100 PROCEEDINGS OF THE AMERICAN ACADEMY 



the Optative with av is equivalent to the primary tenses of the Latin 

 Subjunctive, and not to the secondary tenses : thus, TroiT^crat/xt av is equiv- 

 alent to faciam (not to fecissem, which would be eirolrja-a av). Here 

 likewise the Subjunctive cannot be used in Attic Greek. This analogy 

 between the Optative and the primary tenses of the Latin Subjunctive 

 might lead us even to the view that the latter ought rather to be called 

 an Optative, for which view there are certainly much stronger reasons 

 than for the opposite one which we are considering. The analogy of 

 the Sanskrit also seems to show that the Present Subjunctive of the 

 Latin and the Optative of the Greek are descended from the same 

 original forms. 



In indirect quotations and questions the Optative is used after past 

 tenses, each tense of the Indicative or Subjunctive in the direct discourse 

 being then changed to the corresponding tense of the Optative. Thus, 

 emev on a dvvaiTo noirja-oi, he said that he ivould do toliat he could, implies 

 that the direct discourse was, a av bivajxai TroiT]a-a>. Here the Optative may 

 be the correlative of the Subjunctive ; but it is quite as often the correla- 

 tive of the Indicative, as the Subjunctive can stand after primary tenses 

 only when it would have been used in the direct discourse. One tense 

 of the Optative, the Future, can never represent a Subjunctive, as that 

 mood has no corresponding tense ; but it always represents a Future 

 Indicative. Nothing more need be said to show the absurdity of call- 

 ing this tense a secondary tense of the Subjunctive. The three remain- 

 ing tenses of the Optative can with no more propriety be called secon- 

 dary tenses of the Subjunctive than of the Indicative, for they represent 

 both on precisely the same principles. This is especially obvious in 

 regard to the Aorist, which has two distinct meanings in indirect ques- 

 tions, — one when it represents an Aorist Indicative, and another when 

 it represents an Aorist Subjunctive, the direct form. Thus, rjyvoei ri 

 TToiTjcreifv may mean either he knew not tohat he had done, or he knew 

 not what he should do ; as the direct question may have been either ri 

 firoirjaa ; what did I do? or W ttoitjctco ; {Aov. Suhj.), what shall J do ? 

 Strangely enough, this very class of sentences is supposed to furnish 

 the most striking analogy between the Latin Subjunctive and the Greek 

 Subjunctive and Optative combined. iVbn habet quo se vertat, and non 

 hahehat quo se verteret, are indeed equivalent to ovk. exei ottt] Tpm-qTai. and 

 ovK flxpv onr] rpanoiTO, but a single example like r]pa>Tcov avrbv el dvan^fv- 

 treiev, I asked him whether he had set sail (Dem. in Polycl. p. 1223, 

 21), in which dvan'Kevcreiev represents an Aorist Indicative, shows that 



