Politic. II. 9. (= ΙΙ. 6. p. 56, 8. Ed. Goettl.) λέγουσι δέ, ὡς ἐπὶ μὲν τῶν πρότερον βασιλέων μετεδίδοσαν τῆς πολιτείας, ὥστ ̓ οὐ γίνεσθαι τότε ὀλιγανθρωπίαν πολεμούντων πολὺν χρόνον.

It will be observed that all the four passages are in the orat. obliqua and in all if turned into the orat. directa the indicative would be used, e. g. in the passage cited from Æschines, the (supposed) words of Demosthenes are λήσω μεταλλάξας...τὴν ̓Αθηναίων ἀκρόασιν, καὶ περιστήσω.. ὅταν ἐγὼ δεῦρο παρέλθω,...τοσούτους δὲ καὶ τ. ἐκκαλέσομαι ...τὰς Αἰσχίνου δ. τὴν δι' Αἰσχίνου...ὥστ ̓ οὐδὲ ἀπαντήσεται οὗτος... ἀπολογησόμενος ὅταν... διδῷ ἀλλ' ἀγαπήσει ἐὰν περιπέσῃ ...ζημιῶται. And in Demosth. § 351. οὕτω δὲ ἄτοποί τινες...καὶ δυσχερεῖς ἄνθρωποί εἰσιν, ὥστε οὐκ αἰσχύνονται.

Similar is the usage of wore où with the infinitive after verbs of thinking, hearing. Dem. de Coron. p. 320. § 349. πότερ' οὐχ ἡγεῖ γιγνώσκειν αὐτοὺς ὅστις εἶ; ἢ τοσοῦτον ὕπνον καὶ λήθην ἅπαντας ἔχειν, ὥστ ̓ οὐ μεμνήσθαι τοὺς λόγους οὓς ἐδημηγόρεις ἐν τῷ δήμῳ (where omit ἡγεῖ, and you will have ἅπαντες ἔχουσιν, ὥστ ̓ οὐ μέμνην ται). Plat. Apolog. 26 D. 'Αναξαγόρου οἴει κατηγορεῖν, ὦ φίλε Μέλητε· καὶ οὕτω καταφρονεῖς τῶνδε καὶ οἴει αὐτοὺς ἀπείρους γραμμάτων εἶναι ὥστε οὐκ εἰδέναι ὅτι τὰ ̓Αναξαγόρου βιβλία, τοῦ Κλαζομενίου, γέμει τούτων τῶν λόγων (οὕτως ἄπειροί εἰσιν, ὥστε οὐκ ἴσασι). Thucyd. V. 40. τοὺς γὰρ Βοιωτοὺς ᾤοντο πεπεῖσθαι. τούς τε ̓Αθηναίους εἰδέναι ταῦτα, ὥστε οὐδὲ πρὸς ̓Αθηναίους ἔτι σφίσιν εἶναι ξυμμαχίαν ποιήσασθαι. Dem. III. Philipp. p. 123. § 59. ἀκούω Λακεδαιμονίους...οὕτω δ ̓ ἀρχαίως εἶχον...ὥστε οὐδὲ χρημάτων ὠνεῖσθαι παρ' οὐδενὸς οὐδέν Lys. Theomn. p. 117. St. = 355 R. ὑμᾶς μὲν πάντας εἰδέναι ἡγοῦμαι ὅτι ἐγὼ μὲν ὀρθῶς λέγω, τοῦτον δὲ οὕτω σκαιὸν εἶναι ὥστε οὐ δύνασθαι μαθεῖν τὰ λεγόμενα2.

There are passages however which I cannot bring under this canon, and which I have no right to suppress. In Dem. de Coron. p. 267. § 154. Bekker has no doubt shewn his usual judgment in restoring ὥστ ̓ οὐ δύνασαι (vulg. δύνασθαι, which grates upon the ear, and is not the reading of the best MSS.). But what shall we make of the following? Nicostrat. p. 1246. § 2. οὐδ ̓ αὖ οὕτως ἄπορος ἦν οὐδ ̓ ἄφιλος, ὥστ ̓ οὐκ ἂν ἐξευρεῖν τὸν ἀπογράψοντα. Eurip. Phoen. 1357, 8. οὐ μακρὰν γὰρ τειχέων περιπτυχαὶ | ὥστ ̓ οὐχ ἅπαντά σ ̓ εἰδέναι τα δρώμενα. Soph. Εlectr. 780, 1. ὥστ ̓ οὔτε νυκτὸς ὕπνον οὔτ ̓ ἐξ ἡμέρας | ἐμὲ στεγάζειν ἡδύν. I leave these to the reader, hoping he

1 In Thucyd. VIII. 76. ὥστε οὐδὲ τούτους... χείρους είναι the sentence is in a transition state from oratio obliqua to oratio directa.

2 In both of these passages there is the same latitude of construction which we meet with elsewhere, e. g. Demosth. Nicostr. Ρ. 1254. § 33.

will be able to find the key to the construction, or hit upon a happier emendation than occurs to me. Is it possible that in the first two passages, (which are very similar1), the verb v is extended to the second clause in the former, and in the latter eoTv is to be repeated from eioív? The difficulty which lies in the way of this explanation is, that the substantive verb will be impersonal in the second clause, which is to my mind insurmountable. In the passage from Sophocles a bolder critic than I confess myself to be might cut the knot by reading: ὕπνος...ἔμ' ἐστέγαζεν ηδύς (compare the following words, ἀλλ' ὁ προστατῶν | χρόνος διηγέ μ'). Ι ἔμ' ἐστέγαζεν were once corrupted into éμè σtéyačev, the alteration of the words into their present form would naturally follow2.

APPENDIX C.

THE particle val (answering to our nay, and the Latin næ) appears indeed to belong to the affirmative: yet upon due enquiry I am persuaded it will be acknowledged that it has a sort of negative meaning. The remarks of a writer in the Philological Museum (a book, the sale whereof, I am sorry to say, has not kept pace with its desert) Vol. II. p. 323. are worth quoting: "val is the Latin næ, and as Grimm3 observes, seems, singularly enough, allied to the negative." He purposely avoids entering upon the subject of this connexion. May it not have arisen from the use of the negative, like our "nay," immo, or anzi (ante) in Italian? which are negative inasmuch as they object to the preceding phrase as not being strong enough, whilst they agree with its general meaning, and enhance its force? Certainly in these cases the negative and affirmative senses often approach very near to one another, as, for instance, in the following passage of B. Jonson : "A good man always profits by his endeavour, yea, when absent; nay, when dead, by his example and memory."

1 And are not corrupt, according to the judgment of a friend whom I am not at liberty to mention, as I have not had an opportunity of asking his leave, but in whose opinion, in matters appertaining to Greek philology, I place unlimited confidence.

2 The (supposed) corruption is however sanctioned by no MS. and Eustath. Iliad. p. 632, 31. quotes from Sophocles, ὕπνον ἐμὲ στεγάζειν ἡδύν.

3 I regret to say that I do not possess Grimm's works. I have hitherto been unable to verify the passage referred to.

There appears to me to be great truth in this remark. However, I have at present nothing to do with the investigation of the Latin næ, or the English nay, but will endeavour, from the Greek authors, to ascertain the meaning of vaí. I have observed in the Orators the following passages. Dem. de Coron. p. 310. § 313. ναί, φησίν· ἀλλὰ τὸ τοῦ Κεφάλου καλόν, τὸ μηδεμίαν γραφὴν φυγεῖν. Aristocr. p. 652. § 114. ναί· πῶς ; ἐγὼ διδάξω. [Dem.] Aristog. p. 791. § 84. ναί, φησίν· ἀδίκως γάρ μ' ἐγγέγραφεν. Pantænet. 977. § 49. ναί, φησίν· ἀλλὰ κἂν ἄλλο τι ἀδικῇ τις περὶ τὰ μέταλλα, καὶ τούτων εἰσὶ δίκαι. Æschin. de Coron. p. 57. St. = 423 R. ναί· ἀλλ ̓ ἀντιδιαπλέκει πρὸς τοῦτο εὐθὺς ὡς οὔτ ̓ ἔλαχε τειχοποιὸς οὔτ ̓ ἐχειροτονήθη ὑπὸ τοῦ δήμου. p. 60 = 437. ναί, εἴ γέ σέ τις ἄλλη πόλις στεφανοῖ· εἰ δὲ ὁ δῆμος ὁ ̓Αθηναίων. p. 65 = 478. ναί· ἀλλὰ χαλκοῖς καὶ ἀδαμαντίνοις τείχεσιν, ὡς αὐτός φησι, τὴν χώραν ἡμῶν ἐτείχισε. Dinarch. c. Demosth. p. 91. St. =6 R. ναι· κατέ ψευσται γὰρ ἡ βουλὴ Δημοσθένους. In all these passages I conceive it will be granted that vai has at least a quasi-negative meaning, answering to our "nay;" and in fact does not so much serve the purpose of affirming the preceding position, as of ushering in the subsequent objection.

I now come to the use of the word in other writers. It occurs in Homer's Iliad seven times: I. 234. ναὶ μὲ τόδε σκῆπτρον. I. 286. ναὶ δὴ ταῦτά γε πάντα, γέρον, κατὰ μοῖραν ἔειπες, which line recurs with the necessary variation, VIII. 146. X. 169. XXIII. 626. XXIV. 379. and XVIII. 128. ναὶ δὴ ταῦτά γε, τέκνον, ἐτήτυμον, which is very similar. In the Odyssey we meet with it four times, (in all a repetition of Il. I. 286.) IV. 266. XVIII. 170. XX. 37. XXII. 486. It appears in the Hymn to Mercury 460. ναὶ μα τόδε κρανέινον ἀκόντιον. Thus in the twelve passages from the writings ascribed to Homer, twice it is joined with μά, a particle used in swearing, “Nay by:” and ten times in answering, correcting, and paving the way for a coming objection.

In Æschylus I have observed it only twice: Pers. 738. ναί· λόγος κρατεῖ σαφηνὴς τοῦτο κοὐκ ἔνι στάσις, and 1071. ἰω δῆτα ναὶ ναί, The latter passage may safely be left. In the former the particle is not a mere "yes:" but corresponds to our colloquial " quite clear."

nay, this is

In Sophocles the instances are either in affirmative reply to a question, (Ed. Col. 27. Electr. 845. (and ναιχὶ Cd. Τ. 684.) or (as above) used in bringing forward an objection, Philoct. 372. (quoted on § 118.) Trachin. 424. ναί· κλύειν γ ̓ ἔφασκον. ταὐτὸ δ ̓ οὐχὶ γίγνεται | δόκησιν εἰπεῖν καξακριβῶσαι λόγον.

rous.

In Euripides the examples are, as might be expected, more numeI shall refer to such as bear out the negative signification of the

particle: Med. 1277. ναί, πρὸς θεῶν, ἀρήξατ'· ἐν δέοντι γάρ. Hipp. 605. ναί, πρός σε τῆς σῆς δεξιᾶς εὐωλένου. Alcest. 1119. ναί, σώζε νῦν, καὶ τὸν Διὸς | φήσεις ποτ ̓ εἶναι παῖδα γενναῖον ξένον. Phan. 1665. ναί, πρός σε τῆσδε μητρὸς Ιοκάστης, Κρέον. Iph. Aul. 1247. ναί, πρὸς γενείου σ ̓ ἀντόμεσθα δύο φίλοι. Cyclops, 147. ναὶ δὶς τόσον πῶμ ̓ ὅσον ἂν ἐξ ἀσκοῦ ῥνῇι. In all the above-cited passages I conceive the original idea of a negative is visible.

My examination of Plato also bears testimony to this opinion: but as I am unwilling to carry out this Appendix to further length, I will refer the reader to Ast's Index.

I infer, then, that val is fundamentally connected with a negative idea that its affirmative usage is purely accidental [Compare on the contrary, the apparently) negative senses of μὲν οὖν, μέντοι, μήν], and always, when so employed, serves to strengthen the words coming from the preceding speaker, by way of taking an objection to their sufficiency: that in a considerable number of instances the negative idea predominates: and that in the Orators, at any rate, it is never met with in a strictly affirmative signification. The last remark will of course be confirmed, or rectified, by examples which I shall feel thankful to be furnished with: for I hold myself a lover of truth and for truth's sake, and am wishful to adopt as my motto the words of Socrates with reference to this question or any opinion raised in this Volume: ἐγὼ δὲ τίνων εἰμί; τῶν ἡδέως μὲν ἂν ἐλεγχθέντων, εἴ τι μὴ ἀληθὲς λέγω, ἡδέως δ ̓ ἂν ἐλεγξάντων, εἴ τίς τι μὴ ἀληθὲς λέγοι, οὐκ ἀηδέστερον μέντ ̓ ἂν ἐλεγχθέντων ἢ ἐλεγξάντων· μεῖζον γὰρ αὐτὸ ἀγαθὸν ἡγοῦμαι, ὅσῳπερ μεῖζον ἀγαθόν ἐστιν αὐτὸν ἀπαλλαγῆναι κακοῦ τοῦ μεγίστου ἢ ἄλλον ἀπαλλάξαι. (Plat. Gorg. 458 Α.)

1 Hermann reads from conjecture καὶ δίς τόσον κ. τ. λ. I have not a doubt

but that vai expresses the same meaning much more forcibly.

LIOTHE