change an aspect is when it replaces a presumed
future form, but in such cases an intermediate subordinate clause containing a
subjunctive can usually be postulated, and the aspect of the participle is that of
the subjunctive.
62
V.l. as in the parallel passages Mt 9:6 and Mk 2:11.
 ASPECT IN IMPERATIVAL CONSTRUCTIONS 225

63
equivalent to 8ταν δρης, when you have picked up. There can be no
doubt, however, that in II Tim 2:23 ... παραιτου, είδως ..., keep clear
of... because you know ..., the participle is causal, and not much less
doubt that in II Tim 4:13 ερχόμενος φέρε, when you are coming, bring
with you, it is temporal. In most N T contexts in which participles
are associated with imperatives there can be little doubt that they
represent paratactic imperatives. Such are Mt 2:8 πορευθέντες εξ­
ετάσατε, go andfind out (πορευθηχε καί), 2:13 εγερθείς παράλαβε, get up
and take (έγέρθητι και), 64 6:6 εΰτελθε ... καί κλείσας ... πρόσευξαι, go in
..., shut... and offeryour prayer (κλεΐσον καί), 20:8 άπόδος ... άρξάμενος,
pay ... and, in so doing, begin (και δρξαι), Mk 15:30 σώσον ...
καταβάς,65 save ... and come/by coming down, Lu 6:35 δανίζετε μηδέν
άπελπίζοντες, lend and, when you do, be expecting nothing in return (μηδέν
απελπίζετε), 9:60 άπελθών διάγγελλε, go away and start passing on the
news, Ac 10:20 άναστάς κατάβηθι καί πορευου ... μηδέν διακρινόμενος,
get up, go down, and proceed onyour way without hesitation/and have no
hesitation about it, 16:37 έλθόντες αυτοί ήμας έξαγαγέτωσαν, they must
come themselves and take us out, Jas 5:1 κλαύσατε όλολύζοντες, burst into
tears and go on wailing, 5:14 προσευξάσθωσαν ... άλείψαντες, let them pray
.., and anoint. Of course the same use of participles is found with
first person subjunctive exhortation, as in Heb 12:1 αποθεμένοι ...
τρέχωμεν ... άφορωντες ..., let us lay aside ... and run ..., andas we run
let us be looking intentily...; and this is preceded by another participle
έχοντες which is clearly causal (since we have).
In Mt 28:19, 20 πορευθέντες ... μαθητεύσατε ... βαπτίζοντες ...
διδάσκοντες the command go and make disciples is presented as the
overall programme, with baptizing and teaching as explanatory
details in its working out, and so the imperfective is used for these
last. On the other hand in Ac 22:16 άναστάς βάπτισαι καί άπόλουσαι

63
... have picked up is English idiom: the aorist signals complete action, not com­
parative pastness, and in a slightly different context could idiomatically be
(when/whenover you) pick up.
64
It is to be noted that whereas the imperfectives έγειρε, οπαγε, έρχου, etc., are
common when a separate imperative is used (see p. 6 above), the aorist participles
εγερθείς, έλθών, etc., are normal when the participle stands in place of a preceding
imperative (but less so when it is in place of a temporal clause as in II Tim 4:13).
The reason is not a confusion of the aspects, but rather that when the emphasis
is on beginning or trying to go, etc. an explicit imperatival expression was usually
preferred, and when the completeness of the initial activity was emphasized there
was naturally a kind of briskness which favoured the slighdy tighter relationsship
which the participle could provide.
65
For the v.ll. καί κατάβαΛβηθι cf. Mt 27:40.
226 Κ. L. MCKAY

... έπικαλεσάμενος ..., get up, accept baptism and wash away ... and call
on..., the whole emphasis is on the specific responses involved in
conversion, so the concluding participle is aorist.
In some hortatory contexts it is not surprising that grammatical
anacolutha occur in the use of participles. For example in I Pet 2:
11, 12 παρακαλώ ... άπέχεσθαι... Ιχοντες, I urge you to keep away from
... and keep ..., the participle carries on the exhortation as if άπέχεσθε
had been used instead of παρακαλώ (ύμδς) άπέχεσθαι. So also in Col
3:16 ó λόγος ... ένοικείτω έν ύμΐν ... διδάσκοντες ..., let the word dwell
in you; teach one another ..., the participle is nominative as if
something like ίχετε τον λόγον had preceded. Such adaptations of
grammar to equivalent sense, usually from an impersonal into a
personal construction, are found even in more formally literary
contexts in classical Greek, and are a feature of colloquial speech
in all languages.

Conclusion

As I mentioned earlier, the examples quoted in this article are a

more or less random sample to illustrate the aspectual patterns that
could equally well be illustrated by the larger number not quoted
here. Many of them do not significantly support my hypothesis, but
none is incompatible with it. On the other hand many of them seem
more significant in the light of this hypothesis than in that of any
other. In looking for the relevance of the aspect of all the verb forms
in each context we need to be aware that there may be complex in­
terrelationships, involving the writer's unexpressed assumptions,
and to work from the more obviously significant features to the less
clear, recognizing that some aspectual choices may be subjective
but that none is likely to be completely perverse. The ancient Greek
aspectual system has for long been at least pardy misunderstood,
and we need to set aside traditional misconceptions and re-examine
the texts written by people who knew the language better than we
do. 6 6 At least the logic of the message contained in our texts has not
been totally obscured by such misconceptions, and the effect of our
re-examination will be more the removal of imagined problems and
a fuller appreciation of the emphases intended by the writers than
a complete revision of our understanding of the message.
66
I refer, of course, to the knowledge which goes with familiarity and use rather
than to that which can enable the scholar to explain objectively the syntax of the
language.
 ^ s
Copyright and Use:

As an ATLAS user, you may print, download, or send articles for individual use
according to fair use as defined by U.S. and international copyright law and as
otherwise authorized under your respective ATLAS subscriber agreement.

No content may be copied or emailed to multiple sites or publicly posted without the
copyright holder(s)' express written permission. Any use, decompiling,
reproduction, or distribution of this journal in excess of fair use provisions may be a
violation of copyright law.

This journal is made available to you through the ATLAS collection with permission
from the copyright holder(s). The copyright holder for an entire issue of a journal
typically is the journal owner, who also may own the copyright in each article. However,
for certain articles, the author of the article may maintain the copyright in the article.
Please contact the copyright holder(s) to request permission to use an article or specific
work for any use not covered by the fair use provisions of the copyright laws or covered
by your respective ATLAS subscriber agreement. For information regarding the
copyright holder(s), please refer to the copyright information in the journal, if available,
or contact ATLA to request contact information for the copyright holder(s).

About ATLAS:

The ATLA Serials (ATLAS®) collection contains electronic versions of previously

published religion and theology journals reproduced with permission. The ATLAS
collection is owned and managed by the American Theological Library Association
(ATLA) and received initial funding from Lilly Endowment Inc.

The design and final form of this electronic document is the property of the American
Theological Library Association.