-^'' PA 6604 H44 1917 MAIN UC-NRLF m "i'!:i-i Tfe Sequence of Tenses in Plautus A Thesis Presented to the Faculty of the Graduate School in partial fulfiHraent of the requirements for the degree of Doctor of Philosophy By Edward Hoch Heffner Philadelphia, Pa. 1917 'KQ' Mnttifrait^ of J^^ttnatllwanta The Sequence of Tenses in Plautus A Thesis Presented to the Faculty of the Graduate School in partial fulfillment of the requirements for the degree of Doctor of Philosophy By Edward Hoch Heffner Philadelphia, Pa. 1917 Lancaster, Pa. Intelligencer Printing Co. v4^ The writer takes opportunity here to express his sincere thanks to Prof. J. C. Rolfe, Prof. W. B. McDaniel, Prof. R. G. Kent, Prof. G. D. Hadzsits, Prof. H. B. Van Deventer, and Dr. T. A. Buenger, for the assistance and encouragement given to him during the preparation of this thesis. 3G1263 CONTENTS Chapter I Page Introduction I Chapter II Primary sequence dependent upon the perfect: 1. In final clauses, including substantive clauses developed from the volitive 3 (a) Other than those dependent upon certum est and the perfect of mereo and its compounds 3 (b) Dependent upon certum est 1 1 (c) Dependent upon the perfect of mereo and its compounds 12 2. In indirect questions 13 3. In relative clauses of characteristic 16 4. In consecutive clauses 18 5. In conditional clauses of comparison 22 Chapter III Violations of the sequence principle: 1. Repraesentatio 23 2. Violations caused by other factors 32 Chapter IV Apparent violations of the sequence principle: 1. In connection with the phrase quod sciam 34 2. Other apparent violations 35 Chapter V Instances of mechanical conformity to the sequence principle 37 Chapter VI Shift in the sequence 45 Chapter VII Conclusion 46 The Sequence of Tenses in Plautus Introduction The aim of this study has been to examine the tenses of the dependent subjunctives in Plautus from two points of view: in the first place, to determine to what extent the tense usage is at variance with the familiar doctrine of the Sequence of Tenses, and to explain the exceptions, real or apparent, to the rules of Sequence; and, in the second place, to ascertain what evidence there is against the theory that the dependent, just as the independent, subjunctives denote absolute time and not time relative to that of the verb in the principal clause. (See below.) In 1886-1888 Professor W. G. Hale made a violent attack on the doctrine of the Sequence of Tenses {The Sequence of Tenses in Latin, American Journal of Philology, VII. 446-465; VIII. 46-77; IX. 158-177). His conclusion briefly summed up in his own words is that "in dependent as in independent sub- junctives, the tense conveys meaning, and owes its choice to that fact" {A. J. P., VII. 465), or again, that ''the tenses of the Latin subjunctive, alike in dependent and in independent sentences, tell their own temporal story — that no such thing as is meant by the doctrine of the Sequence of Tenses exists" {A. J. P., VIII. 59). Professor B. L. Gildersleeve in a reply to the first two of Pro- fessor Hale's articles took exception to Professor Hale's conclu- sion (^. /. P., VIII. 228-231). In the third of Professor Hale's articles he restates his doctrine briefly in the words, "the tenses of the subjunctive convey meaning" (A. J. P., IX. 160), and calls attention to the fact that his earlier statement "that no such thing as is meant by the Sequence of Tenses exists'^ was in- cautious. Some years later, A. T. Walker, now Professor of Latin at the University of Kansas, dealt with the subject in his doctoral dissertation (The Sequence of Tenses in Latin: A Study Based on Caesar's Gallic War, Lawrence, Kansas, 1899), and 2 The Sequence of Tenses in Plautus partially refuted Professor Hale's argument. In 1913 the Joint Committee on Grammatical Nomenclature {Report, 57, 60-62, University of Chicago Press, 1913) suggested that the use of the tenses in Latin be taught in accordance with Professor Hale's doctrine. In December, 1914, Professor R. G. Kent read a paper before the American Philological Association at its meeting at Haverford College, entitled The ''Passing'' of the Sequence of Tenses {The Classical Weekly, 9. 2-719-13), in which he examined the material used in Professor Hale's articles together with a certain number of additional examples drawn mainly from the Latin commonly read in the Schools and in the first year of the College {The Classical Weekly, 9. 3, footnote 10). Professor Kent protested against the "the shelving of the principle of the Sequence of Tenses" {The Classical Weekly, 9. 13). In the spring of 191 5, Professor A. T. Walker {Sequence or Harmony of Tenses?, The Classical Journal, 10. 246-251; 291-299) further discussed the subject and argued for a retention of the term sequence. A few brief discussions of Professor Kent's articles by Miss Susan Fowler, of the Brearley School, New York City, by Miss Eliza- beth Mcjimsey Tyng, of the Packer Collegiate Institute, and by Professor B. M. Allen, of the Phillips Andover Academy, with rejoinder by Professor Kent, appeared in the spring of 1916 {The Classical Weekly, 9. 193-198). The text used as a basis in this investigation is the edition by Lindsay in the Scriptorum Classicorum Bibliotheca Oxoniensis. Chief among other editions consulted is that by Goetz and Schoell. Unless otherwise stated, the citations are from Lind- say's edition. Illustrative material drawn from authors other than Plautus has been included, but the examples thus used are not the fruits of a systematic investigation, and have been embodied merely because they seemed to be helpful in explaining the passages in connection with which they are quoted. II Primary Sequence Dependent Upon the Perfect In this Chapter are treated the instances of the perfect indica- tive as a primary tense. Inasmuch as in the great majority of cases the perfect takes the secondary sequence, it seemed ad- visable to bring together here the examples of primary sequence, largely because of their value in a comparative study of the material in Chapter V. The perfect tense is primary when it denotes the present resultant state from a past act, and secondary when it denotes the past act. The line of division between the two capacities in which this tense serves is frequently hard to draw. Occasionally we find examples of the perfect clearly denoting a present state and still taking the secondary sequence, as may be seen from the examples in a portion of Chapter V (pp. 40-44). The dependent subjunctives are treated in order according to the syntax of the clause in which they stand. I. In final clauses, including substantive clauses developed from the volitive: (a) Other than those dependent upon certum est and the perfect of mereo and its compounds. Amph. II sq. (nam uos quidem id iam scitis concessum et datum mi esse ab dis aliis, nuntiis praesim et lucro) : Praesim stands in a substantive clause developed from the volitive and depends on scitis concessum et datum esse. Although the infinitives are in the perfect tense, Mercury wishes to call attention not to the act of giving and granting of certain powers and duties to him, but to his present possession of those powers. The passage therefore means "you know by this time that I have this duty, which has been granted and given to me, of being in charge . . . .", which is a present state and requires the primary sequence. 3 4 The Sequence oj Tenses in Plautus Amph. 64 sqq. nunc hoc me orare a uobis iussit luppiter ut conquistores singula in subsellia eant per totam caueam spectatoribus, si quoi fauitores delegates uiderint, ut is in cauea pignus capiantur togae; In this sentence itissit denotes the present state resulting from the act of ordering. The principal clause means "I have an order that . . . ." The acts denoted by eant and capiantur are future relatively to the time of speaking. Cf. M. G. 971 iussit, p. 9. Amph. 195 me a portu praemisit domum ut haec nuntiem uxori suae, At the time Sosia says these words to Mercury the message with which he has been entrusted is still to be delivered to Alcmena. The principal clause means "I am here, having been sent ahead home from the harbor to announce . . . ," which is a present resultant state and demands the primary sequence. For the converse of this, compare Horace Serm. II. 6. 34 sqq. Ante secundam Roscius orabat sibi adesses ad Puteal eras, de re communi scribae magna atque nova te orabant hodie meminisses, Quinte, reverti. Amph. 869 sq. simul Alcumenae, quam uir insontem probri Amphitruo accusat, ueni ut auxilium/emm; Vent is equivalent to adsum, a present state, and for this reason governs the primary sequence. Aul. 743 at ego deos credo uoluisse ut apud me te in neruo enicem. Here uoluisse means "I have willed," "desire," which is a present state and requires the primary sequence. See on Capt. 267, p. 6. Bacch. 350 sqq. exorsa haec tela non male omnino mihi est: ut amantem erilem copem facerem filium. Primary Sequence Dependent Upon the Perfect 5 ita feci ut auri quantum uellet sumeret, quantum autem lubeat reddere ut reddat patri. As is indicated by the secondary tense sumeret, the speaker has his attention fixed on the past act denoted by feci, and for this reason we might have expected him to use redderet instead of reddat; but the present is substituted by the figure of repraesen- tatio to indicate that the act of reddat is future relatively to that of sumeret. Goetz and Schoell punctuate with a comma at the end of 350 and a period at the end of 351. See p. 26. Bacch. 533 uerum postremo impetraui ut ne quid ei suscenseat. Here the principal clause means "but at last I have him per- suaded," which denotes a present state, and therefore demands the primary sequence in the substantive clause. Bacch 689 sq. MN. ego patrem exoraui. CH, nempe ergo hoc ut faceret quod loquor? MN. immo tibi ne noceat neu quid ob eam vera suscenseat; The tense of faceret shows that the speaker's attention is fixed on the act of beseeching. The presents noceat and sus- censeat are to be explained by repraesentatio. Bacch. 1082 ego dare me meo gnato institui, ut animo opsequium sumere possit; Institui means "my mind is made up," "my decision is," a present state, and therefore governs the primary sequence. Capt. 35 sq. hisce autem inter sese hunc confinxerunt dolum, quo pacto hie seruos suom erum hinc amittat domum. Hunc confinxerunt dolum is the equivalent of "they have this piece of trickery devised," a present state, and therefore takes the primary sequence. 6 The Sequence of Tenses in Plautus Capt. 266 sq. nunc senex est in tostrina, nunc iam cultros adtinet. ne id quidem, inuolucre inicere, uoluit, uestem ut ne inqidnet. Imiolucre .... uoluit .... is equivalent to "he has no napkin on him to prevent staining," a present state, which demands the primary sequence. Nunc, nunc iam enhance the intensity of the present state. See on Aid. 743, p. 4. Capt. 378 sqq. nunc ita conuenit inter me atque hunc, Tyndare, ut tu aestumatum in Alidem mittam ad patrem, si non rebitas hue, ut uiginti minas dem pro te. Conuenit . . . .is equivalent to "now there is this arrange- ment .... that . . ," which is a present state requiring the primary sequence. Capt. 395 sqq. dicito patri quo pacto mihi cum hoc conuenerit de huius filio. PHILOC. quae memini, mora mera est monerier. TY. ut eum redimat et remittat nostrum hue amborum uicem. Redimat and remittat depend on the perfect subjunctive conuenerit in the indirect question in 395, which is equivalent in meaning to "what arrangement there is . . . ," a present state, and thus demands the primary sequence. Capt. 837 nescioquem ad portum nactus es ubi cenes, eo fastidis. Nactus es is equivalent to "you chance to have," which denotes a present state, and therefore requires the primary sequence. See on True. 280, p. 11. Cas. 52 sqq. pater adlegauit uilicum qui posceret sibi istanc uxorem: is sperat, si ei sit data, sibi fore paratas clam uxorem excubias for is; filius is autem armigerum adlegauit suom qui sibi eam uxorem poscat: Primary Sequence Dependent Upon the Perfect 7 Filius .... qui is equivalent to "his armor-bearer has a commission . . .to," which is a present state, and therefore demands the primary sequence. In 52 we should have expected poscat for posceret, for the context of the passage shows that adlegauit there as well as in 55 really denotes a present state, although in the earlier verse the sequence is secondary. The commands of both, that of father and son, are being carried out at the time this passage is being spoken, and the girl's reply is still to be received. The secondary tense posceret is due to a mechanical sequence. Cas. 105 praefeci ruri recte qui curet tamen. Curet is in the present subjunctive because praefeci .... qui'is equivalent to "I have a manager on my estate to," a present state. We see from no that curet is future also relatively to the time of speaking. Epid. 354 sq. nunc iterum ut fallatur pater tibique auxilium apparetur inueni : Inueni means "I know," and therefore takes the primary sequence. Nunc emphasizes the idea of a present state. The acts of fallatur and apparetur are future also relatively to the time of speaking. See on Capt. 569, p. 17; Merc. 254, p. 15; M. G. 767, p. 8 Epid. 444 sq. non repperisti, adulescens, tranquillum locum ubi tuas uirtutes explices ut postulas. Repperisti is equivalent to habes, and therefore takes the primary sequence. Explices is future also relatively to the time of speaking. Epid. 570 sq. AC. Quid est, pater, quod me exciuisti ante aedis? PE. ut matrem tuam uideas, adeas, aduenienti des salutem atque osculum. Ego te exciui ut, to be supplied in thought from exciuisti, is 8 The Sequence of Tenses in Plauius equivalent to "you are here, having been called out by me that . • . . ," a present state, and therefore takes the primary sequence. Cf. Xen. Anab. I. 6. 6. riapexdXeaa u^xaq, avBpeq ^(Xot, oxox; auv uijlTv PouXeu6[X£voi; o tt Slxatov eaxt xal 7cp6q Oewv xal icpit; avOpwzwv TOUTo xpd^o) xept 'Op6vTa toutouI. Jkfe«. 787-8 sq. quotiens monstraui tibi uiro ut morem geras, quid ille faciat ne id opserues, quo eat, quid rerum gerat. Monstraui tibi is equivalent to "you know," and therefore takes the primary sequence. Merc. 667 sqq. Quoniam a uiro ad me rus aduenit nuntius rus non iturum, feci ego ingenium meum, reueni, ut ilium persequar qui me fugit. Reueni means "I am come back," a present state, and there- fore takes the primary sequence. M. G. 295 sq. nam tibi iam ut pereas paratum est dupliciter nisi sup- primis tuom stultiloquium. Paratum is an adjective here, and the sequence is therefore primary. See on Pseud. 579, p. 10. M. G. 728 sq. quae probast mers, pretium ei statuit, pro uirtute ut ueneat, quae improbast, pro mercis uitio dominum pretio pauper et, Statuit means "is fixed," a present state, and therefore takes the primary sequence. The text of 728, however, is dubious. Goetz and Schoell in their editio maior (1890) read quae probast et pretio digna, pro uirtute ut ueneat. M. G. 766 sqq. mihi opus est opera tua, Periplectomene ; nam ego inueni lepidam sycophantiam qui admutiletur miles usque caesariatus, atque uti Primary Sequence Dependent Upon the Perfect 9 huic amanti ac Philocomasio hanc ecficiamus copiam, ut hie earn ahducat habeatque. Inueni is equivalent to scio^ and therefore takes the primary sequence. See on Capt. 569, p. 17; Epid. 355, p. 7; Merc. 254, p. 15. M. G. 971 sq. nunc te orare atque opsecrare iussit ut earn copiam sibi potestatemque facias. Nunc. . . . ^'«55i/ w^ here, as inylw^/t. 64, means "you now have an order to," which is a present state, and requires the primary sequence. The notion of present state is emphasized by nunc. See on Amph. 64, p. 5. M. G. 1 145 sq. nam ipse miles concubinam intro abiit oratum suam ab se ut abeat cum sorore et matre Athenas. Intro abiit is equivalent to intro est, a present state, and there- fore the sequence is primary. M. G. 1238 istuc curaui, ut opinione illius pulchrior sis. Curaui means "provision is made . . . ," a present state demanding the primary sequence. M. G. 1269 induxi in animum ne oderim item ut alias, quando orasti. Induxi in animum means "my mind is made up," a present state demanding the primary sequence. Oderim, being a preter- itive, has the meaning of a present. Most. 941 sq. nisi forte factu's praefectus nouos, qui res alienas procures, quaeras, uideas, audias. Factu's .... qui, "are prefect to . . . ," is a present state calling for the primary sequence. Pers. 325 sq. nam iam omnis sycophantias instruxi et comparaui quo pacto ab lenone auferam hoc argentum. 10 The Sequence of Tenses in Plautus Instruxi and comparaui mean "I have a plan drawn up and prepared . . . ," a present state, and therefore require the primary sequence. lam intensifies the present state. Pers. 620 sq. qur ego hie mirer, mi homo? seruitus mea mi interdixit ne quid mirer meum malum. Mi interdixit, "I am forbidden," a present state, demands the primary sequence. Poen. 772 sqq. nunc hunc inimicum quia esse sciuerunt mihi, eum adlegarunt suom qui seruom diceret cum auro esse apud me; compositast fallacia ut eo me priuent atque inter se diuidant. Compositast means "is arranged," a present state, and therefore demands the primary sequence. Poen. 1018 sqq. palas uendundas sibi ait et mergas datas, ad messim credo, nisi quid tu aliud sapis, ut hortum /oc^ia/ atque ut frumentum metat. Ait datas .... means "says he has," a present state, and therefore takes the primary sequence. Pseud. 579 sqq. nam ego in meo pectore prius ita paraui copias, duplicis, triplicis dolos, perfidias, ut, ubiquomque hostibu' congrediar (maiorum meum fretus uirtute dicam, mea industria et malitia fraudulenta), facile ut uincam, facile ut spoliem meos perduellis meis perfidiis. Paraui .... perfidias is equivalent to "I am ready," which is a present state and demands the primary sequence. See on M. G. 295, p. 8. Rud. 717 non hodie isti rei auspicaui, ut cum iurciiero fahuler. Non .... auspicaui ut means "I am not prepared for," a present state calling for the primary sequence. Primary Sequence Dependent Upon the Perfect ii Rud. 927 fnunc haec tibi occasio, Gripe, optigit ut liberes ex populo praeter tef. The principal clause means "now you have this opportunity," which is a present state and demands the primary sequence. Trin. 15 dedi ei meam gnatam quicum aetatem exigat. Dedi ei is equal to is habet, a present state, and therefore takes the primary sequence. True. 280 ut nacta es hominem quem pudeat probri! Nancta es is equivalent to "you are come upon," a present state, and thus requires the primary sequence. See on Capt. 837, p. 6. The following examples of final clauses in the present subjunctive de- pendent upon the historical perfect are found in the Arguments: Capt. Arg. 7 amittatur; Cure. Arg. 5 mittat; True. Arg. 3 tangat. (b) Dependent upon cerium est. In this section are given a few forms which are regarded as dependent on certum est. Inasmuch as all these verbs belong to the third conjugation, the forms here used may be future indicatives or present subjunctives. We have the future in- dicative in Merc. 472 certumst, ibo ad medicum atque ibi me toxica morti dabo. Morris {A. J. P. XVIII. p. 145) notes that the forms treated in this section "are not quite futures." Durham {Subjunctive Substantive Clauses in Plautus, Cornell Studies XIII. pp. 83-84) refers the substantive clauses dependent upon certum est to the subjunctive of determined resolution. Inasmuch as the phrase certum est always denotes a present state, the sequence is always primary. The examples are: A sin. 248 nam si mutuas non potero, certumst sumam faenore. 12 The Sequence oj Tenses in Plautus Aul. 676 certumst, Siluano potius credam quam Fide. Aul. 681 certum est, malam rem potius quaeram cum lucro. Bacch. 382 certumst iam dicam patri, Capt. 778 sq. nunc certa res est, eodem pacto ut comici serui solent, coniciam in coUum pallium, Cas. 448 certum est, hunc Accheruntem praemittam prius. Stick. 503 sq. certumst amicos conuocare, ut consulam qua lege nunc med — essurire oporteat. — (c) Dependent upon the perfect of mereo and its compounds. The jussive origin of this usage seems to be assured by the fact that we have the negative ne in Men. 1 100. These words sometimes take the primary sequence, meaning "I am deserving of," etc., and at other times they take the secondary sequence, thus indicating that in the latter instance the speaker's attention is fixed upon the past act. The examples of secondary sequence are here given for purposes of comparison. The passages having the primary sequence are: Capt. 421 sq. pol istic me hau centessumam partem laudat quam ipse meritust ut laudetur laudibus. Epid. 442 sq. uirtute belli armatus promerui ut mihi omnis mortalis agere deceat gratias. Men. 1067 non edepol ita promeruisti de me ut pigeat quae uelis [opsequi]. Primary Sequence Dependent Upon the Perfect 13 Men. 1 1 00 promeruisti ut ne quid ores quod uelis quin impetres. Pers. 496 tuom promeritumst merito ut faciam. Poen. 1407 sq. quamquam ego te meruisse ut pereas scio, non experiar tecum. Although meruisse is in the perfect, it means "are worthy of," a present state, and thus takes the primary sequence. The passages having the secondary sequence, showing thereby that the speaker's attention is fixed upon the past act, are the following: Amph. 1 141 sq. tu cum Alcumena uxore antiquam in gratiam redi : hau promeruit quam ob rem uitio uorteres; Aul. 222 nam de te neque re neque uerbis merui uti facer es quod facis. Aul. 735 sq. quid ego demerui, adulescens, mali, quam ob rem ita facer es meque meosque perditum ires liberos? For demerui Goetz and Schoell read commend. Epid. 712 merui ut fierem. Men. 490 quid de te merui qua me caussa perderes? For other instances of the secondary sequence dependent on forms of merui, cf. Terence And. 281; Hec. 580. 2. In indirect questions. A sin. 36 modo pol percepi, Libane, quid istuc sit loci: 14 The Sequence of Tenses in Plautus Percept is equal to scio, a present state, and therefore takes the primary sequence. Bacch. 575 sq. nunc me ire iussit ad earn et percontarier utrum aurum reddat anne eat secum semul. Nunc .... iussit means "I have an order . . ," a present state caUing for the primary sequence. Capt. 46 sqq. sed inscientes sua sibi fallacia ita compararunt et confinxerunt dolum itaque hi commenti de sua sententia ut in seruitute hie ad suom maneat patrem: Commenti is equivalent to " they have this plan thought out," a present state, and therefore takes the primary sequence. Goetz and Schoell, and Morris following them, bracket 48, making maneat depend on 47. Capt. 466 sqq. neque ieiuniosiorem neque magis ecfertum fame uidi nee quoi minu' procedat quidquid facere occeperit, ita[que] uenter gutturque resident essurialis ferias. Vidi is equivalent to "I know," a present state calling for the primary sequence. Cure. 371 sq. Beatus uideor: subduxi ratiunculam, quantum aeris mihi sit quantumque alieni siet: Subduxi ... is equivalent to "my account is balanced," a present state, and therefore takes the primary sequence. Cure. 608 dixi equidem tibi unde ad me hie peruenerit. Dixi equidem tibi is the equivalent of scis, a present state, and therefore takes the primary sequence. See on M. G. 1097 sq., p. 15. Epid. 285 et repperi haec te qui apscedat suspicio. Primary Sequence Dependent Upon the Perfect 15 Repperi is the equivalent of scio, and thus takes the primary sequence. Men. 755 sed id quam mihi facile sit hau sum falsus. Hau sum falsus is equivalent to scio, a present state, and therefore takes the primary sequence. Merc. 253 sq. nisi capram illam suspicor iam me inuenisse quae sit aut quid uoluerit. The principal clause means "I suspect that I now know . . . ," a present state, and therefore takes the primary sequence. See on Capt. 569, p. 17; Epid. 355, p. 7; M. G. 767, p. 8. Merc. 344 neque is quom roget quid loquar cogitatumst. Cogitatumst is equivalent to scio, a present state, and demands the primary sequence. M. G. 867 modo intellexi quam rem mulier gesserit. Intellexi is equivalent to scio, a present state, and thus takes the primary sequence. M. G. 1097 sq. dixi equidem tibi quo id pacto fieri possit clementissume. Dixi .... is equivalent to "you know," a present state calling for the primary sequence. See on Cure. 608, p. 14. Pers. 81 sq. Omnem rem inueni, ut sua sibi pecunia hodie illam facial leno libertam suam. Inueni means "I know," and therefore takes the primary sequence. See on Epid. 355, p. 7; M. G. 767, p. 8. 1 6 The Sequence of Tenses in Plautus Poen. 547 scitis rem, narraui uobis quod uostra opera mi opu' siet. Narraui uobis is equivalent to scitis, and therefore demands the primary sequence. Poen. 557 sqq. itane? temptas an sciamus? non meminisse nos ratu's quo modo trecentos Philippos Collybisco uilico dederis, Ratu's .... is equivalent to "don't you believe that we remember how . . . ," a present state, and thus takes the primary sequence. Ratu's refers to the same time as temptas. Rud. 6ii sq. nunc quam ad rem dicam hoc attinere somnium numquam hodie quiui ad coniecturam euadere. The principal clause means "I am come to no conclusion," a present state calling for the primary sequence. Rud. 924 sq. nam ego nunc mihi, qui inpiger fui, repperi ut piger si uelim siem: Repperi is equivalent to scio, and therefore takes the primary sequence. Rud. 1026 mane, iam repperi quo pacto nee fur nee socius sies. See preceding example and Epid. 285, p. 14. True. 382 sq. sed quod ego facinus audiui adueniens tuom quod tu hie me apsente noui negoti gesseris? Audiui is equivalent to scio, a present state, and thus takes the primary sequence. 3. In relative clauses of characteristic. Some of these relative clauses in addition to the notion of characteristic also contain that of cause or of opposition, as will be noted in the proper places. The examples are: Primary Sequence Dependent Upon the Perfect 17 Amph. 176 sqq. satiust me queri illo modo seruitutem: hodie qui fuerim liber, eum nunc potuit pater seruitutis: Potuit .... means "has put into the power of slavery," a present state, and therefore fuerim in a relative clause of characteristic with the accessory notion of opposition, is in a primary tense. Asin. 85 sq. dotalem seruom Sauream uxor tua adduxit, quoi plus in manu sit quam tibi. Adduxit .... is equivalent to "has here . . . ." Libanus wishes to call attention to the existing state of affairs in Demaenetus's house, due to the former conduct of the old gentleman's wife. The primary sequence is therefore required. Capt. 568 sqq. TY. tu enim repertu's, Philocratem qui superes ueriuerbio. AR. pol ego ut rem uideo, tu inuentu's, uera uanitudine qui conuincas. Repertu's and inuentu's .... are equivalent to "you are known to be one who . . . ," denoting a present state, and therefore they take the primary sequence. See on Epid. 355, p. 7; Merc. 254, p. 8; If. G. 767, p. 15. Epid. 80 numquam hominem quemquam conueni unde abierim lubentius. Numquam .... conueni is equivalent to "I do not know," a present state, and therefore takes the primary se- quence. Men. 473-4 sq. pro di inmortales! quoi homini umquam uno die boni dedistis plus qui minu' sperauerit? Quoi .... dedistis is equivalent to quis habet . , . a present state, and therefore sperauerit, dependent on dedistis, in a clause of characteristic with the added notion of opposition, is in the primary sequence. 1 8 The Sequence of Tenses in Plautus 4. In consecutive clauses. The subjunctive in consecutive clauses is a development of the subjunctive of contingent futurity (Bennett Latin Language §360,6) where some condition is implied. Thus hie liber est talis ut quemvis iuuet originally meant: "this book is of such a nature as anyone you please would like" {i. e., if he should see it). The transition then was readily made to "this book is of such a nature that anyone you please likes it." The subjunctive in consecutive clauses thus takes the place of the indicative logically expected to express actuality. Because of this the tenses in these clauses are used with their indicative values, and to this extent show exceptions to sequence. The examples of the perfect indicative governing consecutive clauses with the primary sequence are: 'Amph. 815 quid ego feci qua istaec propter dicta dicantur mihi? Dicantur denotes a result continuing into the present. Feci is an aoristic perfect, and the primary tense dicantur therefore violates the rule of sequence. Bacch. 605 sqq. et tu, integumentum, uale. in eum [nunc] haec reuenit res locum, ut quid consili dem meo sodali super arnica nesciam, In eum locum denotes a present state, "affairs are in such a state." The sequence of nesciam is therefore regular. Capt. 410 sqq. nam tua opera et comitate et uirtute et sapientia fecisti ut redire liceat ad parentes denuo, quom apud hunc confessus es et genus et diuitias meas: The principal clause means "the arrangements effected through your .... are such," which is a present state. The primary sequence in liceat is therefore regular. Capt. 931 sqq. fecisti ut tibi, Philocrates, numquam referre gratiam possim satis, proinde ut tu promeritu's de me et filio. Primary Sequence Dependent Upon the Perfect 19 The present possim denotes a result continuing into the present and future. Fecisti is an aoristic perfect, and the primary tense possim thus violates the rule of sequence. Cas. 252 sq. sed quid ais? iam domuisti animum, potius ut quod uir uelit fieri id facias quam aduorsere contra? Iam domuisti animum means "is your temper by this time under control," which is a present state. The primary tense facias is therefore regular. Cist. 10 sq. ita in prandio nos lepide ac nitide accepisti apud te, ut semper meminerimus. The perfect meminerimus denotes a result continuing into the present and future. Accepisti is an aoristic perfect, and the primary tense meminerimus thus violates the rule of sequence. Men. 712 quid tandem admisi in me ut loqui non audeam? Audeam denotes a result continuing into the present. Admisi is an aoristic perfect, and the primary tense audeam thus violates the rule of sequence. M. G. 514 sq. ita sum coactus, Periplectomene, ut nesciam utrum me postulare priu' tecum aequiust — The principal clause means "I am in such straits," which is a present state. The primary sequence is therefore required in nesciam. Goetz and Schoell read 515 Vtrum me postulare tecum aequom siet. The different reading, however, does not affect the interpretation of the sequence. Poen. 553 sq. scimus rem omnem, quippe omnes simul dedicimus tecum una, ut respondere possimus tibi. Dedicimus is equivalent to "we know," a present state, and therefore takes the primary sequence. 20 The Sequence of Tenses in Plautus Cf. with the illustrations in the preceding section, the following ex- amples of the present subjunctive dependent upon the perfect indica- tive and expressing a result continuing into the present. Li\'y III. 68. 10 natura hoc ita comparatum est, ut qui apud multi- tudinem sua causa loquitur gratior eo sit, cuius mens nihil praeter pub- licum commodum videt. Livy X. 28. 12 datum hoc nostri generi est, ut luendis periculis pub- licis piacula simus: Tac. Ann. IV. 8. 8 ita nati estis, ut bona malaque vestra ad rem pub- licam pertineant. With the foregoing examples of the present subjunctive (or of its equivalent, the perfect, in the case of preteritives) denoting a result continuing into the present, we ought to compare the following group of examples of the imperfect dependent upon the perfect indicative, and denoting a result exclusively in the present. Since we expect the tenses in result clauses to have their indicative values (p. 18), and thus to form an exception to the principle of sequence, we should have expected the present and not the imperfect in these passages. See on the use of tenses in result clauses Lane Latin Grammar'^ §1757 (2); Gilder- sleeve-Lodge Latin Grammar §513; Allen and Greenough New Latin Grammar §485 c; Bennett Latin Grammar §268, 6, 7. The examples listed in the section below are therefore particu- larly significant as illustrations of mechanical conformity to the principle of sequence. Bacch. 1067 sqq. curatum est — esse te senem miserrumum. hoc est incepta efficere pulchre: ueluti mi euenit ut ouans praeda onustus cederem; Cederem, as is seen from the rest of Chrysalus' speech, denotes a result beginning in the present. We should therefore have expected cedam instead. The imperfect is due to mechanical sequence. Cas. 47 sqq. postquam ea adoleuit ad eam aetatem ut uiris placere posset, eam puellam hie senex amat ecflictim, et item contra filius. The whole context of the passage shows that posset denotes a result beginning in the present. Mechanical sequence caused the imperfect. Primary Sequence Dependent Upon the Perfect 21 Pers. 172 sq. nam equidem te iam sector quintum hunc annum, quom interea, credo, oui' si in ludum iret, potuisset iam fieri ut probe litteras sciret. Sciret in a substantive clause of result dependent on potuisset denotes a result exclusively present, as is shown by iam (173). The sequence is mechanical. Pseud. 13 1 8 sq. hoc ego numquam ratu' sum fore me ut tibi fierem supplex. The begging begins at the time of speaking. Hence we should have expected the present subjunctive. The sequence of fierem is mechanical. Since the subjunctive tenses in result clauses are used with their in- dicative values (p. 18), it follows that the perfect subjunctive, in the same manner as the perfect indicative, is both primary and secondary. Ex- amples of its use as a primary tense are naturally rare. When used as a secondary tense it represents a result merely as a fact without any reference to the continuance of the act. When the writer wishes to re- present the result as continuous he uses the imperfect subjunctive de- pendent upon secondary tenses. The examples of the perfect subjunctive as a secondary tense in result clauses are: Amph. 431 factumst illud, ut ego illic uini hirneam ebiberim meri. M. G. 262 sqq. nam ill' non potuit quin sermone suo aliquem familiarium participauerit de arnica eri, sese uidisse earn hie in proxumo osculantem cum alieno adulescentulo. Pers. 55 sq. nam numquam quisquam meorum maiorum fuit quin parasitando pauerint uentris suos: Pers. 582 sq. nam generi lenonio numquam ullus deu' tam benignus fuit qui fuerit propitius. Compare with the four preceding examples of result clauses in the per- fect subjunctive the following examples of the same usage in Caesar De Bella Gallico (A. T. Walker: The Sequence of Tenses in Latin, p. 40, Law- 22 The Sequence of Tenses in Plautus rence, Kansas, 1899): II. 21. 5 defuerit; III. 15. 4 pervenerit; V. 15. I ftierint and compulerint; V. 54. 4 fuerit; VII. 17. 3 caruerint. This last passage is particularly interesting for the reason that caruerint is coor- dinated with siistentarejit, the former stating a historical fact, the latter representing a resulting condition that continued for some time, namely, as long as the grain was scarce. On the matter of shift of tenses in result clauses dependent upon the same verb, see Gildersleeve-Lodge Latin Grammar §513, Note 2; Lane Latin Grammar'^ §1759- In the first three of the four passages on page 20 the perfect indicative should probably be considered aoristic, although Gildersleeve-Lodge Latin Grammar §513, Note i, say that "examples "^of the use of the aoristic perfect subjunctive after an aoristic perfect indicative > are not found in early Latin." 5. In conditional clauses of comparison. True. 292 sq. itane? erubuisti? quasi uero corpori reliqueris tuo potestatem coloris ulli capiendi, mala! Erubuisti, from the' inceptive erubesco, means "you are red," a present state, and therefore the primary sequence is required in reliqueris in the conditional clause of comparison. Ill Violations of the Sequence Principle I. Repraesentatio. As might be expected from the colloquial character of Plautus, this figure, which involves the retention of a primary tense where the sequence called for a secondary, is fairly common. We find different varieties of the figure, as noted in the discussions on the examples below. Atnph. 72 sqq. siue adeo aediles perfidiose quoi duint, sirempse legem iussit esse luppiter, quasi magistratum sibi alteriue amhiuerit. In the forms duint and amhiuerit we have repraesentatio. The use of this figure in this sentence enables the speaker to preserve the true character of the conditions. Amph. 115 sed ita adsimulauit se, quasi Amphitruo siet. Here, again, as in the preceding example, the use of reprae- sentatio in siet enables the speaker to preserve the character of the condition. Amph. 205 sqq. eos legat, Telobois iubet sententiam ut dicant suam: si sine ui sine bello uelint rapta et raptores tradere, si quae asportassent reddere, se exercitum extemplo domum redducturum, abituros agro Argiuos, pacem atque otium dare illis ; sin aliter sient animati neque dent quae petat, seseigitursummauiuirisque eorumoppidum oppugnassere. Iubet is an historical present. We have repraesentatio in dicant, uelint, sient, dent, and petat. The secondary tense asportassent shows that in 207 the speaker gives iuhet its true value. In sient, dent, and petat he passes back into the former manner of expression. 23 24 The Sequence of Tenses in Plautus Amph. 225 sq. conuenit, uicti utri sint eo proelio, urbem, agrum, aras, focos seque uti dederent. Exeunt and conloquontur (223, 224) show that conuenit is historical present. The primary tense uicti sint is an instance of repraesentatio. This is another passage in which the historical present is viewed first as a primary tense, then as a secondary, as is shown by the imperfect subjunctive dederent. Amph. 486 sqq. sed Alcumenai huius honoris gratia pater curauit uno ut fetu fieret, uno ut labore apsoluat aerumnas duas et ne in suspicione ponatur stupri et clandestina ut celetur consuetio. Apsoluat, ponatur, and celetur depend on fieret. The rule of sequence therefore called for the imperfect subjunctive in these forms. The presents are due to repraesentatio. Curauit denotes a present state, for the sense of the principal clause is "there is this provision . . . ." Instead of j^^re/ we should therefore have expected fiat. The sequence of fieret is mechanical. Goetz and Schoell read: Sed Alcumenae honoris huius gratia Pater curabit uno ut fetu proferat ******* Vno ut labore absoluat aerumnas duas, Et ne in suspicione ponatur stupri Et clandestina ut celetur consuetio. This reading makes absoluat, ponatur, and celetur depend on the primary proferat, and thus makes the sequence regular. The historical present is somewhat more commonly treated as primary, but in the following passages it is conceived according to its sense, that is, as a past, and therefore takes the secondary sequence. Amph. 207 asportassent dependent on uelint reddere, 206, Amph. 215 deducerent Amph. 226 dederent Aul. 319 liceret Bacch. 291 gereretur Bacch. 302 sciscerent Cas. 891 opprimeret respondent, 21^, conuenit, 225, infit, 318, sentio, 290, auferimus, 301, cupio, 891, Violations of the Sequence Principle 25 (editors read variously) Merc, giforet dependent on mittit, 90, M.C. Arg. II. 5 nuntiaret " " navigat, 5, M.G.Arg.U. 7 veniret " " scribit, 7, Pers. Arg. 3 emeret " " suadet, 4, (Goetz and Schoell have emere) Pers. $2)^ facer em dependent on facto, 537, Rud. 602 darem " " uidetur, 601, Trin. 14 aleret " " uideo, 14. In Stick. 366 sq. dum percontor portitores, ecquae nauis uenerit ex Asia, negant uenisse, the historical present indicative takes the perfect subjunctive as a pri- mary tense. Amph. 745 sq. quipp' qui ex te audiui, ut urbem maxumam expugnauisses regemque Pterelam tute occideris. As is seen from 203-247, 412-415, the storming of the city preceded the slaying of king Pterela. Occidisses would naturally mean that the storming and the killing had been synchronous. Occideris is an instance of repraesentatio used of necessity. What the speaker wishes to indicate is that the act of occideris was anterior to audiui but later than oppugnauisses. Cf. with this example Professor R. G. Kent's comment {The Classical Weekly 7. 77) on Caesar B. G. I. 40. 7 Denique hos esse eosdem quibuscum saepe numero Helvetii congressi non solum in suis, sed etiam in illorum finibus plerummque superassent, qui tamen pares esse nostro exercitui non potuerint. Amph. 1 122 sq. is se dixit cum Alcumena clam consuetum cubitibus, eumque filium suom esse qui illas anguis uicerit; Dixit is historical perfect. We should therefore have expected uicisset by the regular rule of sequence. Vicerit is an example of repraesentatio. A sin. 442 sq. aibat reddere quom extemplo redditum esset; nam retineri, ut quod sit sibi operis locatum ecficeret. Quod .... locatum depends on the secondary ecficeret. The sequence rule therefore calls for esset locatum. The perfect sit locatum is due to repraesentatio. 26 The Sequence of Tenses in Plautus AuL 550 pol ego te ut accusem merito meditabar. Inasmuch as accusem precedes meditabar, its tense was prob- ably determined in the speaker's mind before that of the principal verb. On the other hand the force of meditabar may readily extend into the present. We find passages where the transition from parataxis to hypotaxis is going on, and under such con- ditions often the tense of the subjunctive verb in the independent stage is responsible for exceptions in the sequence. The position of the subjunctive clause with respect to the principal clause is a factor to be reckoned with in sentences of this type where there are violations in the sequence. Furthermore the connection between the dependent and the independent clause is occasionally very loose and thus results in irregularities in the sequence. Compare, for example, Terence Phor. 933 sq.: ut filius cum ilia habitet apud te: hoc nostrum consilium fuit. Here, inasmuch as the purpose clause precedes the principal clause, we probably have a mild instance of anacoluthon. Accusem may be plain repraesentatio, but it is possible that its tense is due to one of the other factors mentioned. Cf. Ennius Ann. I. 86 sq. (Vahlen") sic exspectabat populus, atque ora tenebat rebus, utri magni victoria sit data regni. Repraesentatio caused the primary tense sit data. Bacch. 287 occepi ego opseruare eos quam rem gerant. Occepi opseruare called for the secondary sequence. The primary gerant is due to repraesentatio. Bacch. 351 sqq. ut amantem erilem copem facerem filium, ita feci ut auri quantum uellet sumeret, quantum autem lubeat reddere ut reddat patri. The rule of sequence called for the secondary redderet instead of reddat. The primary tense is due to repraesentatio. See p. 5. Violations of the Sequence Principle 27 Bacch. 589 sqq, me misit miles ad eam Cleomachus, uel ut ducentos Philippos reddat aureos uel ut hinc in Elatiam hodie eat secum semul. We should have expected redderet and iret in the purpose clauses dependent on misit. The presents are due to repraesen- tatio. Bacch. 689 sq. MN. ego patrem exoraui. CH. nempe ergo hoc ut faceret quod loquor? MN. immo tibi ne noceat neu quid ob eam rem suscenseat; Noceat and suscenseat depend on exoraui. The rule of sequence called for the imperfect subjunctive instead of the present. The present is due to repraesentatio. Capt. 27 sqq. coepit captiuos commercari hie Aleos, si quem reperire posset qui mutet suom, ilium captiuom : The rule of sequence called for the secondary mutaret depen- dent upon posset. The present mutet is due to repraesentatio. Capt. 1002 sqq. quasi patriciis pueris aut monerulae aut anites aut coturnices dantur, quicum lusitent itidem haec mihi aduenienti upupa qui me delectem datast. The act of delectem is fully past at the time these verses are spoken. The rule of sequence therefore called for delectarem. The present is due to repraesentatio. Cas. 680 sqq. id hue missa sum tibi ut dicerem, ab ea uti caueas tibi. Missa sum means "I am here," a present state, and the sec- ondary dicerem is therefore due to a mechanical adherence to the principle of sequence. Caueas dependent on dicerem violates the rule of sequence. In relation to the time of speaking, caueas 28 The Sequence of Tenses in Plautus represents a future idea, and dicerem a present idea. The rule of sequence of tenses would throw both these verbs into the imperfect subjunctive; but by the use of repraesentatio in caueas, the time distinction is preserved. B^ however, has dicam instead of dicerem. Caueas may be due to parataxis, but this seems less likely than the explanation given. Cist. 1 68 sq. ill' clam opseruauit seruos quo aut quas in aedis haec puellam deferat. The act of deferat is completely past at the time these verses are spoken. The sequence called for deferret. The present is due to repraesentatio. It is to be noted that this and the follow- ing example, together with M. G. 131 deferat, violate the theory of non-sequence. Cist. 565 sqq. immo meretrix fuit; sed ut sit, de ea re eloquar. iam perducebam illam ad me suadela mea: anus ei amplexa est genua plorans, opsecrans ne deserat se: The act of deserat is fully past at the time of speaking. The rule of sequence therefore called for the imperfect subjunctive instead of the present. The present is due to repraesentatio. Cure. 558 sqq. postquam rem diuinam feci, uenit in mentem mihi, ne trapezita exulatum ahierit, argentum ut petam, ut ego potius comedim quam ille. The acts of the dependent subjunctives in this passage are fully past at the time of speaking. The rule of sequence there- fore called for the secondary tenses. The primary tenses are due to repraesentatio. Epid. 414 sqq. te pro filio facturum dixit rem esse diuinam domij quia Thebis saluos redierit. Violations of the Sequence Principle 29 The act of redierit is fully past at the time of speaking. Hence the rule of sequence called for redisset. The primary tense redierit is due to repraesentatio. This example illustrates the commonest form of repraesentatio — that in subordinate clauses in indirect discourse. Cf. Amph. 72 sqq., p. 23; Amph. 209, p. 23; Amph. 1 123, p. 25; Asin. 443, p. 25; Merc. 419, p. 29; Most. 1 124, p. 30; Pseud. 597, p. 30. Men. 453 sq. non ad earn rem otiosos homines decuit deligi, qui nisi adsint quom citentur, census capiat ilico? The text of these lines is bad. If we accept the reading here given, the act of capiat is past at the time of speaking. The rule of sequence called for the imperfect subjunctive instead. The present is due to repraesentatio. Men. 1056 sq. quom argentum dixi me petere et uasa, tu quantum potest praecucurristi obuiam, ut quae fecisti infitias eas. By the rule of sequence we ought to have ires instead of eas. The present subjunctive is due to repraesentatio. H. N. Fowler (edition of Men., p. 173) notes that Brix thought that metrical reasons caused the substitution of eas for ires. Merc. 419 dixit se redhibere si non placeat. The act of placeat is completely past at the time of speaking The rule of sequence called for placeret. The present is due to repraesentatio. M. G. 129 sqq. ego quoniam inspexi mulieris sententiam, cepi tabellas, consignaui, clanculum dedi mercatori quoidam qui ad ilium deferat meum erum, qui Athenis fuerat, qui hanc amauerat, ut is hue ueniret. Dedi is an aoristic perfect. We should therefore by the rule of sequence have had deferret instead of deferat. The present subjunctive is due to repraesentatio. Veniret has the sequence 30 The Sequence of Tenses in Plautus we expect. Deferat violates the non-sequence rule, which would call for deferret. Cf. Cist. 169 deferat, p. 28; Cist. 568 deserat, p. 28. M. G. 962 sq. uah ! egone ut ad te ab libertina esse auderem internuntius, qui ingenuis sati' responsare nequeas quae cupiunt tui? Nequeas dependent on auderem violates the rule of sequence. The present subjunctive is due to repraesentatio. Most. 1 122 sqq. Vbi somno sepeliui omnem atque obdormiui crapulam, Philolaches uenisse mihi suom peregrehuc patrem quoque modo hominem ad seruos ludificatu' sit, ait se metuere in conspecocedere. The rule of sequence called for ludificatus esset. The primary tense ludificatu' sit is due to repraesentatio. However, the text of this passage is very bad. Poen. 600 sqq. scilicet, et quidem quasi tu nobiscum adueniens hodie oraueris liberum ut commostraremus tibi locum et uoluptarium ubi ames, potes, pergraecere. Ames, potes, and pergraecere dependent on the secondary commostraremus violate the rule of sequence which called for the imperfect subjunctive in this clause. The present sub- junctives are due to repraesentatio. Poen. 1056 sq. qui potuit fieri uti Carthagini gnatus sis? Potuit fieri is an aoristic perfect. The rule of sequence there- fore called for gnatus esset instead of gnatus sis. The primary tense is due to repraesentatio. Pseud. 596 sq. ut ego oculis rationem capio quam mi ita dixit eru' meu' miles, septumas esse aedis a porta ubi ille habitet leno . . . Violations of the Sequence Principle 31 Dixit is an aoristic perfect. The rule of sequence therefore called for hahitaret instead of habitet. The present subjunctive is due to repraesentatio. Pseud. 795 sq. quin ob earn rem Orcus recipere ad se hunc noluit, ut esset hie qui mortuis cenam coquat; Coquat dependent on the secondary tense esset violates the rule of sequence, which called for the imperfect subjunctive in both verbs. The present subjunctive is due to repraesentatio. Rud. 124 sqq. PL, die quod te rogo, ecquem tu hie hominem crispum, incanum uideris, malum, peiiurum, palpatorem — DA. plurumos, nam ego propter eiusmodi uiros uiuo miser. PL. hie dico, in fanum Veneris qui mulierculas duas secum adduxit, quique adornaret sibi ut rem diuinam /aaa/, aut hodie aut heri. Faciat dependent on the secondary tense adornaret violates the rule of sequence, which called for the imperfect subjunctive in both verbs. The present faciat is due to repraesentatio. Adornaret depends on ecquem tu uideris implied from 125. The secondary sequence is required in this clause even though it depends on uideris which in turn depends on the primary tense dico, for uideris refers to the past. Peculiar, however, is the use of the subjunctive adornaret side by side with the indicative adduxit. Rud. 408 sqq. ut lepide, ut liberaliter, ut honeste atque hau grauate timidas, egentis, uuidas, eiectas, exanimatas accepit ad sese, hau secus quam si ex se simus natae! In this sentence the use of repraesentatio enables the speaker to preserve the character of the conditional clause of comparison. Adherence to the sequence rule would have resulted in essemus natae, which would have been identical with the tense of the contrary to fact condition. 32 The Sequence of Tenses in Plautus 2. Violations caused by other factors. Cure. 225 sqq. paues parasitus quia non rediit Caria: adferre argentum credo; nam si non fer at, tormento non retineri potuit ferreo quin reciperet se hue essum ad praesepem suam. Ferat stands in a future less vivid condition dependent on potuit. The perfect indicative potuit is irregular, because the verb really refers to the future. We find the perfect subjunctive occasionally in the apodosis of future less vivid conditional sentences, to emphasize the certainty of the conclusion. So Cas. 424 sqq. Because potuit is a verb of possibility, the indica- tive here may be accounted for by the fact that it is the regular mood used with such verbs in the apodosis of unreal conditions. The speaker's mind may be shifting between these two types of conditions. See Gildersleeve-Lodge Latin Grammar §597, Rem. 3 (a). It may well be that the irregularity in the sequence oi ferat is due to the fact that at the time the conditional clause was spoken the speaker had not yet formulated in his mind the verb of the conclusion. If this was the case, we have here a slight anacoluthon. Amph. 815 dicantur See p. 18. Capt. 932 possim See p. 18. Cist. II meminerimus See p. 19. Men. 712 audeam See p. 19. Rud. 217 sq. leibera ego prognata fui maxume, nequiquam fui. nunc qui minu' seruio quasi serua forem nata? V. 217 shows that the speaker has the notion of contrary to fact in mind. The use of the sequent form would have obscured Violations of the Sequence Principle 33 what she meant to say. The violation of sequence is due to the limited power of the tenses. Trin. 991 sq, immo, saluos quandoquidem aduenis — di te perdant, si te flocci facio an periisses prius. We should have expected Sy. to say di te perdant, si te flocci facerem an periisses prius, but for vividness facio is substituted for facerem. What the speaker wishes to say is not . . . "if I should care a straw whether you had been hanged before" (yougot here), but . . . "if I do care . . ." We cannot be absolutely certain as to the exact order in which the clauses of this sentence were formulated in the speaker's mind. The sentence may be regarded as wavering between parataxis and full hypotaxis, the order of thought being "would that you had been hanged before, I'll be hanged if I care a straw whether this had happened to you." The intensity of the thought would naturally tend to check the hypotactic process. Compare with this passage Cic. ad .4/^. VIII. 6. 4 moriar, si magis gauderem, si id mihi accidisset, where the thought runs "I would not (now) more rejoice, if this had happened to me, upon my life I wouldn't." Cicero had to retain the secondary gauderem and accidisset in violation of the sequence principle, for had he made these verbs conform to sequence and written gaudeam and acci- derit the contrary to fact notion would have been sacrificed and that of future less vivid come in to take its place. But that would have meant something quite different from what was in his mind. The non-sequence was necessary. If the sentence be regarded as fully hypotactic, we may explain the tense of periisses as being due to the fact that its association with the more common imperfect in this type of sentence, caused the speaker to disregard the fact that here he was making it depend on the present facio. Cf. Horace Serm. I. 9. 45 sqq. haberes magnum adiutorem, posset qui ferre secundas, hunc hominem velles si tradere; dispeream, ni summosses omnes. IV Apparent Violations of the Sequence Principle I. In connection with the phrase quod sciam. The phrase quod sciam, "as far as I know," is found several times in connection with the perfect tense, seemingly in violation of the principle of sequence. This, however, is a stereotyped expression and sciam does not depend on the perfect with which it stands but on est to be supplied in thought. Thus, for ex- ample, in the first illustration below, the full thought is "indeed it never happened in my presence, {lit.) so far as is that which I know {quoad id est quod sciam), so far as my knowledge goes." Sciam thus stands in a relative clause of characteristic dependent on an implied est, and the sequence really is not violated. The construction is found in: Amph. 749 mequidem praesente numquam factumst, quod sciam. Capt. 172 sq. sed num quo foras ^ uocatus ad cenam? HE. nusquam, quod sciam. Epid. 638 EP. non me nouisti? TE. quod quidem nunc ueniat in mentem mihi. Veniat .... mihi is equivalent to sciam. Men. 500 sq. non edepol ego te quod sciam umquam ante hunc diem uidi neque gnoui; True. 199 nam equidem illi uterum, quod sciam, numquam extumere sensi. In clauses dependent on noui Plautus always uses the primary sequence. The following are the examples. 34 Apparent Violations of the Sequence Principle 35 M. G. 451 sq. ego istam domum neque moror neque uos qui homines sitis noui neque scio. M. G. 924 sq. numquam uidit: qui nouerit me quis ego sim^ Most. 969 scio qua me eire oportet et quo uenerim noui locum. Trin. 283 noui ego hoc saeculum moribus quibu', siet: 2. Other apparent violations. A sin. 7 sciretis See p. 40. Merc. 633 EV. quid ego facerem? CH. quid tu faceres? men rogas? Faceres is a past deliberative retained in its paratactic form and not subordinated to rogas. Inasmuch as faceres precedes rogas it seems better to explain the passage on the basis of parataxis than on that of hypotaxis. Professor W. G. Hale (v4. /. P. VIII. 61, footnote) mentions this form as "an exception to the Law of Sequence." However, even if the subjunctive clause follows the indicative, it is not safe to assume in sentences of this type that the hypotaxis is complete. Cf. Cic. In Vatin. 2. 5 sed quaero a te cur C. Cornelium non defenderem. In this sentence it seems better to understand the order of thought to be "but, I ask you, why was I not to defend Gains Cornelius?" Most. 157 sq. lam pridem ecastor frigida non laui magi' lubenter nee quom me melius, mea Scapha, rear esse deficatam. There is a mild anacoluthon here. Verse 158 continues as if numquam fuit quom lauerim had gone before. Rear is logically parenthetical. The sentence may have begun in the form: nee fuit (or laui) quom, ut reor, melius deficata essem. Or instead of reor the parenthetical verb may have been rear, a so-called deliberative subjunctive. Then too, instead of essem the form 36 The Sequence of Tenses in Plautus may have been fuerim, as we see by comparing Terence Haul. 1025. If the parenthetical verb was in the indicative, it was thrown into the subjunctive upon being subordinated to the impUed fuU (or laui) , because the quom-clause denotes character- istic. Cf. qtiod sciam, p. 34. The present subjunctive was required in the gwow-clause because either /m^/ or laui here denotes time extending into the present. Most. 89 arbitrarer, which is in origin the same kind of subjunctive, depends on cogitaui and follows sequence. Most. 1004 sq. promisi foras, ad cenam ne me te uocare censeas. Censeas is not in subordination to promisi, but is a separate sentence in the process of becoming a parenthetical clause of purpose. Pers. 116 sqq. iam heri narraui tibi tecumque oraui ut nummos sescentos mihi dares utendos mutuos. SAT. memini et scio et te me orare et mihi non esse quod darem. As is seen from heri (116), orare and esse are historical presents and govern the secondary sequence. Rud. 379 si amabat, rogas, quid faceret? See note on Merc. 633, p. 35. Stich. 255 dares See p. 43. True. 292 reliqueris See p. 22. V Instances of Mechanical Conformity to the Sequence Principle In this chapter are treated the dependent subjunctives which follow the rules of sequence in disregard of the claims of absolute time. The first group contains the forms dependent upon a secondary tense and denoting time present or future relatively to that of speaking, but thrown into the secondary sequence because they are dependent upon a secondary tense which denotes present time (e. g., Asin. 589), or else they are by accessory circumstances thrown into dependence upon a secondary tense {e. g., Most. 183). Asin. 588 sqq, LE. attatae, modo hercle in mentem uenit, nimi' uellem habere perticam. LI. quoi rei? LE. qui uerberarem asinos, si forte occeperint clamare hinc ex crumina. Vellem is a potential subjunctive referring to the present. Verberarem in a purpose clause depending on uellem therefore denotes future time. In spite of this fact uerberarem is secondary. The sequence is therefore mechanical. Asin. 674 sq. nimi' bella es atque amabilis, et si hoc meum esset, hodie numquam me orares quin darem: Orares in a contrary to fact conclusion denotes present time, and darem in the substantive clause dependent on orares must refer to time future to orares. If the dependent subjunctive had independent tense value, we should have dem and not darem. Bacch. 553 sqq. PI. opsecro hercle loquere, quis is est. MN. beneuo- lens uiuit tibi. nam ni ita esset, tecum orarem ut ei quod posses mali isicexe faceres . 37 38 The Sequence of Tenses in Plautus Orarem is the condition of a present contrary to fact conditional sentence. Inasmuch as faceres stands in a volitive clause dependent on orarem, it denotes time future to that at which Mnesilochus speaks. Facias would have been required if the dependent subjunctives expressed time independently of the leading verb. Bacch. 1069 cederem See p. 20. Cas. 48 posset See p. 20. Most. 182 sq. SC. ita tu me ames, ita Philolaches tuo' te amet, ut uenusta es. PHILOL. quid ais, scelesta? quo modo adiurasti? ita ego istam amarem? Ita .... amarem, which is the reported form of ita Philolaches tuos te amet, a wish for something in the future, is thrown into the secondary sequence, because it is made to depend on adiurasti. Inasmuch therefore as the wish must refer to the future as regards the time of speaking, nothing but an arbi- trary adherence to a mechanical sequence could cause the speaker to shift the tense of the subjunctive from amet to amaret upon making the subordination. Pers. 173 sciret See p. 21. Poen. 681 sq. CO. uidere equidem uos uellem quom huic aurum darem. ADV. illinc procul nos istuc inspectabimus. Vellem is a potential subjunctive in the present. The tem- poral clause dependent on uellem refers therefore to the future, although we actually have the imperfect, which is due to me- chanical sequence. Darem cannot possibly refer to the past. Note also inspectabimus, which is synchronous with darem. Poen. 1066 patrem atque matrem uiuerent uellem tibi. Mechanical Conformity to the Sequence Principle 39 Vellem is another potential subjunctive referring to the pres- ent. Again uiuerent is mechanically secondary. Poen. 1 25 1 sq. primum, si id fieri possit, ne indigna indignis di darent, id ego euenisset uellem; The text of this passage is uncertain with regard to possit and euenisset. Goetz and Schoell note that the reading of A is POSSI(Me/ E)T. Lindsay gives posset for A. For euenisset Goetz and Schoell note that all the MSS except A have euenire. If the reading posset be accepted, the 5z-clause forms the protasis of a present contrary to fact conditional sentence. In that case darent dependent on posset is an instance of mechanical sequence. Vellem must refer to the present, and we should therefore on the theory that the dependent subjunctives denote absolute time, have expected eueniret and not euenisset. The present subjunctive in a present contrary to the fact protasis is used in Plautus in A sin. 188 habeas; 393 sit; Aul. 523 metiiam; Bacch. 635 sit; Epid. 331 haheam; Most. 555 sit (text corrupt here); Pers. 215 sim; Pseud. 274 possim; also Terence And. 310 sis. See on this point Bennett Syntax of Early Latin, I. pp. 273-274; Lane Latin Grammar'^ §2075; Allen & Greenough New Latin Grammar §517 e. and n. i and 2; Gildersleeve-Lodge Latin Grammar §596. 2, Possit, if accepted as the true reading above, may be regarded as future less vivid. , Pseud. 12,19 fierem See p. 21. True. 138 sqq. ego expedibo. rem perdidi apud uos, uos meum negotium apstulistis. si rem seruassem, fuit ubi negotiosus essem. The perfect indicative fuit is here used instead of the plu- perfect subjunctive in the apodosis of a past contrary to fact conditional sentence to denote what was likely or certain to happen. Essem dependent on fuit is put in the imperfect by mechanical sequence, for the context shows that negotiosus essem means "where I might now be employed." 40 The Sequence of Tenses in Plautus In the following section are grouped the imperfect subjunc- tives dependent upon secondary tenses, but denoting acts distinctly future from the standpoint of the speaker, and thus taking the secondary sequence mechanically and in disregard of the claims of absolute time. Amph. 85 conciderent See p. 45. Amph. 464 sq. amoui a foribus maxumam molestiam, patri ut liceret tuto illam amplexarier. The act of liceret is clearly future relatively to the speaking, and the secondary is therefore mechanical. Amph. 487 fieret See p. 24, A sin. 6 sqq. nunc quid processerim hue et quid mi uoluerim dicam: ut sciretis nomen huiius fabulae; nam quod ad argumentum attinet, sane breuest. nunc quod me dixi uelle nobis dicere dicam: huic nomen graece Onagost fabulae; Sciretis depends on processi to be supplied in thought from processerim, and is mechanically secondary, for the name of the play is not given until 10. Asin. 336 sq. em ergo is argentum hue remisit quod daretur Saureae pro asinis. Inasmuch as the money for the asses is still to be paid at the time of speaking, the present subjunctive was to be expected instead of daretur. The sequence used is mechanical. Asin. 929 iam surrupuisti pallam quam scorto dares? The act of dares would be manifestly future to the time of speaking. The secondary sequence is therefore used mechan- ically, where the primary was to be expected. Mechanical Conformity to the Sequence Principle 41 Bacch. 645 sqq. nunc amanti ero filio senis, quicum ego bibo, quicum edo et amo, regias copias aureasque optuli, ut domo sumeret neu foris quaereret. The acts of sumeret and of quaereret are manifestly future to the time of speaking. The sequence in these verbs is therefore mechanically secondary. Cas. 52 posceret See p. 6. Cas. 604 quin eapse me adlegauit qui istam arcesserem. The act of arcesserem is distinctly future to the time of speak- ing. The secondary sequence is therefore mechanical. M. G. 138 sqq. itaque ego paraui hie intus magnas machinas qui amantis una inter se facerem conuenas. nam unum conclaue, concubinae quod dedit miles, quo nemo nisi eapse inferret pedem, in eo conclaui ego perfodi parietem qua commeatus clam esset hinc hue mulieri; The imperfects facerem and esset denote the relation to the past verbs paraui and perfodi respectively, though the ideas which they denote are definitely thought of as future to the time of speaking. The secondary sequence is therefore mechanical. M. G. 956 nam hoc negoti clandestino ut agerem mandatumst mihi. The act of agerem is clearly future to the time when these words were spoken. The secondary sequence is therefore mechanical. M. G. 1 158 id nos ad te, si quid uelles, uenimus. The idea of uelles is clearly future to the time of speaking. The secondary sequence is therefore mechanical. Cf. Amph. 870 42 The Sequence of Tenses in Plautus feram; and Merc. 669 persequar, where ueni and reueni take the primary sequence though the force of those verbs is not appreciably different from that of uenimus here. Cf. further Poen. 1276 redirent and Caesar B. G. IV. i. 10 Atque in earn consuetudinem adduxerunt, ut locis frigidissimis neque vestitus praeter pellis haberent quicquam, quarum propter exiguitatem magna est corporis pars aperta, et lavarentur in fluminibus. Haberent and laverentur in result clauses dependent upon addux- erunt are indisputably present, as is shown by est, and the im- perfect subjunctive is due to mechanical sequence. Most. 420 sqq. iussit maxumo opere orare ut patrem aliquo apsterreres modo ne intro iret ad se. The imperfects are used to denote the relation to the iussit, though the acts denoted by the subjunctives are definitely thought of as future to the time of speaking. The secondary sequence is therefore mechanical. Most. 714 sqq. tempus nunc est senem hunc adloqui mihi. hoc habet! repperi qui senem ducerem, quo dolo a me dolorem procul pellerem. accedam. The acts of ducerem and pellerem are thought of as distinctly future to the time of speaking. The secondary sequence is therefore mechanical. Perfect forms of reperio taking the primary sequence are found as follows: Capt. 568 superes; Epid. 285 apscedat; Epid. 445 explices; Rud. 1026 sies. True. 81 daret depends on the aoristic perfect. For the sequence of verbs governed by perfect forms of inuenio, which have approximately the same meaning as the perfect forms of reperio, see pp. 7, 8, 15, and 17. Most. 1 126 sq. nunc ego de sodalitate solus sum orator datus qui a patre eiius conciliarem pacem. Mechanical Conformity to the Sequence Principle 43 The act of conciliarem is thought of as distinctly future to the time of speaking. The secondary sequence is therefore mechani- cal. Pers. 304 sq. quae dixi ut nuntiares, satin ea tenes? The act of nuntiares is distinctly thought of as future to the time when these words are spoken. The sequence is therefore mechanical. Pseud. 55 sqq. ea caussa miles hie reliquit symbolum, expressam in cera ex anulo suam imaginem, ut qui hue adferret eiius similem symbolum cum eo simul me mitteret. ei rei dies haec praestituta est, proxuma Dionysia'. CALL eras ea quidem sunt: The imperfect subjunctives are used to denote the relation to reliquit, though the acts which they denote are thought of as definitely future to the time of speaking. The secondary se- quence is therefore mechanical. Pseud. 1233 qui ilium ad med hodie adlegauit mulierem qui abduceret. The act of abduceret is manifestly future to the time of speak- ing. The secondary sequence is therefore mechanical. Stich. 247 sqq. CR. Panegyris rogare iussit ted ut opere maxumo mecum simitu ut ires ad sese domum. GE. ego illo mehercle uero eo quantum potest. iamne exta cocta sunt? quot agnis fecerat? CR. ilia quidem nullum sacruficauit. GE. quo modo? quid igitur me uolt? CR. tritici modios decem rogare, opinor, te uolt. GE. mene, ut ab se[se] petam? CR. immo ut a uobis mutuom nobis dares. The act of ires is manifestly future to the time of speaking. The sequence is therefore mechanical. Dares depends on iussit 44 The Sequence of Tenses in Plautus me rogare carried along in thought from w. 248-9, and the sequence is thus normal. lussit me rogare denotes a request to be made on Gelasimus manifestly after the speaking of v. 255. Hence the secondary tense dares is due to mechanical sequence. Trin. 442 sqq. meu' gnatus me ad te misit, inter te atque nos adfinitatem ut conciliarem et gratiam. tuam uolt sororem ducere uxorem ; et mihi sententia eademst et uolo. The act of conciliarem is thought of as definitely future to the time of speaking. The imperfect is used to denote the relation to misit, and the sequence is therefore mechanical. VI Shift in the Sequence By virtue of the fact that the perfect indicative may be either primary or secondary, it sometimes happens that one and the same verb is regarded in these two senses in the same sentence, and the sequence changes to suit the particular viewpoint that the speaker or writer has for the moment. Amph. 64 sqq. nunc hoc me orare a nobis iussit luppiter ut conquistores singula in subsellia eant per totam caueam spectatoribus, si quoi fauitores delegatos uiderint, ut is in cauea pignus capiantur togae; siue qui ambissent palmam histrionibus seu quoiquam artifici (seu per scriptas litteras seu qui ipse amhisset seu per internuntium), siue adeo aediles perfidiose quoi duint, sirempse legem iussit esse luppiter, quasi magistratum sibi alteriue amhiuerit. Eant and capiantur denote the present state resulting from Jupiter's act, whereas ambissent and amhisset indicate that the speaker's point of view has shifted and that he now thinks of the actual ordering. Duint and amhiuerit are primary by repraesentatio. In 81 sqq.: hoc quoque etiam mihi in mandatis dedit ut conquistores fierent histrionibus: qui sibi mandasset delegati ut plauderent quiue quo placeret alter fecisset minus, eius ornamenta et corium uti conciderent. the repraesentatio has been dropped and the speaker has the past act in mind. Moreover, conciderent is slavishly secondary, for the actual punishing of the offenders would take place after the speaking of these verses, that is, at the end of the performance. Amph. 205 sqq. See p. 23. Amph. 225 sqq. See p. 24. 45 VII Conclusion In summarizing this investigation we must bear in mind that the large majority of dependent subjunctives can readily be explained according to either theory of tense usage which one happens to prefer, that of sequence or that of non-sequence. What conclusions then are we to draw from a study of the excep- tions to the two respective theories? In so far as the exceptions to the doctrine of sequence are concerned, it is to be noted that if we exclude repraesentatio, of which there are forty-six instances pp. 23-31, there are but seven examples (four of which are in result clauses) remaining (pp. 32-33), all of which can be explained on logical grounds. A few exceptions, moreover, would, in the nature of things, be expected. As to the exceptions to the doctrine that the dependent sub- junctives denote absolute time and not time relative to the verb on which they depend, we note that their number is much larger, thirty-eight examples (pp. 37-44), which are arranged in two groups of eleven and twenty-seven respectively. In the ex- amples of the first group nothing but a mechanical application of the rule of sequence can have caused the secondary subjunc- tives where the claims of absolute time called for a primary tense. The examples of the second group (pp. 40-44) will not all seem equally convincing to all readers, but when studied as a group they give conclusive evidence of a tendency to use the imperfect tense of the subjunctive to denote an act which from the standpoint of the speaker is distinctly and definitely future. Naturally this tendency can be seen in other but less cogent examples. How strongly the feeling for the secondary imperfect dependent upon the perfect was established may be observed in the examples of clauses in the imperfect subjunctive denoting a result lying exclusively in the present (pp. 20-21). Again, in the case of the secondary forms dependent upon the imperfect subjunctive in contrary to fact conditions and in potential subjunctives (pp. 37-39), where the requirements of absolute time called for a primary tense, to explain the tense by saying 46 Conclusion 47 that "the modal feeling in the speaker's mind which expresses itself in the main sentence is, in the nature of things, very likely to continue in the speaker's mind in the subordinated sentence or sentences, either quite unchanged or but slightly shaded" (W. G. Hale, ^. /. P. VIII. 54), is to grant that there is such an influence as we commonly call the sequence of tenses. BIBLIOGRAPHY The following list includes the works to which reference was made more or less constantly. Works less freely consulted are not listed here, although the writer made use of numerous editions of the plays of Plautus other than the ones mentioned below, as well as of all the standard Latin grammars in English, in French, and in German. Allen and Greenough, New Latin Grammar, New York, 1903. Auden, H. W., Pseudolus of Plautus, Cambridge, 1896. Bennett, C. E., Latin Grammar, Boston, 1908. Bennett, C. E., The Latin Language, Boston, 1907. Bennett, C. E., Syntax of Early Latin, The Verb, Boston, 1910. Brix, Julius, Ausgewdhlte Komodien des T. M. Plautus, Leipzig, 1873. Draeger, A., Historische Syntax der Lateinischen Sprache, Leipzig, 1874. Durham, C. L., Subjunctive Substantive Clauses in Plautus, Cornell Studies in Classical Philology, XIIL Elmer, H. C, T. Macci Plauti Captivi, Boston, 1900. Fennell, C. A., T. Macci Plauti Stichus, Cambridge, 1893. Fowler, H. N., The Menaechmi of Plautus, Boston, ^910. Gildersleeve, B. L., The Sequence of Tenses in Latin, American Journal of Philology, VI n, 228-231, 1887. Gildersleeve-Lodge, Latin Grammar, third edition. New York, 1905. Gray, J. H., T. Macci Plauti Asinaria, Cambridge, 1894. Gray, J. H., T. Macci Plauti Epidicus^ Cambridge, 1893. Hale, W. G., The Sequence of Tenses in Latin, American Journal of Philology, VIL 446-465, 1886; VIIL 46-77, 1887; IX. 158-177, 1888. Hale-Buck, Latin Grammar, Boston, 1903. Holtze, Fr. W., Syntaxis Priscorum Scriptorum Latinorum, Leipzig, 1861. Kent, R. G., Some Tense Sequences in Caesar, De Bella Gallico, The Classical Weekly 7. 77-78, 1913. Kent, R. G., The "Passing" of the Sequence of Tenses, The Classical Weekly, 9. 2-7; 9-13, 1915. Kiihner, Raphael, A usfuhrlicheGrammatik der Lateinischen Sprache, Hannover, 1912. Lane, G. M., Latin Grammar for Schools and Colleges (revised by Morgan), New York, 1903. Lindsay, W. M., T. Macci Plauti Comoediae, Oxford, 1903, 1910. Lindsay, W. M., Syntax of Plautus, Oxford, 1907. Loewe, Goetz, Schoell, T. Macci Plauti Comoediae, ed. mai., Leipzig, 1884. Lorenz, Aug. O. Fr., Ausgewdhlte Komodien des T. M. Plautus, Berlin, 1866, 1876. Morris, E. P., Captives and Trinummus of Plautus, New York, 1898. Morris, E. P., Pseudolus of Plautus, Boston, 1895. 48 Bibliography 49 Palmer, Arthur, T. Macci Plauti Amphitruo, London, 1890. Sonnenschein, E. A., T. M. Plauti Mostellaria, second edition, Oxford, 1907. Sonnenschein, E. A., Rudens, ed. min., Oxford, 1901. Stoltz u. Schmalz, Lateinische Grammatik, Miinchen, 19 10. Tyrrell, R. Y., Miles Gloriosus of T. Maccius Plautus, London, 1899. Wagner, Wilhelm, T. Macci Plauti Aulularia, London, 1901. Walker, A. T., The Sequence of Tenses in Latin: A Study Based on Caesar's Gallic War, Lawrence, Kansas, 1899. Walker, A. T., Sequence or Harmony of Tenses?, The Classical Journal, 10. 246-251; 291-299, 19 1 5. INDEX LOCORUM Page Amph. 12 praesim 3 " 64 iussit 4 " 66 eant 4) 45 " 67 uiderint 45 " 68 capiantur 4. 45 " 69 ambissent 45 " 71 ambisset 45 " 72 duint 23, 45 " 74 ambiuerit 23, 45 ' 82 fierent 45 " 85 conciderent 40, 45 " 115 siet 23 " ^77 fuerim 17 195 nuntiem 4 " 205 dicant 23 " 206 uelint 23 " 207 asportassent 23, 24 209 sient, dent, petal 23 *' 215 deducerent 24 225 uicti sint 24 226 dederent 24 " 431 ebiberim 21 465 liceret 40 " 487 fieret 24 " 488 apsoluat 24 " 489 ponatur 24 " 490 celetur 24 " 746 occideris 25 " 749 sciam 34 " 815 dicantur 18 " Sjoferam 4. 41 " 1 123 uicerit 25 " 1142 uorteres 13 Asin. 7 sciretis 35, 40 36 sit 13 86 sit 17 " 188 habeas 39 " 248 sumam 11 " 336 daretur 40 " 393 sit 39 " 443 sit locatum 25 " 589 uerberarem 37 Page " 675 darem 37 " 929 dares 40 Aul. 319 liceret 24 ' 523 metuam 39 ' 550 accusem 26 ' 676 credam 12 ' 681 quaeram 12 ' 736 faceres, ires 13 ' 743 uoluisse 4 ' 743 enicem 4 Bacch. 287 gerant 26 " 291 gereretur 24 " 302 sciscerent 24 353 reddat 5, 26 " 382 dicant 12 " 533 suscenseat 5 555 fi^ceres 37 " 576 reddat, eat 14 " 590 reddat 27 " 591 eat 27 " 607 nesciam 18 " 648 sumeret, quaereret 41 " 690 noceat, suscenseat. . . 5, 27 " 1069 cederem 20, 38 " 1082 possit 5 Capt. Arg. 7 amittatur 11 " 28 mutet 27 " 36 amittat 5 " 49 maneat 14 " 173 sciam 34 " 267 uoluit 6 " 267 inquinet 6 " 379 mittam 6 " 381 detn 6 " 397 redimat, remittant 6 " 411 liceat 18 " 422 laudetur 12 " 467 procedat, occeperit 14 " 568 superes 17 " 569 inuentu's 7, 8, 15, 17 50 Index Locorum 51 Page Capt. 570 conuincas 17 " 779 coniciam 12 " 837 cenes 6 " 837 nanctus es 6, 11 " 932 possim 18 " 1004 delectem 27 Cas. 48 posset 20, 38 " 52 posceret 6, 41 " 56 poscat 6 " 105 curet 7 " 25s facias 19 ' ' 424 luserim 32 " 425 fecerim 32 " 426 creauerim 32 " 448 praetnittam 12 " 604 arcesserem 41 " 681 dicerem 27 " 682-3 caueas 27 " 891 opprimeret 24 Cist. 1 1 meminerimus 19 " 169 deferat 28 " 568 deserat 28 Cure. Arg. 5 tnittat 11 " 226 ferat 32 " 372 sit, siet 14 " 559 abierit, petam 28 " 560 comedim 28 " 608 peruenerit 14 " 608 dixi 14 Epid. 80 abierim 17 " 285 apscedat 14 " 331 habeam 39 " 354fallatur, apparetur 7 " 355 inueni 7, 8, 15, 17 " 416 redierit 28 " 443 deceat 12 " 445 explices 7, 42 " 571 uideas, adeas, des 7 " 638 ueniat 34 " 712 fierem 13 Men. 454 capiat 29 " 475 sperauerit 17 Page Men. 490 perderes 13 " 500 sciam 34 " 712 audeam 19 " 755 «■< 15 " 787-8 geras 8 " 789 opserues 8 " 1057 eas 29 " 1067 pigeat 12 " 1 100 ores 13 Merc. 91 foret 25 " 254 inuenisse 7, 8, 15 " 254 sit, uolueril 15 " 344 loquar 15 " 419 placeat 29 " 472 ibo, dabo il " e^ifaceres 35 " 669 persequar 8, 42 M.G.Arg.ll. 5 nuntiaret 25 " " 7 veniret 25 " " 131 deferat 29 " " 12,9 facerem 4^ " " 143 esset 41 " " 263 participauerit 21 " " 295 pereas 8, 10 " " 295 paratum est 8 " " 452 sitis 35 " " 514 nesciam 19 " " 728 ueneat 8 " " 729 pauperet 8 " " 767 inueni 8, 15 " " 768 admutiletur 8 " " 769 ecficiamus 9 " " 770 abducat, habeat 9 " " 867 gesserit 15 " " 925 sim 35 " " 956 ogerew 41 " " 963 nequeas 30 " " 971 iussit 4 " " 972 facias 9 " " 1097 dixi 15 " " 1098 possit 15 " " iii\6 abeat 9 " " 1158 uelles 41 " " 1238 sis 9 " " 1269 oderim 9 52 The Sequence of Tenses in Plautus Page Most. 89 arbitrarer 36 158 rear 35 183 amarem 37, 38 421 apsterreres 42 422 iret 42 555 sit 39 715 ducerem 42 716 pellerem 42 942 procures, quaeras, uideas, audias 9 969 uenerim 35 1005 censeas 36 1 124 ludificatu sit 30 1 127 conciliarem 42 Pers. Arg. 3 emeret 25 " 56 pauerint 21 " 82 facial 15 " 119 darem 36 " 173 sciret 21, 38 215 sim 39 " 304 nuntiares 43 " 326 auferam 9 " 496 faciam 13 " 538 facerem 25 583 fuerit 21 " 621 mirer 10 Poen. 547 siet 16 554 possimus 19 559 dederis 16 603 ames, pates, pergraecere 30 681 darem 38 775 priuent, diuidant 10 1020 fodiat, metat 10 1057 gnatus sis 30 1066 uiuerent 38 1252 darent, euenisset 39 1276 redirent 42 1407 pereas 13 Page Pseud. 57 adferret 43 " 58 mitteret 43 " 274 possim 39 " 579 po-raui 10 " 583 uincam, spoliem 10 " 597 habitet 30 " 796 coquat 31 " 1233 abduceret 43 " 1319 fierem 21, 39 Rud. 129 adornaret 31 130 faciat 31 218 forem nata 32 379 faceret 36 410 simus natae 31 602 darem 25 61 1 dicam 16 ■jl-j fabuler 10 925 siem 16 927 liberes 11 1026 sies 16 Stick. 249 ires 43 " 255 dares 36, 43 " 366 uenerit 25 " 503 consulam 12 Trin. 14 aleret 25 " 15 exigat II 283 siet 35 " 443 conciliarem 44 " 992 periisses 33 True. Arg. 3 tangat 11 " 81 daret 42 " 140 essem 39 " \()() sciam 34 " 280 nancta es 1 1 " 280 pudeat II " 292 reliqueris 22, 36 " 383 gesseris 16 UNIVERSITY OF CALIFOENIA LIBRARY, BERKELEY THIS BOOK IS DUE ON THE LAST DATE STAMPED BELOW Books not returned on time are subject to a fine of 50c per volume after the third day overdue, increasing to $1.00 per volume after the sixth day. Books not in demand may be renewed if application is made before expiration of loan period. IMl P 19J0 6J^jl491B STACKS JUN 6 1979 DEC 6 1979 JUN ' 1988 J^i K . ) 7 AUTO DISC ^ IBfs 50m-7,'29 UNIVERSITY OF CALIFORNIA LIBRARY UC.BEBKELEYUBIIHH1ES llllllllllll