The combination ἄλλο τι ἤ—; “is there any thing else than-?" necessarily anticipates an affirmative response. The is very often omitted, and aλOT alone is then equivalent to nonne? as in Plat. Resp. p. 369: ἄλλοτι γεωργὸς μὲν εἶς, ὁ δὲ οἰκοδόμος, “ of course "of course one is a husbandman, and the other a builder." There is a great risk of missing this idiom in some. passages, e. g. in Plat. Theaetet. p. 159 D: órav dè àσlevoûvтa, äλλoti πρῶτον μὲν τῇ ἀληθείᾳ οὐ τὸν αὐτὸν ἔλαβεν; where the denial of identity suggests another rendering to the uninitiated student.

536 If a mere hypothesis is called in question, the answer expected is necessarily negative; thus,

(a) “Exλŋv toú éσTI, "I suppose he is a Greek."

(5) οὔτι που Ἕλλην ἐστί, “I suppose he is not a Greek. Hence interrogatively,

(ε) ή που Ἕλλην ἐστί; num Graius est? i.e. “he is not a Greek, is he?" or "he is not a Greek, I suppose." "No!"

537 Since, therefore, un forbids or negatives an assumption, its appearance in an interrogation presumes a negative reply; thus, åρа μý éστiv åσlevýs; num ægrotat? i.e. "he is not ill, I suppose" or "he is not ill, is he?"

In questions μn is often combined with our under the form μov, and we have sometimes even μov ovv, as Plat. Soph. 250 D: μῶν οὖν ἐν ἐλάττονί τινι νῦν ἐσμὲν ἀπορίᾳ, “ surely we are not in a less difficulty now?" And μov has become so entirely an interrogative particle, that it is followed by either μý or où, according as the answer expected is negative or positive; as Plat. Phædo, 84 c: μῶν μὴ δοκεῖ ἐνδεῶς λέλεχθαι; and Id. Soph. 234 Α: μῶν οὐ παιδιὰν νομιστέον ;

We have also the combination un où in questions both with the indicative and with the subjunctive, as Plat. Meno, p. 89 c: ἀλλὰ μὴ τοῦτο οὐ καλῶς ὡμολογήσαμεν, " but perhaps we have done rightly in making this admission;" Cratyl. 436 в: ảλλà μǹ oux ouтws ëxn, “but perhaps this is not so." Here the interrogation is virtually lost, and un, like the later Greek μýоте, means simply "perhaps." It is common to consider these phrases as elliptical or presuming the loss of φοβοῦμαι oι ὅρα (below, 538,

539). But the construction with verbs of fearing, being that c the indirect and dependent interrogative, must be subsequent to that of the direct question or prohibition.

§ XIII. Mý after Verbs of Fearing, &c.

538 These interrogations with μn and the indicative mood, like the prohibitions and deprecations with μn and the subjunctive or optative (529, (c), (d)), are appended to verbs of fearing and circumspection, with this difference:

(a) If the object of our fear or forethought is regarded as certain, we use the indicative.

(b) If uncertain, we use the subjunctive or optative.

Thus,

(α) μὴ ἀμφοτέρων ἡμαρτήκαμεν, “we have not lost both, have we?" Answer, "No!"

But by prefixing poßoîμai, we do away with the negative, so that poßoûμai-un signifies forsitan, "perhaps ;" and poßoûμai-μǹ ἀμφοτέρων ἡμαρτήκαμεν = “ I fear we have missed both—it is but "I too probable" (Thucyd. III. 55).

In the same way with a verb of circumspection:

μὴ δόκησιν εἴχετ ̓ ἐκ θεῶν, “ you had not a mere fantasm sent from the gods, had you?" Answer, "No!" But if we prefix σκοπεῖτε, we annul the particle μή, which had negatived the hypothesis, so that

σκοπεῖτε-μὴ δόκησιν εἴχετ ̓ ἐκ θεῶν

(Eurip. Helen. 119)

will signify "look to it, if you had not some vision sent from the gods"-i.e. "it is more than probable that you had." Similarly μǹ πailwv čλeyev; "he did not speak in jest, did he?" Answer, "No!" But if we prefix opa the negation is annulled, and the sense of probability is introduced, so that ὅρα-μὴ παίζων ἔλεγεν (Plat. Theatet. p. 145 B) will signify "it is probable that he spoke in jest."

(b) μn Oáva, "let me not die"-nego suppositionem me moriturum esse, vel pono me moriturum non esse.

Sédoixa-μn Oáva, "I fear I shall die-it is but too probable."

Similarly ἔφη δεδοικέναι μὴ θάνοι, " he said he was afraid he should die;" for the optative, being by nature an indeterminate tense, is properly used after other past tenses (above, 292, 513; below, 607).

Obs. There is the same difference between poßoûμai μn and ovk oid ei, as between forsitan and haud scio an: the former signifies that it is probable ; the latter that it is unlikely; thus, οὐκ ἂν οἶδ ̓ εἰ δυναίμην (Plato, Tim. p. 263) = poßoûμai pyjov Súvwμai. The apodotic av shows that our old' ei is adverbial.

539 We may also say in the indicative usage, (a) opa μǹ ovx OUTW TaûT' exei (Plat. Alcib. II. p. 139 D), "perhaps this is not the case; and with the subjunctive or optative, (b) poßoûμaι-μǹ). οὐ-θάνω, “I fear I shall not die;” ἐφοβούμην-μὴ οὐ-θάνοιμι, “Ι feared I should not die," according to 534.

Obs. That these usages do not belong to the syntax of the illative or final sentence appears (1) from the sense, for the meaning is not "with the consequence that it is not so" (below, 602, (d)), or "to the end that it may not be so" (below, 611), but simply "whether it be so ;" (2) from the omission of the particles wore or iva, oπws, &c.; (3) from the analogy of the Latin; for vereor ut veniat means "I fear how he can come," i.e. "I fear he will not come;" but efficio ne veniat for efficio ut ne veniat would mean "I manage to the end that he may not come;" so that the negative in the one case is expressed by ut alone, and in the other by ne for ut ne.

§ XIV. Construction of où μń.

540 (a) Ov with the Future or Subjunctive in Interrogations.

When the interrogative où is used with the future tense, the result is a positive command (523); when it is used with the subjunctive, the result is a deliberation nearly amounting to a resolve (521). The former construction most frequently occurs in the second person, the latter in the first; as

où μéveis; quin manes? "will you not remain ?" i.e. "stop!" and it is expected that the person addressed will do so (535);

ovк tw; nonne ibo? "shall I not go?" which implies "of course I shall."

541 (b) Mn with the Future Indicative or Aorist Subjunctive.

But if we prefix un to the future indicative or aorist subjunctive, the result is, of course, a prohibition (529, (c)); thus,

in Soph. d. Τ. 216: τἄμ' ἐὰν θέλῃς ἔπη κλύων δέχεσθαι, ἀλκὴν λάβοις ἂν κανακούφισιν πόνων, “ if you shall be willing to hear and receive my words, you would get succour and an alleviation of your troubles," where the mind supplies the intervening consequence, "I will speak, and perhaps my words might produce the effect." We have the protasis of III. with the apodosis of I. (502, (a), (cc)), or of IV. (a), without av (502, (8), (cc)), in Xen. Cyr. II. 1, § 9: ἐγὼ μὲν ἂν εἰ ἔχοιμι ὡς τάχιστα ὅπλα ἐποιούμην πᾶσι Πέρσαις, as this is followed by the second case: κἂν ταῦτα παρασκευάσῃς ἡμῖν μὲν ποιήσεις, κ. τ. λ., and immediately preceded by the third case: οὐδ ̓ εἰ πάντες ἔλθοιεν Πέρσαι, πλήθει γε οὐχ ὑπερβαλοίμεθ ̓ ἂν τοὺς πολεμίους, before which we have in succession (§ 8): εἰ οὕτως ἔχει, τί ἂν ἄλλο τις κρεῖττον εὕροι; and εἴ τι πείσονται Μῆδοι, εἰς Πέρσας τὸ δεινὸν ἥξει, we may see that in the whole passage the protasis and apodosis are changed to suit the various shades of confidence or uncertainty with which the assumptions are put forth.

504 The apodosis is very often used in cases III. and IV. without any protasis, and with the same distinction of meaning as if a protasis had been expressed; thus we have in Soph. Aj. 88:

μένοιμ' ἄν· ἤθελον δ ̓ ἂν ἐκτὸς ὢν τυχεῖν,

where the optative is used, as it very often is, to express a constrained future, “I suppose I must remain," and the indicative expresses, "but if it were possible, I should like to be out of the way."

505 The most common substitutes for ei, in all these cases of protasis, are the participle without the article, and the relative with indefinite antecedent. Thus we can say, with scarcely any difference of meaning:

On the other hand, the infinitive and participle may take the place of the finite verb in the apodosis, whenever the latter appears in a dependent sentence, which requires either of these verb-forms (below, 593, 594). Thus (a) the particle av is very often found with the infinitive after such verbs as οἴομαι, δοκῶ, νομίζω, ἡγοῦμαι, ἐλπίζω, ὑπολαμβάνω, denoting opinion or expectation, and also after verbs like λέγω, φημί, ὁμολογῶ, ὑπισχνοῦμαι, ὄμνυμι, denoting the expression of the thoughts in words with reference to something conditional, as Thucyd. II. 20: τοὺς ̓Αθηναίους ἤλπιζεν ἴσως ἂν ἐπεξελθεῖν, καὶ τὴν γῆν οὐκ ἂν περιϊδεῖν τμηθῆναι, because in the independent sentence we should have had ἴσως ἂν ἐπεξέλθοιεν καὶ οὐκ ἂν περιΐδοιεν. It is more than doubtful whether the future infinitive is ever used with av. Instances are found in some of the existing texts, but they seldom stand the test of criticism (see Preface to Thucydides, p. xi). We find äv with the infinitive used substantively, as Thucyd. VII. 62: Sià Tò βλάπτειν ἂν τὸ τῆς ἐπιστήμης, “ on account of the fact that it would be a hindrance to the application of our skill." (b) The apodotic use of the participle with av is generally found in objective, relative and causal sentences; as Thucyd. 1. 76: εὖ ἴσμεν μὴ ἂν ἧσσον ὑμᾶς λυπηροὺς γενομένους, “ we are quite convinced that you would not have been less vexatious," where the protasis is εἰ ὑπομείναντες απήχθησθε. Plat. Crit. p. 48 c: τῶν ῥᾳδίως ἀποκτιννύντων καὶ ἀναβιωσκομένων γ' ἄν, “ of those who would without hesitation slay and restore to life again.” Thucyd. I. 73: ἀδυνάτων ἂν ὄντων πρὸς ναῦς πολλὰς ἀλλήλοις βοηθεῖν, " as they would have been unable to assist one another when opposed to so many ships.” Xen. Anab. I. 1, § 10: ὡς οὕτω περιγενόμενος ἂν τῶν ἀντιστασιωτών, “on the ground that he would in this way have got the better of his political opponents." On the repetition of av with the participle when it really belongs to the verb of the sentence, see below, 508, (a).

66

4