The Subjunctive in Independent Sentences in Plautus. I

STOP

Early Journal Content on JSTOR, Free to Anyone in the World

This article is one of nearly 500,000 scholarly works digitized and made freely available to everyone in
the world by JSTOR.

Known as the Early Journal Content, this set of works include research articles, news, letters, and other
writings published in more than 200 of the oldest leading academic journals. The works date from the
mid-seventeenth to the early twentieth centuries.

We encourage people to read and share the Early Journal Content openly and to tell others that this
resource exists. People may post this content online or redistribute in any way for non-commercial
purposes.

Read more about Early Journal Content at http://about.jstor.org/participate-jstor/individuals/early-
journal-content .

JSTOR is a digital library of academic journals, books, and primary source objects. JSTOR helps people
discover, use, and build upon a wide range of content through a powerful research and teaching
platform, and preserves this content for future generations. JSTOR is part of ITHAKA, a not-for-profit
organization that also includes Ithaka S+R and Portico. For more information about JSTOR, please
contact support@jstor.org.

AMERICAN
JOURNAL OF PHILOLOGY

Vol. XVIII, 2. Whole No. 70.

I.— THE SUBJUNCTIVE IN INDEPENDENT SEN-
TENCES IN PLAUTUS.

I. — Facts.

This paper consists of three parts : first, a presentation of the
facts of usage ; second, a discussion of some of the forces which
appear to have affected the mode ; third, remarks upon the poten-
tial and the optative uses and upon the subjunctive in general.

No precise line can be drawn between the independent sentence
and the dependent clause, but for my purpose I have included
most of the cases of the subjunctive in parataxis. Of indirect
questions, however, I have taken only those into which uis, win
and similar words are inserted ; other indirect questions, though
many of them are actually paratactic, are not included in the lists.
The paratactic prohibition with ne passes at once into the ne
clause, and is therefore not given. No sentences are included
which contain a formal protasis ; this involves the omission of a
considerable number of cases where the subjunctive is undoubt-
edly independent of the protasis, but they are in all respects
similar to the cases given in the lists and the material is sufficient
without them.

In the presentation of the facts there is no classification by
function, such as is usually made. The division is by tense,
person and number, with a subdivision according to the form of
the sentence. This, in full, would be as follows :

I. Affirmative.

A. Independent.

a) Non-interrogative.

b) Interrogative.

134 AMERICAN JOURNAL OF PHILOLOGY.

B. Para tactic.

a) Non-interrogative.
F) Interrogative.
II. Negative (with the same sub-classes).

For all practical purposes this scheme is followed closely
enough by making for each person and number four classes : —
Independent, Interrogative, Paratactic and Negat ve with ne.
Under each class the uses to which the form may be put are
described and illustrated, and this leads in the end to a partial
classification by function. Full lists are given only where the
usage is important or infrequent, but complete statistics will be
found in the tables.

Present, isi sing.

A. Non-interrogative , independent. — As the same form is used
in the 3d and partly in the 4th conjugation for pres. subj. and fut.
indie, the undoubted subjunctive forms are given first and used
as a criterion for distinguishing pres. from fut. where the form is
the same.

Three cases are marked as subjunctive by utinam, Aul. 433,
Epid. 196, Trin. 618. These are all wishes. Pers. 575, modo ut
sciam quanti indicet, is marked by ut, but in meaning is like those
which follow.

The subj. form is found in the following cases : Ba. 1049 quod
perdundumst, properem perdere ; 1058 sed crepuit foris : ecfertur
praeda ex Troia. taceam nunciam ; Trin. 1136 quid ego cesso
hos conloqui? sed maneam etiam, opinor ; Pers. 542 uideam
modo mercimonium. Cf. 575, above, and Ter. Heaut. 273 hoc
quod coepi primum enarrem, Clitipho: post istuc ueniam. All
express the speaker's desire or choice in regard to an act of his
own. The first three are in soliloquy ; Pers. 542 and 575 are in
a dialogue and are answered, but they are also half-soliloquizing.

All the remaining cases (except those of the 3d conjug.) are
hypothetical, dealing with the speaker's action in a supposed
case. The phrase non meream (merear) is found in Ba. 1184
quern quidem ut non hodie excruciem, alterum tantum auri non
meream, Men. 217, Poen. 430, and in a like sense, non emam,
Capt. 274. Almost identical is floccum (ciccuni) non interduim,
Rud. 580, Trin. 994, and, with negative implication, Aul. 672 tam
duim quam perduim. All these assert the speaker's conviction in
regard to his own action in a supposed case, corresponding in
general to the English, ' I would not do it for the world.'

THE SUBJUNCTIVE IN PLAUTUS. I35

In two cases the speaker assures the hearer in regard to his
action in a supposed case. Capt. 237, quod tibi suadeam, suadeam
meo patri, is not above suspicion (suadeo, Cam., Sch.), but Trin.
758 is a clear statement of intention.

The assurance is scarcely more than a view or opinion in True.
495 sine uirtute argutum ciuem mihi habeam pro praefica, and
Aul. 230, Merc. 125 nimis nili tibicen siem, are clearly only
expressions of opinion. In Capt. 237, Trin. 758, Aul. 230, a
clause with conditional coloring precedes, and the same thing is
implied in True. 495.

The text is doubtful in several passages. In Aul. 570 there is
a hiatus and Seyffert supplies quod . . . habeo. This makes good
sense, but I think it sure that non potent could mean the same
thing as nolo potare, which Goetz suggests in the notes. In Trin.
749 the difficulty is in reconciling the text to the following verses,
but adeam, edoceam are correct in themselves, and precisely
similar to cases which will be mentioned in 3d sing., where a half-
indirect subjunctive is used in proposing a plan of action. Men.
982, if si (Bx.) be not supplied, is an extreme case, scarcely
parallel to any of the preceding. St. 208* is a gloss.

Of clearly subjunctive forms, then (omitting tielini), PL uses
only 20, with perhaps two to be added from doubtful passages.
Functionally, they are of four kinds. They express a) the
speaker's choice or desire in regard to his own action; U) his
conviction as to his (negative) action in a supposed case ; c) his
assurance to another person in regard to his action in a supposed
case; and d) his opinion about a supposed case. In c and d
there is usually a clause or phrase which gives hypothetical tone.

As to the verbs which use the same form for 1st sing, of fut.
indie, and pres. subj., it may be said, by way of preliminary, that
the fact that language has found no special form to express
futurity in the first person dicam, as it has in dices, dicas, is clear
proof that there was no strong distinction in meaning. It is
therefore an error to force every case of faciam or dicam either
into a future or a pres. subjunctive.

Some of the frequently recurring soliloquies in which a speaker
conceals himself as he hears the door opening, like Ba. 610 sed
hue concedam, Epid. 103, Cas. 434, closely resemble Trin. 1136,
so that with a different verb we might have had a clear subj. form,
e.g. abeam. The situation in Ba. 798, Rud. 1356 sed conticiscam,
is exactly like Ba. 1058 taceam nunciam. As the periphrastic

I36 AMERICAN JOURNAL OF PHILOLOGY,

form in Ba. 1049 (above) helps to define properem, so in Cist. 657
faciundumst puerile officium : conquiniscam ad cistulam, the two
phrases refer to the same act, toward which the speaker's mental
attitude is the same, and the periphrastic strongly suggests the
sense 'must' for the subj. Cf. the parataxis with necessest, below.
In Asin. 605 sermoni iam finem face tuo : huius sermonem acci-
piam, Aul. 405, Cas. 516 nunc amicine anne inimici sis imago,
. . ., sciam, the verb expresses desire, not futurity, and Merc. 881.
recipiam me illuc, expresses choice, as in parataxis with optumumst.

The modal shading in these cases is faint and it would no doubt
be possible to translate them all as futures, but they are parallel
to the subj. forms given above and to the paratactic uses to be
given later. With them, though less clearly, may be classed the
cases of the subj.-fut. forms with potius, like Epid. 149 ne feceris :
ego istuc accedam periclum potius, Cas. 999 hercle opinor potius
uobis credam quod dicitis, Aul. 767 i refer: dimidiam tecum
po'tius partem diuidam. There is a slight modal shading, deter-
mination taking the form of preference, but there is nothing in
the context to give a hypothetical tone. Cf. also the difference
between uelim and malim, below, and see the careful treatment
of these expressions in Neumann, de fut. in prise. Lat. . . . ui et
usu, Breslau, 1888.

non dicam dolo, Men. 228, Trin. 480, is an introduction to a
following remark and is allied to the frequent use of dicam,
eloquar to introduce a statement. Trin. 90 haud dicam dolo is
exactly similar, though the statement is postponed to 94 ; it
should be followed by a colon, not by a period. The ordinary
punctuation makes it wrongly a reply to the preceding question.
Of the same nature is the phrase deum (maiorum) uirtute dicam,
M. G. 679, Pers. 390, Ps. 581, Trin. 346. It is parenthetical and
introductory to the statement which follows. 1 In Aul. 283 dicam
seems to be similar, though the text is hopelessly corrupt. It is
possible enough to translate ' I may say,' but instead of trusting
to the very uncertain test of translation I prefer for the present to
say that these cases lie in the borderland between the future and
the subjunctive, having resemblances to both.

In a few cases the context shows that the mode expresses an
opinion in regard to a supposed case. In Aul. 232 there is an

1 Cramer, de perf. coniunct. usu potentiali ap. prise, script. Lat., Marburg,
iSS6, p. 54, calls this optative and comp.^a« tua dixerim. Brix on Trin. 346
calls it future.

THE SUBJUNCTIVE IN PLAUTUS. 1 37

ubi clause; in Ps. 358 the hypothetical suggestion is in the
sentence itself; in M. G. 845 it is suggested by the other speaker ;
Amph. 156, 161 are in a long hypothetical passage containing a
si clause. Men. 985 is confused and Bx. is probably right in
making a protasis.

To summarize, there are in 1st sing. pres. the following :

With utinam,

3

modo ut,

1

Subj. forms —

of desire or choice,

4

conviction (neg.),

7

assurance,

2

opinion,

3

Doubtful forms —

of desire and choice,

9

with potius,

(?)

non dicam dolo,

3

deum uirtute dicam,

4

of opinion,

5

34

To this number perhaps two or three should be added from
passages of doubtful text, and if the cases with polius were added,
the number would be somewhat increased.

The remaining cases under this heading are all cases of uelim
and compounds, of which full lists are given.

uelim with parataclic subjunctive. — In the 2d sing, pres., Cas.
234 enicas. || uera dicas uelim; Rud. 511 pulmoneum edepol
nimis uelim uomitum uomas ; Rud. 1067, Men. 909. The taunt-
ing verse in Trin. 351 quod habes ne habeas et illuc quod non
habes habeas is followed by uelim malum in A, by malum in BD.
To the reasons given by Bx. Anh. for believing malum to be the
gloss, I would add the close resemblance of this passage to the
others above and the fact that the 2d sing, always has utinam
or uelim in wishes except in certain formulas and once in the
marriage song, Cas. 822. With the 3d pers., Rud. 877 perii. ||
uerum sit uelim ; and six cases with ueniat, Aul. 670 nimis hercle
ego ilium coruom ad me ueniat uelim, Cas. 559, Most. 1074 nunc
ego ille hue ueniat uelim, Poen. 1288, Ps. 1061, True. 481. In
five of these nunc is used and the person is sometimes in the

138 AMERICAN JOURNAL OF PHILOLOGY.

nom. with ueniat, sometimes in ace. (Aul. 670, Cas. 559, Ps. 1061).
The order is always ueniat uelim. With the 3d pers. of the peri.,
Ba. 334 nescit quid faciat auro. || mihi dederit uelim, Poen. 1206
. . . quod haruspex . . . dixit. || uelim de me aliquid dixerit, Rud.
662, Poen. 570. Once, in Most. 632 nihilo plus peto. || uelim
quidem hercle ut uno nummo plus petam, an ut is inserted
between uelim and the subjunct., with the same challenging and
hostile sense that appears in most cases with paratactic subjunctive.

In ten cases uelim takes an infin. Without subject ace, Cist.
497 di me perdant — || quodcumque optes, tibi uelim contingere ;
Asin. 274. With me, Ba. 530, Cas. 287, Epid. 120, Most. 218 in
anginam ego nunc me uelim uorti. With te, Aul. 120 uelim te
arbitrari med haec uerba . . . tuae rei causa facere; Fragm. 41
uelim ted arbitrari factum. With other subjects, St. 587 edepol
ne ego nunc mihi medimnum mille esse argenti uelim ; Trin. 433.
With a perfect participle, Aul. 504 moribus praefectum mulierum
hunc factum uelim ; Ba. 603 sufflatus ille hue ueniet. || disruptum
uelim ; Cas. 326 ego edepol illam mediam disruptam uelim ;
Cure. 83, St. 191, 613. With an adj., Amph. 834 uera istaec
uelim ; cf. Rud. 877, above. With a direct object, Amph. 1058
animo malest, aquam uelim; Most. 266 nimis uelim lapidem, qui
. . . diminuam caput; Ps. 598, Rud. 211. Absolute, Cas. 464 ut
tibi, dum uiuam, bene uelim plus quam mihi ! (This is the only
passage where ut, which here shows plainly its exclamatory
character, is used with uelim) ; Most. 742 (R. 1 L. 2 iielleni), Pers.
629, Ps. 1070, Trin. 58, Fragm. 38.

As to the meaning of these 44 cases, the following points may
be noted : 1) With a few exceptions, the speaker does not expect
that the expression of his desire will bring about the realization
of the desire. This is the general characteristic which, with
some modifications, is common to all the various forms of
optation. 1 The exceptions are Men. 909, perhaps Rud. 1067,
Aul. 120, Fragm. 41, in all of which the person addressed is to be
the actor. An expression of desire thus directly addressed to
the person who has it in his power to act is of necessity very little
removed from a direct expression of will, and Men. 909 adeas
uelim is only a trifle more polite than adeas or adeas uolo. But
in Rud. 1067 . . . ne uideas uelim it is not in the power of the
second person to avoid seeing, and in Aul. 120, Fragm. 41, where

1 1 venture to use this word in a technical sense, in order to avoid confusion
from the different meanings of the English word wish.

THE SUBJUNCTIVE IN PLAUTUS. 1 39

the dependent phrase is te arbilrari, the thinking or believing is
not strictly an act which depends on the will of the believer.
Possibly Most. 742 should also be called an exception, but the
passage is peculiar in other ways also. 2) In more than half of
these cases the content of the sentence is a curse or a wish which
involves a threat. 3) The paratactic subjunctive is, with perhaps
two exceptions, the kind of subjunctive which in other connections
would be recognized as optative.

In general, the understanding of a modal usage must depend
upon a correct interpretation, and full lists of uelim have been
given in order to enable the reader to see for himself that the
ordinary explanation of uelim as a potential or a subjunctive of
modesty or of mild assertion is absolutely inconsistent with the
facts of usage in Plautus. To translate it 'I should like' is simply
to introduce confusion by the use of a peculiar English idiom.
The subjunctive is optative in character, but the fuller discussion
of this must be postponed until the other paratactic verbs are
given.

mauelim, malim. With paratactic subjunctive, Poen. 1150
abeo igitur. || facias modo quam memores mauelim, Poen. 1184.

With ut clause, Trin. 762 malim hercle ut uerum dicas quam
ut des mutuom.

With infin., Asin. 811 emori me malim, quam haec non eius
uxori indicem, Ba. 465, 490, 514, 519 (a gloss), Men. 720, Merc.
356, 889, Pers. 4, Poen. 827, True. 260, 743 (a somewhat uncertain
conjecture), Vid. no. The subject is me, expressed or implied,
except in the last case.

With direct object, Poen. 151 istuc mauelim.

With adj. or ptc, implying esse, Aul. 661, Epid. 119, Poen.
1 214, True. 742.

With dependent phrase implied in the context, though not
expressed, Capt. 858, Rud. 570, True. 422.

It will be noted that malim takes the infin., with me implied or
expressed, much more generally than uelim, and that it has the
paratactic subjunctive only twice, in 2d sing. pres. With these
exceptions, the relationship to uelim appears clearly ; the desire
(preference) is expressed without expectation that the expression
will lead to its realization, as in all optations, and a large number
of cases contain a kind of self-curse {emori me malim, quam;
perire me malim; mendicum malim mendicando uincere ; arare
mauelim, etc.). The composition with mage, however, by reduc-

140 AMERICAN JOURNAL OF PHILOLOGY.

ing the wish to a preference somewhat obscures the optative
sense and gives more apparent reason for calling malim potential.
Compare what is said later on the effect of potius.

nolim is used three times : Amph. 86 and Capt. 943 with infin.,
Merc. 539, absolutely.

peruelim also occurs three times: with infin., Cas. 862, Epid.
536 ; with ptc. (esse implied), Cure. 102.

In tabular form these uses are :

Parat. subj.

ut

Iniin.

Ptc.

and adj.

Obj.

Absol.

uelim, 16

1

10

7

4

6 = 44

malim, 2

1

13

4

1

3 = 24

nolim,

2

1 = 3

peruelim,

2

1

= 3

74

B. Interrogative sentences, with pres. subj., 1st sing. — Since the
effect of the interrogative form of sentence upon the meaning of
the mode must come up later for discussion, the statement of
usage is made as brief as possible.

I. Questions with quis, arranged (with the exception of the
dicam questions) according to the form and construction of quis.

The pronoun occurs 5 times in the accus. (quern 3, quam 1,
quod 1) ; two of these are in soliloquy and the question is delib-
erative, the others follow an impv. or its equivalent and seek to
learn the desire of the person addressed, quam ob rem (3 cases,
if we include Mil. Glor. 360) follows and repudiates a suggestion
or command, quo modo (pactd), 3 cases, always implies nullo
modo, and in M. G. 1206 seems to imply 'can.'

The adverbs of place are like the pronominal forms, not
idiomatic, ubi, 3 times in soliloquy with quaeram, requiram,
inueniam, all deliberative, once in a question as to the will of the
person addressed, quo follows an impv. 4 times and an implied
suggestion once, in true questions; in soliloquy, deliberative, 3
times, unde (2), once after an impv., with rejecting force, once in
soliloquy.

The adverbs which mean 'how' or 'why' are more idiomatic.
qui is used only in the formula qui ego isiuc credam (tibi) ?, Cure.
641, Merc. 627, 902, 'how do you expect me to believe that?,'
with repudiating force, quin, M. G. 426, is repudiating, ut, Ba.
149, is more nearly hypothetical, under the influence of lubens.

THE SUBJUNCTIVE IN PLAUTUS. 141

quor {cur) is used 13 times. The verb is always of mental action,
posiulem, rogem, negem, cur em (2), mirer, mentiar, miniier,
quaeram (2), suscenseam, adflicter (Ps. 1295), perpetrem (Cas.
701). A few of the questions are addressed directly to the other
speaker, but nearly all have a half-soliloquizing tone, and all
imply a rejection, as in English when the why is made emphatic
with a falling inflection. The negative is non.

Of the 139 quis questions with the 1st sing, pres., 81 are intro-
duced by quid.

quid is the direct object in 54 cases. Of the 18 cases with
other verbs than faciam and agam (which are given separately
below), about half repeat an impv., as in Aul. 651 redde hue. ||
quid reddam ?, with a tone which varies from repudiation to a
distinct question as to the desire of the speaker (cf. True. 789
quid loquar? with Epid. 584 quid loquar uis?). About half a
dozen are in soliloquy and are deliberative.

quid faciam serves so well to illustrate the history and meaning
of quis questions that I give the lists in full.

a) quid faciam? Ba. 634, Cist. 63, 301, Epid. 98, Merc. 207,
565, M. G. 459, Most. 523, Pers. 42, Poen. 357. Of these, Ba.
634 and Merc. 207 are in soliloquy and are deliberative ; the rest
mean ' What do you want me to do ? '

b) quid nunc faciam ? Men. 834, Ps. 1229, both asking for
advice or direction.

c) quid ego faciam? Cure. 589 (deliberative), Pers. 26
(addressed to another person, but half-soliloquizing).

d) quid ego faciam nunc? Epid. 255, like Pers. 26.

e) quid ego nunc faciam? Ba. 857 {nunc ego*), Cas. 549, Cure.
555, Men. 963, M. G. 305, Most. 371. Both Ba. 857 and Most.
371 are answered, but the question was not necessarily addressed
to the other person ; the rest are in soliloquy.

f) quid faciam with other added words : aliud Merc. 568, hoc
. . .postea Most. 346, tibi Ps. 78, huic homini Ps. 13 16. The last
is in pretended deliberation.

There are also some cases with appended protasis.

quid agam is used in the same way. a) quid agam? in
dialogue, Aul. 636, M. G. 363. b) quid nunc agam?, Amph.
1046, Cas. 952, Poen. 351 {agam nunc), all in soliloquy, c) quid
ego agam?, Most. 378 (in soliloquy), Trin. 981 (repudiating an
impv.). d) quid ego nunc agam?, Aul. 274, 447, Cist. 528, all in
soliloquy. /) quid ego nunc own illoc agam?, Men. 568, for
advice.

142 AMERICAN JOURNAL OF PHILOLOGY.

In the accus. of compass and extent quid is generally indistin-
guishable from quid 'why.' Capt. 556 . . . etiam huic credis ? ||
quid ego credam huic ? || insanum esse me, is clearly the pronoun,
and where the pronominal force is distinct the questioning force
is also most distinct. But in general these cases, 20 in all, are
like those with quor, above, in implying that there is no reason
for acting as the other person desires, or at least in implying a
disinclination toward the action. Three cases, Capt. 536, Rud.
447, Trin. 1024, are in soliloquy, and Amph. 41 is in a prologue.

In quid ni (7 times), quippe ?ii (once) the ni has been suffi-
ciently shown to be negative, not conditional, 1 and these sentences
are merely the negative form of those given above.

Cas. 454 deosculer. || quid 'deosculer'? is not really a question
with subjunctive verb, and in Epid. 281, Merc. 887 the text is
entirely uncertain.

Beside these there are 13 cases in which dicam is inserted and
one similar case with prae dice m. They are introduced by various
forms of quis. True. 689 quam esse dicam hanc beluam ?, Cas.
616 qua, abl.; quid, ace, Ps. 744 sed quid nomen esse dicam ego
isti seruo?, Asin. 587, Merc. 516, Pers. 400; quid 'why,' M. G.
1 201 quid te intus fuisse dicam tam diu?, St. 288; quo, Capt. 533
quo ilium nunc hominem proripuisse foras se dicam ex aedibus ?,
Cure. 1, 12; unde, Ps. 966, Rud. 264, Ps. 1305 f. sed die tamen,
. . . unde onustam eeloeem agere te praedicem ? In the last case
it does not matter whether the question is regarded as direct or
indirect ; see Becker in Studemund's Studien, I, p. 160, and cf.
Ps. 709. The introducing word in these questions is for the most
part of a kind that does not greatly influence the meaning of the
mode, and the questions differ from those given above only in
the fact that dicam with the infin. is used as a periphrasis for the
simple verb. In the simpler form these questions would be quae
est haec belua f quid nomen est isti seruo f quid intus fuisti tam
diuf The inserted dicam expresses the same thought as the
English ' What kind of a creature am I to suppose this to be ? '
' What would you have me call that slave of yours ? ' That dicam
is subj. and not future is plain from Ps. 1306 and from the thought.

II. Sentence questions are for the most part exclamatory,
corresponding in form to exclamations with the indicative rather
than to true interrogations.

1 O. Brugmann, Ueber den Gebrauch des condicionalen ni, Leipzig, 1887.

THE SUBJUNCTIVE IN PLAUTUS. 143

egone is used 26 times, including Cas. 117 egon quid faciam
tibi ? and True. 276, where Schoell reads tene ego. All are repu-
diating exclamations, and the will of the other person, which is
repudiated, is frequently expressed in a preceding imperative.
The verb is usually a repetition of the preceding (. . . mecum i
potatum. || egone earn ?) or an amplification of it (Cure. 10 lautus
luces cereum. || egon apicularum congestum opera non feram ?).
In a few cases a verb of speaking or keeping silence is used, or
the force of the repudiation is intensified by patiar, possim (Asin.
810 egon haec patiar aut taceam? emori me malim).

Other forms of pronouns with -ne (tene 2, eamne, tuane, mene,
uosne, meosne) have exactly the same meaning, and patiar is used
in 3 of the 7 cases.

Parallels to these, in which the indicative occurs in exclamatory
repudiation of a statement of fact, may be found in any play.

Questions with -ne appended to a noun are not idiomatic, either
with the indie, or with the subj. Pers. 26 deisne follows quid ego
faciam ? and is half-deliberative. In Poen. 730 quid turn ? homi-
nemne interrogem . . . ? would be a question for advice, but -ne
is a conjecture of Ritschl, following A ; the passage is given
below under parataxis (censen for quid turn, with Pall.).

The two cases with etiamne are for advice, Rud. 1275, 1277.

Without a particle, there are 8 cases of the verb alone or with
introductory quid? repeating and repudiating an expressed impv.,
as in Most. 579 abi quaeso hinc domum. || abeam?, Merc. 749
abi. || quid, abeam?

Four cases with non at or near the beginning of the sentence
repeat and repudiate a negative suggestion, Epid. 588 quor me
igitur patrem uocabas? || non patrem ego te nominem, ubi . . . ?

With the verb at the beginning of the sentence, the exclamatory
and rejecting force is especially clear. Such sentences begin like
those which consist of the verb alone (e. g. Asin. 838 an tu me
tristem putas? || putem ego quem uideam esse maestum ...?),
but run off into added details. Where the verb is not at the
beginning the sentence is short (tibi ego dem? loricam adducamf),
because the exclamatory tone cannot be long sustained. In three
or four cases (Ba. 903 hodie exigam aurum hoc?||exige ac
suspende te, Men. 539 dicam curare? || dicito, and Pers. 26, after
quid ego faciam f) the question is not repudiating. The first two
are, of course, not to be distinguished from futures ; the last is
partially deliberative. In Most. 664 GS. rightly use a period.

1 44 AMERICAN JO URNAL OF PHIL OLOG Y.

Most. 556 quid nunc faciundum censes ? || egon quid censeam ? is
an indirect quotation, and censeam is not properly the verb of the
question.

Amph. 813 quor istuc, mi uir, . . . ex te audio? || uir ego tuos
sim? is defended by Ter. Andr. 915, Hec. 524. It repudiates the
claim implied in mi uir.

The one case with anne, Cist. 518, is an impatient demand ; the
disjunctive questions (4 cases) are all deliberative.

Of the 73 cases of sentence question, four are asked for the
sake of getting advice or direction, and four (the disjunctive
questions) are deliberative. Beside these the only deliberative
questions are the two in Pers. 26, which in form seem to delib-
erate, but in content (deisne aduorser, cum eis belligerem) are
plainly rejecting.

C. Present, isi singular, in parataxis.— The. value of these
examples for the interpretation of the subjunctive is so great that
full lists are given.

a~) Depending upon an impv. — Upon sine, Ba. 29 (24 GS.) sine
te amera; 1027 sine perlegam, 1176, 1199, Cas. 136, Cist. 454,
Ep. 204, M. G. 1084 (sinite), Most. 1180, Pers. 750, Poen. 142,
261, Poen. 375 (3 cases), Ps. 61, 239 (2 cases). Total, 18.

These are all short sentences, usually only sine and the verb.

With fac, Epid. 567 fac uideam ; Poen. 893 fac ergo id ' facile '
noscam ego ; the rest are all fac sciam, taking the place of the
impv. of a verb meaning 'to cause to know'; Cure. 414, 617,
Men. 890, M. G. 277, Ps. 696, Rud. 1023, Trin. 174. The verbs
are all of knowing. Total, 9.

With caue, St. 37 tace sis : caue sis audiam ego istuc posthac
ex te. True. 942 is a conjecture.

There is a small but remarkable group in which a paratactic
subj. goes with the impv. of a verb of action. Cure. 313 uin
aquam ? || si frustulentast, da, obsecro hercle, obsorbeam ; True.
367 deme soleas : cedo bibam ; Most. 373 uigila. || uigilo: cedo
bibam (MSS cedo ut bibam). Cf. Verg. Aen. IV 683 f. date
uolnera lymphis abluam, where date is not equivalent to sinite, as
Lade wig thinks. Most. 849 mane sis uideam, and perhaps such
cases as Cure. 427 concede inspiciam quid sit scriptum, though
this is usually printed with a colon after concede. The impv. in
these uses expresses the action which is necessary as an antecedent
to the subj. verb. With the impv., 32.

THE SUBJUNCTIVE IN PLAUTUS. 145

b) With an impersonal phrase. — optumumst, Asin. 448 nunc
adeam optumumst ; Cas. 949, 950, Epid. 59 sed taceam optu-
mumst ; Rud. 377 capillum promittam optumumst occipiamque
hariolari.

With necesse est, Poen. 1244 pro hoc mihi patronus sim
necessest. True. 817 is a conjecture.

With concessum, datum, Amph. 12 nam uos quidem id iam
scitis concessum et datum mi esse ab dis aliis, nuntiis praesim et
lucro.

With decretumst, Poen. 501 profestos festos habeam decretumst
mihi.

With certumst, Asin. 248, Aul. 681, Ba. 382, Capt. 779 (certa
res), Cas. 448. These are all verbs of 3d conjug., but they are
not quite futures.

Possibly Cist. 519 non remittam definitumst is similar.

The single case of licet with 1st sing., Asin. 718 licet laudem
Fortunam, tamen . . . shows by the use of tamen that licet is felt
as a conjunction. With impersonals, 14.

c~) With indicatives. — Rud. 681 quae uis uim mi adferam ipsa
adigit, Trin. 681 meam sororem tibi dem suades sine dote.
Amph. 9 is in a dependent clause in a long sentence, uti bonis
uos uostrosque omnis nuntiis me adficere uoltis, ea adferam, ea
uti nuntiem, and I have no doubt that, with colloquial freedom,
the infin., an ut clause, and the paratactic subj. adferam are used
as parallel constructions.

The effect of indirect discourse, which will be felt in these cases,
will appear also where other persons and numbers are used
paratactically with an indie, verb. With indicative, 3.

d) Paratactic questions. — In quis questions uis is inserted, Aul.
634 redde hue sis. || quid tibi uis reddam? (cf. 651 redde hue. ||
quid reddam?), Ba. 692 nunc hoc tibi curandumst, . . . || quid uis
curem?, Epid. 19, 584 quid taces? || quid loquar uis?, Merc. 158
quid uis faciam?, M. G. 300, Most. 578, St. 115.

uis stands between quid and the verb except in Epid. 584.
The function of uis is evidently to bring out more clearly the
inquiry as to the will of the person addressed which is contained
by implication in, e. g., quid reddam? With uis inserted, 8.

In sentence questions, which are almost invariably repudiating,
uin is inserted or prefixed in order to emphasize the true inter-
rogative character of the sentence.

Capt. 360 uin uocem hue ad te ? || uoca ; 858 uin te faciam
fortunatum ? ; Asin. 647, Cas. 272, 544, Men. 606, Merc. 486 (2),

146 AMERICAN JOURNAL OF PHILOLOGY.

721, M. G. 335, 1399 (but the text is not sure), Pers. 575, Poen.
439, 990, 1226, Ps. 324, 522, St. 397, 486, Trin. 1092, True. 502
uin adeam ad hominem ? || uolo, 924 (2). In Trin. 59 uin conmu-
temus? ego tuam ducam et tu meam ? the force of uin passes
over to ducam. In Merc. 728 etiam uis nomen dicam ? the
question is introduced by etiam, and -ne is not needed. In Capt.
121 the position of -ne is changed for emphasis to mene uis dem.
In Poen. 730 GS., following Pall., have censen hominem interro-
gem f A has quid turn.

Of these, Men. 606 has something of repudiating force (men
rogas? || uin hunc rogem?), but none is either an exclamation or
a deliberative question, uin is prefixed in order to exclude the
ordinary meaning of sentence questions with the subjunctive.

With uin, 27.

The 1st person sing, is not used with ne.

Present subjunctive, 2d person singular.

A. Non-interrogative , independent. — Wishes with utinam occur
Men. 1 104 utinam efficere quod pollicitu's possies, Cist. 555 utinam
audire non queas, both with verbs meaning 'to be able.' In the
marriage song, Cas. 821 uir te uestiat, tu uirum despolies, the
circumstances give something of optative force, which comes out
more clearly in the following verses. Trin. 351 will be given
under parataxis.

saluos (salud) sis is used 17 times as a form of greeting and
ualeas 7 times in parting. The meaning gives them optative force.

Postponing for a moment the hypothetical uses and the indefi-
nite 2d person, there remain 121 cases of the subjunctive express-
ing some kind of will or desire. In a broad sense of the word
these might be called jussive, but not more than a tenth of the
number are true commands and about as many more are
demands. Advice, serious or sarcastic or urgent, is the most
common kind of use, not far from 50 cases coming fairly under
this head. There are 15 or 20 requests and about as many
expressions of permission. Invitation, challenge, petition, expres-
sion of obligation, curse, are used each a few times, and there is
one asseveration, Most. 182. But where the form is unchanged it
is useless to make purely functional distinctions. I prefer to note
the usage of certain of the more common verbs, accipias (4
cases) is used only in requests and advice ; agas (2) and uel aias
uel neges (2) in a challenging demand ; dicas (6) is in all cases

THE SUBJUNCTIVE IN PLA UTUS. 1 47

but one advice, as part of a plan; habeas (11) is generally a
sarcastic permission ; ignoscas (3) is a petition ; iubeas (3), advice ;
laceas (6) a command in all but one case.

Hypothetical uses are rare, and in every case some preceding
or accompanying phrase gives the hypothetical tone. In Capt.
599 sapias mag-is is in answer to the question quid si . . . iusserim ?
In Rud. 1229 si sapias, sapias: habeas quod di dant, habeas is
only a continuation and expansion of the apodosis. In Aul. 231
the ubi clause contains a protasis and in Asin. 180, Trin. 554,
quouis and quamitis prepare for the hypothetical use.

una opera, in its peculiar Plautine sense, is used three times
with postules and twice with iubeas, all hypothetical.

There are 21 cases, also, in which the subject seems to be the
indefinite second person: Aul. 506, 517, 520, Capt. 420, Cas. 562,
M. G. 94, 689, 761, Most. 278, Poen. 585, 831, 836, 1416, Ps. 137,
1176 (?), Trin. 671 (2 cases), 914, 1031, 1052, 1054. Two or three
of these, which are not in soliloquy, might be questioned. If
they are not indefinite, they should be added to the list of hypo-
thetical passages. In many of the cases, some phrase or clause
precedes which sets the hypothetical tone. Thus in Aul. 506
quoquo uenias, Cas. 562 quom aspicias, Ps. 1176 ubi aspicias.
The verbs are uideas (5), censeas (3), audias (2), nescias (2),
scias, inuenias, conspicias, cupias, tie lis, desideres, all of mental
action, and noceas and perdas. The last two are preceded by
quom ferias and by duarum rerum exoritur : uel perdas . . . uel
. . . amiseris. 2d sing, independent, 179.

B. Questions. — There are only 5 quis questions, and the small
number and sporadic character make a precise interpretation
difficult. Rud. 1322 quid dare uelis, qui istaec tibi inuestiget
indicetque? and Asin. 558 edepol uirtutes qui (how) tuas nunc
possis conlaudare, sicut ego possim ? appear to be hypothetical,
but the use of the auxiliary verbs as expansions of quid des, qui
conlaudes, complicates the phrases. Pers. 638 quid (why) eum
quaeras qui fuit? is like Rud. 1322, i. e., might have been
expanded for greater clearness into quid quaerere uelis. Epid.
693 quid ago? || quid agas? mos geratur means 'what should you
(ought you to) do ? ' as the mode of mos geratur shows. M. G.
554 fateor. || quidni fateare, . . .? goes with other cases of quid
ni. Rud. 767 is not a quis question, but a relative qui with ne.

Sentence questions ail repudiate the expressed or implied
desire of the person addressed. There are 8 cases with -tie, all

I48 AMERICAN JOURNAL OF PHILOLOGY.

tun or tuin (gen.), one with an (or four, if Asin. 813 is read an
hi), and 10 without a particle. Aul. 431 is an indirect question.

C. Parataxis. — a) With imperative, fac (13), Amph. 976 hue
fac adsis; Capt. 439 fac fidelis sis fideli ; Cas. 421 et quamquam
hoc tibi aegrest, tamen fac accures. || licet; Cure. 521, Merc. 498,
M. G. 812 {face follows subj.), 1360, Pers. 196, 198, Poen. 1035
{face follows), Ps. 236, 481.

With facito, Asin. 238 syngraphum facito adferas ; Cas. 523,
Most. 216, Poen. 1084, 1278, 1418, 1414 leno, tu autem amicam
mihi des facito aut mihi reddas minam ; Trin. 485. The impv.
precedes the subj. in 17 of the 20 cases.

With sine, Asin. 902 sine reuenias modo domum : faxo scias . . .
GS. punctuate sine: reuenias, but cf. Cas. 437 sine modo rus
ueniat; Most. 11 sine modo adueniat senex, with exactly the
same threatening tone, sine has lost something of its verbal
force.

With uide, Asin. 755 adde et scribas uide plane et probe ; Poen.
578 uide sis calleas.

With caue, 10 cases. Capt. 431 caue tu mi iratus fuas; 439
caue fidem fluxam feras; Cas. 530, Epid. 437, Most. 810, 1025,
Pers. 51, 816 caue sis me attigas, ne tibi . . . malum magnum
dem ; Rud. 704. Aul. 660 is also a case of caue, though the rest
of the sentence is confused. With imperatives, 33.

b) With indicatives. — nolo, Capt. 383 ergo animum aduortas
uolo ; 388, 430, M. G. 546, Poen. 279, 1 197, Rud. 1414, Trin. 372.
uolo follows in 6 of the 8 cases.

nolo, Cas. 233 ted amo. || nolo ames ; Most. 1176 sine ted exo-
rarier. || nolo ores. || quaeso hercle. || nolo, inquam, ores. || nequi-
quam neuis ; Pers. 245, Trin. 945. nolo precedes the verb in all
cases.

malo, Ps. 209 taceo. [| at taceas malo multo quam tacere dicas.

faxo, Asin. 876 iam faxo ipsum hominem manufesto opprimas ;

Men. 113 faxo foris uidua uisas patrem ; 644 faxo scias; Most.

1133 ego ferare faxo; Ps. 949, Trin. 62 ne tu hercle faxo haud

nescias quam rem egeris ; 882 faxo scias. In Cure. 587 BJ have

faxo reperias, E repcries; the future is of course possible, but

the subjunctive is perfectly good. In Asin. 902 the MSS have

faxo ut scias, and as this construction is not infrequent, there is

no sufficient reason for omitting ut. In 7 of the 8 cases faxo

precedes.

THE SUBJUNCTIVE IN PLAUTUS. 149

c) With uelim and faxim. — Cas. 234 enicas. || uera dicas uelim,
Men. 909, Rud. 511, 1067. Compare especially Poen. 1150 abeo
igitur. || facias modo quam memores mauelim, with Ps. 209, quoted
above with malo.

With faxim, Amph. 511 ilia si sciat . . ., ego faxim ted Amphi-
truonem esse malis, quam Iouem.

d) With impersonals. — optumumst, Aul. 568 turn tu idem optu-
mumst loces ecferendum. With licet, Epid. 471 estne empta mihi
istis legibus? || habeas licet; Most. 713, Trin. 1179. Rud. 139 is
especially noteworthy because saluos sis is usually so distinctly
optative: meperiisse praedicas. || mea quidem hercle causa saluos
sis licet.

e) In questions. — Most. 322 uisne ego te ac tu me amplectare ?
Here the parataxis is really due to the omitted ampleciar. With
potin, Cas. 731 potin a med abeas?, Pers. 297.

C. Present, 2d person singular, with ne. — The distinction
between the independent sentence and the dependent clause is
nowhere more difficult than in sentences with ne. No thought is
really independent of the preceding thought, and the connection
may increase in closeness until it is one of real dependence with-
out finding expression in language. It is only when language
begins, so to speak, to run in ruts, to form fixed phrases giving
evidence of dependence, that we know that the line has been
crossed.

Such a phrase has been formed in Plautus in the clauses in
which ne is used with a verb of mistaking, of thinking wrongly.
ne frustra sis occurs 7 times, ne postules 6 times, ne censeas twice,
ne erres once, ne speres twice, and other forms (exisiumes, arbi-
trere, opinere, etc.) with verbs of thinking and saying (praedices,
dicas), and even occasionally with other verbs (metuas, territes (?),
quaeras), are found once or twice each. These may fairly be
excluded as semi-dependent, though they of course show some-
thing of prohibitive force.

molestus ne sis (10 times) shows in a few cases (Asin. 469
abscede hinc, molestus ne sis, Aul. 458), especially where it
follows immediately after another command, a tendency to
dependence.and the same beginnings of a feeling of purpose may
be suspected in other cases, e. g. M. G. 1361 i, sequere illos: ne
morere ; Pers. 318 emitte sodes, ne enices fame : sine ire pastum ;
M. G. 1 215 moderare animum, — ne sis cupidus. But whether

150 AMERICAN JOURNAL OF PHILOLOGY.

these are included or not, they would not change the result. In
the use of the pres. 2d sing, with ne there is the same range of
function as in independent uses without ne. The cases vary from
sharp and emotional warning {ne atligas me, ad portum ne bitas,
dico iam tibi, molestus ne sis) to prayers {Amor, amicus mihi ne
fuas~) and mild warning and advice. So far as I can see, the
proportion of emotional cases is not smaller than in the corres-
ponding uses without ne, where also the prevailing tone is one of
advice or suggestion. The number given in the table (61)
includes the partially dependent cases.

Present, 3d person singular.

A. Non-interrogative, independent. — Certain forms of wish are
so well marked in the 3d pers. that they can be set apart with
precision, differing in this from wishes in 1st or 2d pers.

With utinam there are 4 cases, Asin. 418, M. G. 1009 f., Most.
233, Rud. 158, all general in content, not like the specialized
forms of wish to be given below.

With ut, Cas. 238 ut te bonus Mercurius perdat, an unusual
kind of wish with ut. In Poen. 912 ualeas beneque ut tibi sit,
two forms of wish are put together. Pers. 290, Cure. 257 have ut
in an expression of desire, not a wish.

The phrase quae res bene {male) uorlal, quod bonum atque
fortunatum sit, etc., is found 10 times. It is introductory, as in
classical Latin, in only 3 cases.

Other impersonal forms of wish are bene {male) sit, uae tibi sit,
bona pax sit, male istis euenat, in all 7 cases.

The wishes which contain the name of a god are especially well
marked. Mars adiuuet, me facial quod uolt Iuppiter, Iuppiter te
seruet, Hercules te infelicet (after a repetition of licet), and
especially Iuppiter te perdat {perduit) ; 13 cases. There are also
6 cases of asseveration, ita . . . amet; Most. 182 has a lover's
name instead of a god's, and Capt. 877 f. combines ita amabit
with ita condecoret.

Beside these, Pers. 269 uapulet is a curse, Most. 374 pater
aduenit . . . || ualeat pater is the 3d pers. of ualeas, and Cas. 822
tua uox superet tuomque imperium : uir te uestiat is defined as a
wish by the fact that it is used in a marriage song. Wishes, 45.

The formal contract read by the parasite in Asin. 751 ff. con-
tains 9 cases of the 3d sing. These, like the other subjunctives

THE SUBJUNCTIVE IN PLAUTUS. 15 I

in that passage, express that kind of obligation which is involved
in a contract and deserve separate mention.

The remaining cases (except the hypothetical) require a some-
what careful analysis, because they imply in use much more than
the verb-form is capable of expressing, and the implied but
unexpressed elements lead to a considerable extension of the
meaning of the mode. Nothing in the verb-form defines in any
way the relation of the hearer to the will or to the action, though
that relation may be both real and close, and the variety of
possibilities in regard to the subject of the verb (a person, a
thing, impers., etc.) further complicates the matter.

When the subject of the verb is a definite person, the relation
of the hearer to the action suggests the following groups of
usage: — a) The hearer is to convey the speaker's will to the
third person, the actor. Am ph. 951 euocate hue Sosiam : guber-
natorem . . . Blepharonem arcessat, i. e. ' tell him that I want him
to call Blepharo ' ; Poen. 905 manu eas adserat, suas popularis,
liberali causa. Cf. M. G. 1037 adeat, siquid uolt. (i. e. 'tell her
to come here') || siquid uis, adi, mulier, where the wish (permis-
sion) is immediately conveyed to the actor.

b~) The hearer is to bring about the performance of the action
by the third person. Most. 920 octoginta debentur huic minae? ||
... || hodie accipiat 'see that he gets them'; M. G. 1304 omnia
composita sunt quae donaui : auferat ' have her carry them off' ;
M. G. 1 100, Cas. 697. With these should go the large number of
cases in which the speaker is advising the hearer as to the way in
which a third person is to act in order to carry out a plan. Pers.
151 sed longe ab Athenis esse se gnatam autumet; Trin. 764 ff.
scitum consilium inueni : homo conducatur . . . : is homo graphice
exornetur . . . : salutem ei nuntiet uerbis patris ; M. G. 792, Ps.
753 f., St. 299.

c) The speaker wishes the hearer to permit the third person to
act. The will may not extend to the third person, who may be
ready of himself to do the act. Merc. 989 redde filio : sibi
habeat. || iam, ut uolt, per me habeat licet; Merc. 991, Pers. 447,
Rud. 1 121 aliud quidquid ibist, habeat sibi.

cT) The speaker expresses his indifference in regard to an act
of the third person and implies that the hearer also is to be
indifferent. Poen. 264 erus nos . . . mantat. || maneat pol : mane,
i. e. 'never mind if he is waiting'; Ba. 224 adueniet miles. ||
ueniat quando uolt.

152 AMERICAN JOURNAL OF PHILOLOGY.

e) In some cases — the large majority, probably, in formal style
— the second person is merely a hearer, the recipient of the
speaker's confidence. Ba. 502 ilium exoptauit potius ? habeat ;
Amph. 300 clare fabulabor : hie auscultet quae loquar, where LG.,
ed. crit., suggest ut hie, unnecessarily, I think. In such sentences
the expression of speaker's desire is the important thing, and its
effect upon the action of the third person is of little consequence,
so that these sentences may approach a wish; St. 711 modo
nostra hue arnica accedat.

As only definite persons are involved, the nature of the will —
command, advice, permission — is much the same as in 2d sing.

With definite actor, 48.

Passages in which the actor is not a definite person fall also into
several groups, but the difference in meaning is slight. When
the subject is described in a relative clause, the meaning is almost
the same as when the subject is definite. M. G. 81 qui autem
auscultare nolet, exsurgat foras ; Rud. 486 ; with quisque, Pers.
373 dicat quod quisque uolt : ego non demouebor. In a number
of cases the subject is an ideal or typical person, a true lover (Ps.
307 det, det usque : quando nil sit, simul amare desinat), a genuine
woman, by the standards of comedy (M. G. 190 qui arguat se,
eura contra uincat iure iurando suo), an ideal slave (Amph. 960
proinde eri ut sint, ipse item sit : uoltum e uoltu conparet ; Ba.
656, Aul. 599 f.). In Pers. 125 cynicum esse egentem oportet
parasitum probe: . . . pallium, marsuppium habeat, the decline of
the subjunctive in these cases to a mere sense of artistic propriety
is illustrated by the parallel of esse oportet. In nearly all of these
cases the second person, if one is present, is disregarded and the
speaker addresses the audience ; i. e. both the actor and the
hearer are indefinite. In all, 25 cases.

If the subject is a thing (which occurs rarely) or if the verb is
passive, the subject is not the actor.

When the subject is a definite person, the hearer is usually to
be the actor, and the expression of will may be essentially the
same as in the cases above under b or even in 2d sing. So
Fragm. 50 (Carbon. II) patibulum ferat per urbeni, deinde adfi-
gatur cruci; Trin. 767 is homo exornetur; M. G. 1401 iamne ego
in hominem inuolo ? || immo etiam prius uerberetur fustibus ; Capt.
609, M. G. 1418.

The cases in which the subject is a thing are nearly all in the
plural, but beside a few verbs of passive meaning, maneat, stet,

THE SUBJUNCTIVE IN PLAUTUS. 153

supersit, defiai, the phrase cena detur occurs twice, mos geratur
4 times zndjiat 27 times. Of the last, three cases have a subject,
but the rest are all in connection with an expression of desire, to
which fiat gives assent. In these cases the previous speaker, the
second person, has already expressed the desire, and the speaker
of fiat, who is really to be the actor, selects a form which leaves
the person who wills, the hearer, the actor and the nature of the
act indefinite, so that, in truth, nothing is expressed except assent
to the desire, as if he said 'your will is mine.' This leaves fiat so
weak that it is little more than a future, and in fact fiet is also used
(Men. 186, Merc. 302, M. G. 908) in the same sense, as mos tibi
geretur is used, Ps. 22, for the subjunctive. With passives, 45.
The 3d person singular is also used, though not frequently, of
a supposed case. In six or eight passages the hypothetical tone
is set by a protasis and is continued through the following
sentences. These passages are not counted. There are also 12
cases where there is no distinct protasis in the immediate context.

Ba. 139 non par uidetur neque sit consentaneum praesens

paedagogus una ut adsiet, is an excellent illustration of the
meaning of this use of the mode, par uidetur is equal to sit
consentaneum, the meaning of uideri exactly expressing the
opinion or view which in the second phrase is expressed by the
mode. Ba. 97 ego opsonabo : nam id fiagitium meum sit, mea te
gratia . . . facere sumptum; True. 221 stultus sit, qui id miretur;
M. G. 736 qui deorum consilia culpet, stultus inscitusque sit ; St.
24 ioculo istaec dicit: neque ille sibi mereat Persarum montes,
. . ., ut istuc faciat (cf. non meream) ; M. G. 691 hoc numquam
. . . audias: uerum priusquam galli cantent, . . ., dicat 'da, mi
uir'; Capt. 208 nos fugiamus? quo fugiamus ? || in patriam. ||
apage, haud nos id deceat fugitiuos imitari ; True. 907 numquam
hoc unum hodie ecficiatur opus, quin opus semper siet ; Ps. 432
fors fuat an istaec dicta sint mendacia; Amph. 1060 nee me
miserior feminast neque ulla uideatur magis. In Trin. 441 hie
postulet frugi esse : nugas postulet, the first verb is in sense a
protasis and the subjunctive is not hypothetical ; the second verb
might perhaps be omitted from this list, as being influenced by
the protasis. Asin. 465 Sauream non noui. || at nosce sane. || sit,
non sit : non edepol scio, is very peculiar and perhaps unparal-
leled in Plautus, but the meaning is clear; cf. Capt. 964, St. 31 ff.
In Most. 984 possiet is a conj. of Cam.

154 AMERICAN JOURNAL OF PHILOLOGY.

B. Questions. — The quis questions are introduced by quis (3),
quid (3), qui 'how' (2), unde (1), and are all of one pretty well
defined class, implying impossibility. The only cases which call
for notice are quid hoc sit hominisf, Amph. 576, 769, and quid
hoc sit negoti . . .?, Asin. 407 (Cam. reads est, but sit is defended
by the other cases). In these the relation to the subjunctive of
desire is somewhat more apparent than in some of the other
cases, though none is strictly potential.

The sentence questions have ne in 3 cases, and are without a
particle in 8. All are repudiating exclamations, but all show the
same leaning toward the potential which appears in the quis
questions, and non is found three times. Rud. 728, where dei,
not det, is the correct reading, and Men. 763 are not included.

Questions, 20.

C. Parataxis. — As the vagueness of the 3d pers. sing, of the
subjunctive is chiefly in the undefined relation of the hearer to
the will and the action, it will be found that the leading verb
serves mainly to define what the mode alone leaves undefined.
The classes below are arranged in the same order as those above,
under A.

a) With iube, indicating that the hearer is to convey the
speaker's will to the actor. Most. 930 die me aduenisse filio. ||
... || curriculo iube in urbem ueniat; Pers. 605 iube dum ea hue
accedat ad me ; Rud. 708 iube modo accedat prope ; with a more
polite addition, Most. 680 euoca dum aliquem ocius, roga circum-
ducat. Cf. Amph. 951 euocate Sosiam : . . . arcessat . . .

b) With fac,facito,facite. The hearer is to cause the subject
of the verb to act. Rud. 1219 et tua filia facito oret: facile
exorabit; Pers. 445 facito mulier ad me transeat; Most. 854 age
canem istam a foribus aliquis abducat face. Some cases with fac
have a verb of passive sense, often with a thing for the subject,
and correspond to the uses with passive verbs in which the hearer
is the real actor. So Ps. 157 aquam ingere: face plenum ahenum
sit coco ; Men. 866 facitote sonitus ungularum appareat ; Pers.
438, Men. 867, 992, Rud. 621, 1215.

c) With sine ; the second person is to permit the third person
to act. M. G. 1244 sine mulier ueniat, quaeritet, desideret,
exspectet; Cas. 206 sine amet, sine quod lubet faciat ; Cist. 734,
Ep. 36, Ps. 478. With sine modo, Amph. 806, Cas. 437, Most. n.
In Ps. 159 at haec (securis) retunsast. || sine siet, and Asin. 460 ne

THE SUBJUNCTIVE IN PLAUTUS. 155

duit, si non uolt. sic sine adstet, the speaker is indifferent, as in
the following class. Cf. Poen. 264 erus nos mantat. || maneat pol.

d) With licet; the speaker is willing or indifferent. Merc. 989
redde filio : sibi habeat. || iam, ut uolt, per me habeat licet ; Capt.
303 the speaker is helpless.

e) With uolo, emphasizing the speaker's will and leaving the
second person out of the action. Ps. 11 23 leno argentum hoc
uolo a me accipiat atque amittat mulierem, Asin. 77, Rud. 1332,
True. 473. But in Poen. 1151 patruo aduenienti cena curetur
uolo, as the verb is passive with a thing for subject, the hearer is
to be the actor and the will is a command. Pers. 832 at enim
quod ille meruit, tibi id obsit uolo, continues the curse expressed
in 831, and therefore approaches the meaning of the subj. with
uelim.

The cases with malo (M. G. 1333, with somewhat uncertain
text) and nolo (Merc. 107, Ps. 436, St. 734) call for no comment.

uelim with the subjunctive has been given above. The passages
are Aul. 670, Cas. 559, Most. 1074, Poen. 1288, Ps. 1061, Rud.
877, True. 481. With malim. following an asseveration, Poen. 289.

With faxo, emphasizing the speaker's determination to bring
about the act, Amph. 972, Ba. 864, Most. 68, True. 643. With
faxim in an ui clause, True. 348. With faciam, Amph. 63, 876.

Scattering cases are St. 757 si quidem mihi saltandumst, turn
uos date bibat tibicini (cf. da bibam) ; Merc. 1004 nihil opust
resciscat; Cure. 461 leno, caue mora in te sit mihi. Capt. 961
quod ego fatear, credin pudeat quom autumes? is the only para-
tactic question. It gives a good basis for interpreting M. G. 614
quodne uobis placeat, displiceat mihi? and shows that these
questions deal with an opinion. In parataxis, 60.

D. With ne. — The contract in Asin. 751 ff. contains 17 cases
with ne, and a proclamation in Poen. prol. 17 ff. has 3 more.
The rest are nearly all ?ie quis, quisquam, and call for no remark.

With ne, 28.

Present, 1st person plural.

A. — The hortatory use is so well marked and so well known
that nothing need be said of it here. There are 94 cases. The
verb eamus, with its compounds, is used 42 times (evidently
because of a recurring dramatic situation), agamus is used 4
times and other verbs once or twice each, utinam is used only

156 AMERICAN JOURNAL OF PHILOLOGY.

once, Asin. 615. The use of age, agite, and of a vocative {eamus,
mea germand), and especially of tu (True. 840 eamus tu in ius),
show that the sense of the 2d person was felt.

B. — Questions occur only 3 times : with quo, Capt. 208, with
uter, deliberative, St. 696, and in a repudiating exclamation, Capt.
208.

C. Parataxis. — With nolo, Ba. 708, St. 670 uolo eluamus hodie
peregrina omnia. With cetiseo, Merc. 1015 immo dicamus senibus
legem censeo. With suades, Asin. 644 proinde istud facias ipse,
quod faciamus nobis suades ; cf. Trin. 681 dem suades. With
orant, Amph. 257 uelatis manibus orant ignoscamus peccatum
suom. There are 2 questions with uin, St. 736, Trin. 59.

The proportion of paratactic to independent uses,- 7 out of 105,
is much smaller than in other persons.

D. — There is one case with ne, Poen. 251.

Present, 2d person plural.

A.-r-Poen. 623 fortunati omnes sitis is a wish; M. G. 1341 is a
petition or request ; Cure. 632 quid istuc ad uos attinet ? quaeratis
chlamydem et machaeram hanc unde ad me peruenerit, follows
two requests that he should tell where he got a certain ring, and
is ironical, 'ask me where I got my cloak.' But the addition of
una opera (as with postules) would bring out clearly the under-
lying sense, 'you might as well ask me.'

B. — There are no questions.

C. Parataxis. — The cases are so exactly like those in the sing,
that I give only the numbers, uolo 3, facite 3, oro obtestor 2,
modo faciatis oro 1, caue 1.

D. — With ne, 13 cases, of which 8 are addressed to the audi-
ence ; ne expecietis 5, (ad~)miremini 2, uereamini 1.

Present, 3d person plural.

The differences between 3d sing, and 3d plur. are : 1) the
absence of cases in which the subject of the verb is a definite
person ; 2) the large number of wishes and curses ; 3) the
peculiar paratactic uses.

THE SUBJUNCTIVE IN PLAUTUS. I 57

A. Non-interrogative, independent. — Wishes. With utina??i,
Asin. 841, Pers. 289, Ps. 108, all general in character. With ut
and di perdant {perduint), Aul. 785, Merc. 710, Pers. 298, Fragm.
(Boeot.) 21. Rud. prol. 82 ualete, ut hostes uostri diffidant sibi
(cf. Poen. 912 ualeas beneque ut sit tibi) ; Cist. 202 (prol.) ualete
et uincite . . . seruate . . . socios, . . . parite laudem et lauream :
ut uobis uicti Poeni poenas sufferant; perhaps also Poen. prol.
128 ualete atque adiuuate: ut uos seruet Salus, though it is
usually punctuated adiuuate ut. These seem to be all cases oi ut
in a wish, but I fear that my list is incomplete.

Wishes and curses with di {deaeque) or the names of gods.

There are 25 cases of di te {ilium, istam) perdant and 5 of di
me perdant; 11 of di te (istum, illos) perduint and 1 with vie.
Other forms are di te infelicent (5), malum quod isli di deaeque
duint(i), di deaeque . . . te . . . excrucient. Of good wishes there
are di te ament (14), di te seruent ( 1), sospitent{i), di bene uortant
(7), di tibi dent quaequomque optes {quae uelis) (10), di duint
quaequomque optes (2), omnia optata offerant (1), di bene {male,
melius) faciant (8). Asseverations with ita are ita me di ament
(20), ita me di seruent (1), and wishes ita di faciant occur 4 times.

Two other cases, Cure. 575 ita me machaera et clypeus . . .
bene iuuent and M. G. 1316 saluae sient, show by their content
and relation to other phrases that they are wishes.

The whole number of wishes in 3d plur. is 127, of which 118
contain the names of gods or the word di. With these utinam is
used 3 times (in wishes of a general character), ut 4 times, qui 7,
quin 2, at 6, o once, ah once, ita 27 times.

Beside the wishes the subjunctive is used 21 times in other
expressions of will or desire. Most of these are active, and the
subject is never inanimate and never definite individuals, but
always a class, reges, haruspices, matronae, inimici, or, more
vaguely, alii, omnes, or a class described in a qui clause. The
person addressed is usually the audience or a person present on
the stage and treated as a representative of the public ; in a few-
cases it might be said that the third persons are really the persons
addressed, indirectly and impersonally, as in laws (e. g. Poen.
prol. 32 ff. matronae tacitae spectent etc.). As both actors and
persons addressed are thus vaguely conceived, there is no possi-
bility of cases of direct command, such as appear in 3d sing.
The 3d plur. expresses only the more general kinds of desire,
amounting usually to no more than a statement of obligation or

158 AMERICAN JOURNAL OF PHILOLOGY.

propriety. In two cases, Asin. 671 . . ., ni genua fricantur. || quid
uis egestas imperat: fricentur, Ba. 1133 cogantur quidem intro,
where the verb is passive, the person addressed is really the actor,
and these express more direct forms of will, almost equivalent to
f rices, cogamus. So also in Capt. 115, where uti adseruentur
expresses a command and is not a dependent clause.

The hypothetical uses are found chiefly in four long passages,
Amph. 155 ff., Aul. 228 ff., Merc. 407 ff. (10 verbs), Rud. 978 ft'.,
where a protasis has preceded or has been plainly implied. In
M. G. 1369 f. the protasis is implied in caue isiuc feceris ; in Asin.
602 qui sese parere adparent huius legibus suggests a protasis ;
in Trin. 703, 740, 743 the implication is plain before the subjunctive
verb is reached.

Counting all the verbs there are 23 cases, occurring in 8 passages.

B. — The 3d plural is used only twice in questions. Poen. 860
is a repetition of a preceding di ament ; Ps. 205 is a repudiating
exclamation.

C. Parataxis. — The cases are arranged as in 3d sing.

a) iube, Men. 956 tu seruos iube hunc ad me ferant ; St. 396 i
intro . . . : iube famulos rem diuinam mi apparent (cf. for ace.
some cases of ueniat ueliiri).

b) With fac, f actio, faciie, Aul. 402 (BDJ, but Non. has a
different tradition, which GS. follow) ; Aul. 407 facite totae plateae
pateant ; Cas. 521 fac uacent aedes ; Cas. 527 fac habeant linguam
tuae aedes. || quid ita?||quom ueniam, uocent ; Ps. 166, 181,
Fragm. 70 (Cornic. VII). With ttide, Amph. 629 uide ex naui
efferantur.

c) With sine, Ba. 11 34 sic sine adstent.

d) With licet there are no cases.

e) With nolo, Pers. 293 eueniant uolo tibi quae optas ; nolo,
True. 585 uasa nolo auferant ; with faxo, Amph. 589, Men. 540 ;
with faciam, M. G. 1399 uin faciam quasi puero in collo pendeant
crepundia? (notice the double parataxis); viithfaxim, Aul. 495,
Merc. 829, Pers. 73, Trin. 221, 222.

Thus far, with three or four exceptions, the verbs are passive in
meaning or in form and the subjects are things. In the indepen-
dent uses, on the contrary, the verbs are nearly all active and the
subjects are classes of persons. The paratactic uses in 3d plur.,
therefore, are not parallel to the independent uses, extending

THE SUBJUNCTIVE IN PLAUTUS. 1 59

them along the same lines, but supplementary, expressing ideas
which the independent uses express rarely or not at all. Plautus
did not say fac sese domi contineant,fac matronae tacitae specie?ii,
because the subjects of the verbs were both the persons addressed,
though indirectly, and the actors ; nor did he say uacent aedes or
ex naui efferantur quae imperaui omnia, without fac or uide,
because such phrases would omit the actor entirely.

A few other scattering cases are found. Ps. 938 si exoptem,
quantum dignus, tantum dent ; Capt. 694 nil interdo dicant (or
interdico aiant, Fl. GS.) ; Ps. 207 prohibet faciant (a gloss) ;
Poen. prol. 22 decet . . . stent . . . temperent; Ba. 1033 caue . . .
fuant. Amph. 632 utinam di faxint, infecta dicta re eueniant tua
is a paratactic wish. St. 31 ff. ipsi interea uiuant, ualeant, ubi
sint, quid agant, ecquid agant, neque participant nos neque
redeunt presents two curious indirect questions, uiuant, ualeant,
depending paratactically upon participant without an interrogative
particle.

D. — With ne there are only three cases, Poen. prol. 23, 29, 38,
entirely like the other independent uses in this passage.

Imperfect, 1st person singular.

A. — In independent non-interrogative uses only wishes with
utinam and expressions of desire with uellem, mauellem occur.

With utinam, Amph. 575 utinam ita (i. e. ebrius) essem ; Rud.
533 utinam fortuna nunc hie anetina uterer. Both express a
present wish contrary to the fact, impossible of fulfilment.

uellem is used 9 times: — with infin., Asin. 589, Poen. 681, Cist.
93 ; with perf. ptc, Cist. 506 ; with adj., Most. 980, Ps. 309 ; with
obj., St. 713; with parat. subj., Poen. 1066 patrem atque matrem
uiuerent uellem tibi. || an mortui sunt ? || factum ; St. 312 nimis
uellem haec fores erum fugissent. Of these cases, Poen. 681
uidere equidem uos uellem, quom huic aurum darem expresses a
desire in the past, but still felt by the speaker, in regard to a
future act (cf. 682 illinc procul nos istuc inspectabimus). The
rest are all like uelim in that they express a wish, but the impli-
cation that it will not be fulfilled is clear in some cases (Ps. 309,
Poen. 1066, Cist. 506) and possible in all.

mauellem (mallem) is used 8 times: — with infin., Amph. 512,
Ba. 198, 452, Ps. 1057 ; with perf. infin. act., Cure. 512 ; with infin.
implied, Ba. 1201, Ps. 131; with parat. subj., Ba. 1047 ne ille

l6o AMERICAN JOURNAL OF PHILOLOGY.

edepol Ephesi multo mauellem foret, . . ., quam reuenisset domum.
The sense of desire or will is present in all ; some express a curse
(Ba. 198, Ps. 1057, 131); in a few cases (Cure. 512, Ps. 131, Ba.
1047) the suggestion of non-fulfilment is quite distinct.

B. Questions. — There are two quis questions, both referring to
a past obligation or desire on the part of the second person.
Cist. 94 I should take as a dependent clause.

Sentence questions are mainly rejections of an expression or
implication of obligation in the past. In Trin. 177 paterer is
balanced by indicare me aequom fuit in the preceding verse. In
Most. 183 ita ego istam amarem? is an exclamatory repetition of
ita Philolaches tuos te amet in 182. The negative in Most. 455
is non.

C. — In parataxis there is only one case, St. 177 hoc nomen
repperi eo quia paupertas fecit ridiculus forem.

Imperfect, 2d singular.

A. — With utinam there is one case, Rud. 494 f. utinam ... in
Sicilia perbiteres, with distinct past reference.

In one case the subjunctive is plainly hypothetical; Men. 160
edepol ne tu, ut ego opinor, esses agitator probus. The reference
to the past is certainly not clear, but cf. Merc. 125 nimis nili tibicen
siem, where the future reference is apparent.

In all other independent 2d sing, cases (13) the subjunctive
expresses an obligation which the actor should have felt in the
past. Merc. 633 ff. quid ego facerem ? || . . . men rogas ? requae-
reres, rogitares; Merc. 637, Poen. 387, 391, Rud. 842 quin occi-
disti extemplo? || gladius non erat. || caperes aut fustem aut lapi-
dem ; Trin. 133 ff. (4 verbs); Pers. 710 animus iam in nauist
mihi. || eras ires potius, hodie hie cenares deserves special mention,
because it expresses a past obligation in regard to a future action ;
' You should have made up your mind (then) to go to-morrow
(not to-day).' Ba. 432 . . . ubi reuenisses domum, ... in sella
apud magistrum adsideres is in a passage describing old customs ;
the other verbs are in the impf. indie, and the subj. adsideres is
due to the ubi clause. Ps. 494 should be printed with a period,
as in GS., otherwise there is no justification for the past tense.

B. Questions. — With quis, Merc. 884 quo nunc ibas ? || exula-
tum. || quid ibi faceres ? Goetz, ed. crit., supplies ut, but the text

THE SUBJUNCTIVE IN PLAUTUS. l6l

as it stands (GS.) gives the proper sense, 'What were you
intending to do there?'

There are no sentence questions. Capt. 713 is a continuation
of a conditional sentence. For Ps. 494 see above.

C. Parataxis. — Asin. 503 si esses percunctatus . . ., scio pol
crederes is not counted, and Ba. 635, if non is dropped from the
text, is exactly similar. The only clear case is St. 624 ueni
(impv.). || hucine? || immo in carcerem. || quid igitur? || dixi equi-
dem in carcerem ires. This is really a quotation, dixi ' in carcerem
i,' remarkable for the use of dixi instead of iussi.

D. — With ne there are 3 cases, expressing past obligation : Ps.
437, Ba. 29, 30 (16 f. in ed. crit.).

Imperfect, 3d singular.

A. — With utinam, Merc. 823 utinam lex esset eadem, quae
uxorist, uiro. There are 2 cases in Rud. 379 f. of past obligation.

There are 3 passages where the mode appears to be hypo-
thetical, Cas. 910, Rud. 1262 (2), Ba. 314. In the last, nimio hie
priuatim seruaretur rectius, a slight sense of obligation is produced
by rectius.

B. — The one quis question, Rud. 379, is a question in regard
to a past obligation.

The sentence questions are all associated with a protasis in the
context, though Ps. 288 is a repudiating exclamation. The others
are Trin. 178, Capt. 714, Trin. 954.

C. — In parataxis, mauellem is used with foret, Ba. 1047, in a
wish. Trin. 115 si inimicus esset, credo haud crederet is not
counted. The rest are all of the nature of indirect quotations.
Ba. 551 ille, quod in se fuit, accuratum habuit, quod posset mali
faceret in me, inconciliaret copias omnis meas. The MSS have
ineonciliare, which in such colloquial style is not impossible;
faceret expresses a past intention, quoted by accuratum habuit.
Pers. 634 tactus lenost, qui rogarat [rogabat?], ubi nata esset,
diceret; that is, the leno asked {die) ubi tu nata' s 9 Trin. 591
tandem impetraui abiret is as if for uolui (iussi) abiret et tandem
impetraui ; Merc. 536 f. inter nos coniurauimus, . . ., neuter stupri
causa caput limaret; in Epid. 316 me iussit senex conducere
aliquam fidicinam . . . : dum rem diuinam faceret, cantaret sibi,

l62 AMERICAN JOURNAL OF PHILOLOGY.

there are various conjectures, but I take cantaret to be a part of
the order, fidicinam conducas: ea mihi cantet, as often in pres.
tense.

D. — The only case of ne has a protasis.

There are no cases of ist or 2d plur.

Imperfect, 3^ plural.

A. — With utinam, Capt. 537 utinam te di prius perderent,
quam periisti e patria tua, with distinct reference to the past.
Trin. 1028 f. may perhaps also refer to the past, but is better
taken as an ordinary unfulfilled wish.

The other cases all occur in passages where the speaker is
giving the details of a plan made in the past. In M. G. 731 f.
itidem diuos dispertisse uitam humanam aequom fuit: . . ., uitam
ei longinquam darent : . . ., is adimerent animam cito, and in
Epid. 386 aequom fuit clearly defines the meaning of the mode.
Poen. 1 139 f. hodie earum mutarentur nomina, facerentque . . .
quaestum corpore is part of a plan made by the leno, not by the
speaker, and would be introduced paratactically by lenoni decre-
lumst or some similar phrase.

There are no hypothetical or interrogative uses.

C. — In parataxis, Poen. 1066 depends upon uellem ; Merc. prol.
52 (48 G.) pater clamitare . . . et praedicere (histor. infin.), omnes
timerent mutuitanti credere; M. G. 54 at peditastelli quia erant,
siui uiuerent.

The temporal force of the imperfect is plain in those uses which
most nearly resemble the direct expressions of will in the present ;
all expressions of obligation refer to the past, even when the act
to be performed is still in the future. But in wishes the shift of
temporal force, by which unfulfilled conditions and wishes in the
present take an imperfect subjunctive, had already begun. Cases
have been noted above where the reference to the past seemed
most distinct. See also Blase, Geschichte des Irrealis, pp. 3-5.

Perfect, ist person singular.

A. — Asia. 491 praefiscini hoc nunc dixerim. Cramer, 1 pp. 47
ff., doubts the genuineness of the passage, but calls it a subjunc-

X Y. Cramer, de perfecti coniunctiui usu potential! ap. prise, script. Lat.
Marburg, 1886.

THE SUBJUNCTIVE IN PLAUTUS. 1 63

tive of will, correctly, translating "Dies wiinsche ich jetzt unbe-
rufen zu sagen." Cf. the like uses of dicam, haud dolo dicam, etc.
True. 349 follows a si clause in 344 ff. and is hypothetical.

B. — In quis questions, Amph. 748 ubi ego audiuerim ? repeats
and rejects audiuistin. There are 3 cases of nouerim, Cure. 423,
Men. 299, M. G. 923, which are classed here, though noui is
present in sense.

In True. 625 there is a repudiating repetition of a subjunctive.

There are beside these few cases many perfects 1st sing, which
are called potential (v. Brix, n. on Capt. 309), but they are all in
subordinate clauses or in conditional sentences.

Perfect, 2d singular.

The functions of perf. subj. and fut. perf. indie, differ so slightly
in independent uses, in any person except the 1st sing., that one
set of forms suffices for both. The confusion which might be
expected from this is, however, much less than that between fut.
indie, and pres. subjunctive.

A. — In independent expressions of will memineris is used twice
(M. G. 807, Pers. 856) and noueris once (True. 163 dum uiuit,
hominem noueris: ubi mortuost, quiescat), both really present in
sense. Ba. 840 meretricemne esse censes? || quippini ? || frustra's. ||
quis igitur opsecrost? || inueneris, i.e. 'find out for yourself.'
Trin. 1053 duarum rerum exoritur optio: uel illud quod credi-
deris perdas uel ilium amicum amiseris is exactly like a number
of cases mentioned in the pres. tense.

Other cases have more of future meaning. They are Capt.
1028 (see Brix, note), Cure. 665, Trin. 760, Most. 1152 (see
Lorenz 2 , note), Trin. 61. In Capt. 1028 and Cure. 665 there is a
hypothetical tone, though it is not distinct enough to find expres-
sion in a clause; Trin. 61 is influenced by the preceding faxo
dederis. Most. 1152 and Trin. 760 are the most distinct futures.

B. — There is only one question, a repudiating exclamation,
Amph. 818 tun mecum fueris ?

C. — In parataxis, faxo dederis (Trin. 60) and faxo haud come-
deris (Men. 521) lie between the subjunctive and the clearly
future uses. So also ?nemineris facito St. 47, though it is more
closely related to the independent uses of -memineris.

The remaining cases are all prohibitions, either with caue or
with ne.

1 64 AMERICAN JOURNAL OF PHILOLOGY.

The verbs used with caue are feceris (Cas. 332, M. G. 1368,
Poen. 1023, St. 285, Trin. 513), dixeris (Pers. 389, Trin. 555),
siueris, siris (Ba. 402, Epid. 400, Most. 401), praeuorteris (Merc.
113), rettuleris (Epid. 439), sumpseris (Cist. 300), fueris (Aul.
618), intromiseris (Aul. 90), responderis (Amph. 608).

D. — With ne and its compounds the verbs are fueris (Asin.
839. Epid. 595), dixeris (Cist, no, Merc. 402), feceris (Epid. 148,
Men. 415), parseris (Pers. 572, Poen. 993), attigeris (Pers. 793),
induxeris (Trin. 704), destiteris (Trin. 1012), ostenderis (Rud.
1 1 55), interueneris (? M. G. 1333).

Other forms of prohibition are nil monueris Cure. 384, minume
feceris Most. 272. In Rud. 1135, Pers. 395, the futures in the
context seem to show that nullum ostenderis, milium acceperis
are also future.

Perfect, 3d singular.

A. — There are two cases with uiinam, Cas. 398 f., Poen. 799,
both referring to the past, and one, Trin. 753 nam certo scio,
locum quoque ilium omnem, ubi situst, comederit, which is
hypothetical.

B. — There are two quis questions, M. G. 925, Trin. 1050.

C. — In parataxis: with uelim, Ba. 334, Poen. 1206; with faxo,
Aul. 578, Poen. 346 (either might be called fut. perf.) and Capt.
801 (so far as the corrupt, text makes a judgment possible) ; with
caue, Men. 994 caue quisquam uostrum . . . fecerit, really a second
person. The text of Cure. 27, M. G. 926, True. 429 is too uncer-
tain for use.

There are no cases of the 1st person plural.

Perfect, id plural.

The only case is M. G. 862 ne . . . dixeritis, addressed to the
audience.

Perfect, 3d plural.

Aul. 542 meminerint is an expression of propriety, like those in
pres. 3d plur.; St. 385 perierint is a curse, and both verbs are in
reality present in meaning. Poen. 617 is a future perfect.

In parataxis there are two cases with ttelim, Poen. 570, Rud.
662, both curses.

ne di sirinl (siuerinf) is used in Ba. 468, Merc. 613, Merc. 323.

THE SUBJUNCTIVE IN PLAUTUS. 165

Of the forms with perfect ptc, Amph. 979 fac conmentus sis is
deponent and perfect. The rest are all presents: paratus sis
Trin. 1189, occlusae sint Asin. 759, curata fac sint Aul. 273,
Amph. 981, facito opsonatum sit Ba. g6,facite deductus siet Capt.
•jT,6,/ace occlusae sient Most. \oo,fac sit delaium Ps. 190. They
do not differ in use from active forms.

Pluperfect.

In 1st sing., Rud. 497 f. utinam cubuissem. In 2d sing., utinam
parsisses True. 375 ; fuisses (conj.), hypothetical, M. G. 1 1 1 2.
In 3d sing., Cas. 996, Poen. 1252 are useless. M. G. 721 si ei
forte fuisset febris, censerem emori: cecidissetue ebrius aut de
equo uspiam, metuerem ne ibi diffregisset crura, is a protasis
without si, under the influence of the preceding si (cf. Pseud. 863).
Epid. 628 is hypothetical. In 3d plur. Amph. 386 is a wish with
utinam and St. 312 with uellem.

Subjunctive forms in -s-.
1st person singular.

haud (non) ausim, Aul. 474, Ba. 1056, Poen. 1358; all of
hypothetical statement.

haud negassim, Asin. 503.

non empsim, Cas. 347, M. G. 316; also hypothetical.

ausim in questions, Merc. 154 egon ausim tibi usquam quic-
quam facinus falsum proloqui; Most. 923 f. egone te ioculo modo
ausim dicto aut facto fallere? || egone aps te ausim non cauere,
. . . ? ; Poen. 149. These are, as the form of question shows,
repudiating exclamations, and ausim is an insertion for fuller
expression; in such a phrase as egon tibi usquam quicquam
facinus falsum proloquar? the idea of wishing or desiring (the
proper sense of audeo) is latent. Cf. the similar insertion of
patiar and of uis, uin. But Merc. 301 sed ausimne ego tibi
eloqui fideliter ? is a true question, answered by audacter.

faxim is used in male faxim lubens Poen. 1091, 1093, and is
hypothetical, though no protasis is found in the context. In all
other cases of faxim a protasis is expressed in the context (Trin.
221, cf. 217-20) or in the sentence itself (Amph. 511, Aul. 494,
Merc. 814 G. (826 GS.), Pers. 73) ; these cases are included in the
list for completeness, though strictly they would be excluded by
the protasis. In True. 63' the text is somewhat uncertain, though
faxim is sure. True. 892 hostissim is a conjecture and of no value.

166 AMERICAN JOURNAL OF PHILOLOGY.

All cases of forms in -sim are, it will be noticed, hypothetical.

id person singular.

All forms are in prohibitions, either with ne, neque, numquam,
nil, or in parataxis with caue. They are as follows : Asin. 839 ne
dixis, Aul. 744 ne dixis, Capt. 149 numquam dixis neque animum
induxis, Men. 611 ne comessis (Bx.; comesses MSS GS.), M. G.
283 ne dixis, 1007 nil amassis, Most. 526 nil curassis, 1097 ne
occupassis, n 15 ne /axis, Poen. 553 ?ie curassis, Ps. 79 ne par sis,
232 nil curassis, Rud. 1028 neque indicassis, St. 149 neque
celassis, Trin. 627 neque occultassis, True. 606 ne responsis.

With caue, Asin. 256 /axis, 467 supplicassis, 625 /axis, Aul.
608 indicassis, Ba. 910 parsis, 1188 amissis, Cas. 404 obiexis,
Merc. 484 rfwM, M. G. 1125, 1245, 1372 /axis, Most. 523 respexis,
808 /axis, True. 943 /axis, Vidul. 83 dixis, 91 demuiassis. In
Rud. 982 a««'j (Seyff. Sch.) would have no precise parallel ;
Sonnenschein reads aiisu's. Most. 518 is a conjecture. M. G.
669 optassis is conjectural, as apodosis to a si clause, and is not
counted. 1

3d person,

haud ausit, M. G. 11, is hypothetical and exactly like hand
ausim. In Ba. 697 non ausit has an expressed protasis.

All other cases, sing, and plur., are in wishes or asseverations.
Capt. 622 it a rex deorum /axil, Cist. 742 at uos Salus seruassit,
Most. 398 ila /axil Juppiter, Ps. 14 Iuppiter prohibessit, 923 ita
/axit Iuppiter, Pers. 330 perennitassit.

In the plur. all are with names of gods or di, diui. With
utinam, Aul. 50 adaxint, Amph. 632 /axint. Also di te {ilium)
/axint, Most. 463, Vidul. 86 ; di te seruassint, Asin. 654, Cas. 324,
Ps. 37, Trin. 384; ita di me seruassint, St. 505; ita di /axint,
Aul. 149, 257, 788, Capt. 172, Pers. 652, Poen. 909, 911; di
/axint, Cist. 151 ; di meliora (menus') /axint, Poen. 1400, Ps. 315,
Merc. 285; me amassint, Cure. 578. There is a protasis with
Cist. 523.

1 Of the 34 cases of prohibition with perfect forms (19 with caue, 15 with ne
or compounds), all but one, Cure. 384 nil monueris, are of the 3d conjugation.
Of the 33 prohibitions with sigmatic aorist forms, 10 are of the 1st conjugation.
In other words, the perfect in -ui is used in prohibitions only 4 times (monueris,
siueris, siris 2) out of 67, though verbs which make their perfects in -ui are
used 14 times. This can scarcely be accidental ; it must indicate some
relation between the aoristic forms and the prohibitory use.

THE SUBJUNCTIVE IN PLAUTUS.

167

It should be noted that of the forms in -s- all in the 1st pers.
are hypothetical, all in the 2d pers. are prohibitions, and all but
one in 3d pers. are optative.

Present.

Singular.

Plural.

ISt.

2d.

3d.

ISt.

2d.

3d.

A. Of will,

17

148

172

94

3

148

582

uelim,

74

74

Hypothetical,

17

3t

12

23

83

B. Questions,

212

24

20

3

2

261

C. Paratactic,

84

69

60

7

10

30

260

D. With ne,

61

28

1

13

3

106

A.

uellem,

Hypothetical,

B.

C.

D.

404

2

17
o

8
1
o

333

14

292

105

26

Imperfect.

1 4

1 6

1 6

3 o

206

28

20

19

Perfect.

10

A.

1

5

2

2

10

Hypothetical,

1

1

2

B.

5

1

2

8

C.

19

6

2

27

D.

15

I

3

19

7

40

II
Pluperfect.

I

7

A.

1

1

1

2

5

Hypothetical,

1

1

2

B.

O

C.

D.

1

2

2

2

7

With perf. ptc,

9

Forms in -s-.

A.

6

21

27

Hypothetical,

14

1

15

B.

4

4

C.

16

16

D.

16

16

32

1366

26
17

5
15
11

3

77

66

— 16

21 78

1603

E. P. Morris.