THE IMPERFECT INDICATIVE IN EARLY LATIN. i . Introductory. It is the purpose of this paper to define and differentiate the various uses of the imperfect indicative, to discover if possible their origin and trace their interrelations, to outline in fact the history of the tense in early Latin. The term ' early Latin ' is used somewhat elastically as including not only all the remains of the language down to about the time of Sulla, but also the first volume of inscriptions (to 44 B. c.) and the works of Varro, for Varro belongs distinctly to the older school of writers in spite of the fact that the Rerum rusticarum libri were written as late as 37 B. c. But exact chronological periods are of little meaning in matters of this sort, and the present outline, being but a frag- ment of a more complete history of the tense, may stop at this point as well as another. Before proceeding to the investigation of the cases of the imperfect occurring in early Latin it is necessary to describe briefly the system by which these cases have been classified. In the first place all cases of the same verb have been placed together so that the individual verb forms the basis of classification. 1 Then verbs of similar meanings have been combined to form larger groups. There result three main groups (and some subdivisions) which for the better understanding of this paper may be tabulated thus: I. Verbs of physical action or state. 1. Motion of the whole of a body, e. g. eo, curro. 2. Action of a part of a body, e. g. do, iacio. 3. Verbal communication, e. g. dico,promilto. 4. Rest or state, e. g. sum, sto, sedeo. II. Verbs of psychic action or state. 1. Thought, e. g.puto, scio, spcro. 2. Feeling, e. g. metuo, atno. 3. Will, e. g. volo, nolo. 1 Cf. Trans. Am. Philolog. Ass., XXX, 1899, pp. 14-15. 164 AMERICAN- JOURNAL OF PHILOLOGY. III. Auxiliary verbs, i. e. verbs which represent such English words as could, should, might, &c, &c, e. g. possum, oportet, decet. Such a system has, of course, many inconsistencies. The verb ago, for instance, may be a verb of action (I. 2) or of verbal com- munication (I. 3), but since instances of this sort were compara- tively rare and affected no important groups of verbs it has seemed best not to separate cases of the same verb. Again I. 3 is logically a part of I. 2, or the verbs grouped under III might perhaps have been distributed among the different subdivisions of I and II. But the object of the classification, to discover the function of each case, has seemed best attained by grouping the verbs as described. By this system verbs of similar meaning, whose tenses are therefore similarly affected, are brought together and this is the essential point. In a very large collection of cases a stricter subdivision would doubtless prove of advantage. 2. The Facts 1 of Usage. There are about 1400 cases of the imperfect indicative in the period covered by this investigation. Of these, however, it has been necessary to exclude 2 from 175 to 180 leaving 1226 from a consideration of which the results have been obtained. The tense appears, therefore, not to have been a favorite, and its comparative infrequency which I have noted already for Plautus and Terence 3 may here be asserted for the whole period of early Latin. About three-quarters of the total number of cases are supplied by Plautus, Terence, and Varro (see Table I). A study of these 1226 cases reveals three general uses of the imperfect indicative : I. The progressive or true imperfect. II. The aoristic imperfect. III. The' shifted' imperfect. Let us consider these in order. 1 In the following pages I have made an effort to state and illustrate the facts, reserving theory and discussion for the third section of this paper. a These are cases doubtful for one reason or another, chiefly because of textual corruption or insufficient context. For the latter reason perhaps too many cases have been excluded, but I have chosen to err in this direction since so much of the material consists of fragments where one cannot feel absolutely certain of the force of the tense. 3 Trans. Am. Philolog. Ass., XXX, p. 22. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 1 65 The true imperfect shows several subdivisions : I A. The simple progressive imperfect. I B. The imperfect of customary past action. I C. The frequentative imperfect. Of these I A and I B include several more or less distinct variations, but all three uses together with their subdivisions betray their relationship by the fact that all possess or are immediately derived from the progressive ' function. This pro- gressive idea, the indication of an act as progressing, going on, taking place, in past time or the indication of a state as vivid, is the true ear-mark of the tense. The time may be in the distant past or at any point between that and the immediate past or it may even in many contexts extend into the present. In duration the time may be so short as to be inappreciable or it may extend over years. The time is, however, not a distinguishing mark of the imperfect. The perfect may be described in the same terms. The kind of action * remains, therefore, the real criterion in the distinction * of the imperfect from other past tenses. I A. The Simple Progressive Imperfect. Under this heading are included all cases in which the tense indicates simple progressive action, i. e. something in the 'doing', ' being ', 4 &c. The idea of progression is present in all the cases, but there are in other respects considerable differences according to which some distinct varieties may be noted. All told there are 680 cases of this usage constituting more than half the total (1226). I I have chosen progressive as more expressive than durative which seems to emphasize too much the time. 2 'Kind of action' will translate the convenient German Aktionsart while ' time ' or ' period of time ' may stand for Zeitstufe. % Herbig in his very interesting discussion, Aktionsart und Zeitstufe (I. F. '896), §107, comes to the conclusion that 'Aktionsart ' is older than ' Zeitstufe ' and that though many tenses are used timelessly none are used in living speech without 'Aktionsart.' * The progressive effect is also found in the present participle (and in parti- cipial adjectives), and indeed the imperfect, especially in subordinate clauses, is often interchangeable with a participial expression, falling naturally into participial form in English also. How close the effect of the imperfect was to that of the present participle is well illustrated by Terence, Heaut. 293-4 nebat . . . texebat and 285 texentem . . . offendimus. Cf. Varro R. R. Ill, 2. 2 cited on p. 167. 166 AMERICAN JOURNAL OF PHILOLOGY. Of these 449 are syntactically independent, 231 dependent. 1 In its ordinary form this usage is so well understood that we may content ourselves with a few illustrations extending over the different groups of verbs. I.i. Verbs of motion. Plautus, 2 Aul. 178, Praesagibat mi animus frustra me ire, quom exibam domo. 1 With the principles of formal description as last and best expressed by Morris (On Principles and Methods of Syntax, 1901, pp. 197-8) all syntacticians will, I believe, agree. Nearly all of them will be found well illustrated in the present paper. For purposes of tense study, however, I have been unable to see any essential modification in function resulting from variation of person and number, although some uses have become almost idiomatic in certain persons, e. g. the immediate past usage with first person sing, of verbs of motion (p. 15). Just how far tense function is affected by the kind of sentence in which the tense stands I am not prepared to say. In cases accompanied by a negative or standing in an interrogative sentence the tense function is more difficult to define than in simple affirmative sentences. It is easier also to define the tense function in some forms of dependent clauses, e. g. temporal, causal, than in others. This is an interesting phenomenon, needing for its solution a larger and more varied collection of cases than mine. At present I do not feel that the influence upon the tense of any of these elements is definite enough to call for greater complexity in the system of classification. While, therefore, I have borne these points constantly in mind, the tables show the results rather than the complete method of my work in this respect. ' In the citation of cases the following editions are used: Fragments of the dramatists, O. Ribbeck, Scaenicae Romanorum poesis fragmenta (I & II), Lipsiae 1897-8 (third edition). Plautus, Goetz and Schoell, T. Macci Plauti comoediae (editio minor), Lipsiae 1892-6. Terence, Dziatzko, P. Terenti Afri comoediae, Lipsiae 1884. Orators, H. Meyer, Oratorum romanorum fragmenta, Turici 1842. Historians, C. Peter, Historicorum Romanorum fragmenta, Lipsiae 1883. Cato, H. Keil, M. Porci Catonis de agricultura liber, Lipsiae 1895, and H. Jordan, M. Catonis praeter lib. de re rustica quae extant, Lipsiae i860. Lucilius, L. Mueller, Leipsic 1872. Auctor ad Herennium, C. L. Kayser, Cornifici rhetoricorum ad C. Heren- nium libri tres, Lipsiae 1854. Inscriptions, Th. Mommsen, C. I. L. I. Ennius (the Annals), L. Mueller, Q. Enni carminum reliquiae, Petropoli 1884. Naevius (Bell, poen.), L. Mueller, Q. Enni carminum reliquiae, Petropoli 1884. Varro, H. Keil, M. Terenti Varronis rerum rusticarum libri tres, Lipsiae 1883. Varro, A. Spengel, M. Terenti Varronis de lingua latina, Berolini 1885. Varro, BUcheler, M. Terenti Varronis saturarum Menippearum reliquiae, Lipsiae 1865. THE IMPERFECT INDICA TIVE IN EARL Y LATIN. 1 67 Id. Amph. 199, Nam quom pugnabant maxume, ego turn fugiebam maxume. Lucilius, Sat., XVI. 12, l ibat forte aries' inquit; I. 2. Verbs of action. Ex incertis incertorum fabulis (comoed. pall.) p. 137, XXIV. R., sed sibi cum tetulit coronam ob coligandas nuptias, T\b\ ferebat; cum simulabat se sibi alacriter dare, Turn ad te ludibunda docte et delicate detulit. Plautus, True. 198 . . . atque opperimino : iam exibit, nam lavabat. Cf. id. Men. 564 (ferebam), Mil. 1336 (temptabam), Epid. 138 (mittebam); Terence, Andr. 545 (dabam); Auctor ad Herenn. 4, 20, 27 (oppetebat). I. 3. Verbal communication. Plautus, Men. 1053, Quin modo Erupui, homines qui ferebant te . . . Apud hasce aedis. tu clamabas deum fidem, Ex incert. incert. &c. 282. XXXII. R., Vidi te, Ulixes saxo sternentem Hectora, Vidi tegentem clipeo classem Doricam : Ego tunc pudendam trepidus hortabar fugam. I. 4. State. Plautus, Aul. 376, Atque eo fuerunt cariora, aes non erat. Id. Mil. 181, Sed Philocomasium hicine etiam nunc est? Pe. Quom exibam, hie erat. Varro, R. R. III. 2. 2., ibi Appium Claudium augurem sedentem invenimus . . . sedebat ad sinistram ei Cornelius Merula . . . Cf. also Plautus, Rud. 846, (sedebanf), Amph. 603 (stabam) &c. &c. II. 1. Verbs of thought. Hist. frag. p. 70, 1. 7, Et turn quo irent nesciebani, ilico manserunt. Plautus, Pseud. 500-1, Non a me scibas pistrinum in mundo tibi, Quom ea muss[c]itabas ? Ps. Scibam. Cf. also Plautus, Rud. 1 186 ,(credebam); Varro R. R. I. 2. 25. (ignorabat), &c. II. 2. Feeling. Plautus, Epid. 138, Desipiebam mentis, quom ilia scripta mittebam tibi. 1 68 AMERICAN JOURNAL OF PHILOLOGY. Id. Bacch. 683, Bacchidem atque hunc suspicabar propter crimen, Chrysale, II. 3. Will. Lucilius, Sat. incert. 48, fingere praeterea adferri quod quis- que volebat: In these cases the act or state indicated by the tense is always viewed as at some considerable distance in the past even though in reality it may be distant by only a few seconds. The speaker or writer stands aloof, so to speak, and views the event as at some distance and as confined within certain fairly definite limits in the past. If, now, the action be conceived as extending to the im- mediate past or the present of the speaker, a different effect is produced, although merely the limits within which the action progresses have been extended. This phase of the progressive imperfect we might term the imperfect of the immediate past 1 or the interrupted 2 imperfect, since the action of the verb is often interrupted either by accomplishment or by some other event. A few citations will make these points clearer : Plautus, Stich. 328, ego quid me velles visebam. Nam mequidem harum miserebat. — '\ was coming to see what you wanted of me (when I met you) ; for I've been pitying (and still pity) these women.' In the first verb the action is interrupted by the meeting ; in the second it continues into the present, the closest translation being our English compound pro- gressive perfect, a tense which Latin lacked. The imperfect ibam is very common in this usage, cf. Plautus, True. 921, At ego ad te ibam = l was on my way to see you (when you called me), cf. Varro, R. R. II. 11. 12; Terence, Phorm. 900, Andr. 580. But the usage is by no means confined to verbs of motion (I. 1) alone. It extends over all the categories: I. 2. Motion. Plautus, Aulul. 827 (apparabas), cf. Andr. 656. 1 In Greek the aorist is used of events just past, but of course with no pro- gressive coloring, cf. Brugmann in I. Miiller's Handbuch, &c, II 2 , p. 185. * E. Rodenbusch, De temporum usu Plautino quaest. selectae, Argentorati 1888, pp. n-12, recognizes and correctly explains this usage, adding some examples of similar thoughts expressed by the present, e. g. Plautus, Men. 280 (quaeris), ibid. 675 (quaerit), Amph. 542 (numquid vis, a common leave-taking formula). In such cases the speaker uses imperfect or present according as past or present predominates in his mind, the balance between the two being pretty even. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 1 69 I. 3. Verbal communication. Terence, Eun. 378 (iocabar), Heaut. 781 (dicebam) ; Plautus, Trin. 212 (aibanf). I. 4. Rest. Plautus, Cas. 532 (eratn), cf. Men. n 35. Terence, Eun. 87 (stabam), Phorm. 573 {cotnmorabar). II. 1. Thought. Terence, Phorm. 582 (scibam), cf. Heaut. 309. Plautus, Men. 1072 (censebam), cf. Bacch. 342, As. 385 &c. II. 2. Feeling. Plautus, Stich. 329 (miserebaf) ; Turpilius, 107 V R. (sperabam). II. 3. Will. Plautus, As. 392 and 395 (volebatn), Most. 9, Poen. 1231. 1 III. Auxiliary verbs. Plautus, Epid. 98 (so/ebam), cf. Amph. 711. Terence, Phor- mio 52 (conabar). In this usage the present or immediate past is in the speaker's mind only less strongly than the point in the past at which the verb's action begins. The pervading influence of the present is evident not only because present events are usually at hand in the context, but also from the occasional use with the imperfect of a temporal particle or expression of the present, cf. Plaut. Merc. 884, Quo nunc ibas = ' whither were you (are you) going ? ' Terence, Andr. 657, immo etiam, quom tu minus scis aerumnas meas, Haec nuptiae non adparabanfur mihi, ' Rodenbusch (p. 26) labors hard to show that this case is like the preceding and not parallel with the cases of volui which he cites on p. 24 with all of which an infinitive of the verb in the main clause is either expressed or to be supplied. Following Bothe, he alters deicere to dice (which he assigns to Adelphasium) and refers quod to the amabo and amflexabor of I230 = 'meine Absicht'. But there is no need of this. Infinitives occur with some of the cases cited by Rodenbusch himself on p. II, e. g. Bacch. 188 (189) Istuc volebatn . . . fercontarier, Trin. 195 Istuc voUbam scire, to which may be added Cas. 674 Dicere vilicum volebatn and ibid. 702 illud . . . dicere volebatn. It is true that the perfect is more common in such passages, but the imperfect is by no means excluded. The difference is simply one of the speaker's point of view: quod volui = ' what I wished * (complete) ; quod valebant = ' what I was and am wishing ' (incomplete). As. 212, which also troubles Rodenbusch, is customary past. 12 170 AMERICAN JOURNAL OF PHILOLOGY. Nee postulabat nunc quisquam uxorem dare. Merc. 197, Equidem me tarn censebam esse in terra atque in tuto loco : Verum video . . . In the last two cases note the accompanying presents, set's and video. The immediate past also is indicated by a particle, e. g. Plautus, Cas. 594 ad te hercle ibam commodum. There are in all 207 l cases of this imperfect of the immediate past. They are distributed pretty evenly over the various groups of verbs as will be seen from the following table: No. of Cases. I. I Verbs of motion, 26 I. 2 it " action, 17 I. 3 (i " verbal communication, 31 I. 4 " " state, 35 II. 1 it " thought, 36 II. 2 " " feeling, 35 II. 3 " " will, 13 III. Auxil iary verbs, 14 207 The verbs proportionately most common in this use are ibam and volebam which have become idiomatic. The usage is especially common in colloquial Latin, but 16 cases 5 occurring outside the dramatic literature represented chiefly, of course, by Plautus and Terence. By virtue of its progressive force the imperfect is a vivid tense and as is well known, became a favorite means in the Ciceronian period of enlivening descriptive passages. It was especially used to fill in the details and particulars of a picture (imperfect of situa- tion). 8 This use of the tense appears in early Latin also, but with much less frequency. The choice of the tense for this purpose is a matter of art, whether conscious or unconscious. At times, indeed, there is no apparent reason for the selection of an imper- fect rather than a perfect except that the former is more graphic, 1 Somewhat less than one-third of the total (680) progressive cases. 5 These cases are Ennius, Ann. 204, C. I. L. I. 201. 1 1 (3 cases), Varro, L. L. 5. 9 (1 case), and Auctor ad Herenn. 1. 1. 1 (2 cases), 1. 10. 16, 2. 1. 2, 2. 2. 2 (2 cases), 3. 1. 1 (2 cases), 4. 34. 46, 4. 36. 48, 4. 37. 49. All of these are in passages of colloquial coloring, either in speeches or, especially those in auctor ad Herenn., in epistolary passages. 3 I use this term for all phases of the tense used for graphic purposes. THE IMPERFECT INDICATIVE IN EARL Y LA TIN. 171 and if it were possible to separate in every instance these cases from those in which the imperfect may be said to have been required, we should have a criterion by which we might dis- tinguish this use of the imperfect from others. But since the progressive function of the tense is not altered, such a distinction is not necessary. Statistics as to the frequency of the imperfect of situation in early Latin are worth little because the chief remains of the language of that period are the dramatists in whom naturally the present is more important than the past. The historians, to whom we should look for the best illustrations of this usage, are for the most part preserved to us in brief fragments. Nevertheless an examination of the comparatively few descriptive passages in early Latin reveals several points of interest. In Plautus and Terence the imperfect was not a favorite tense in descriptions. Bacch. 258-307, a long descriptive passage of nearly 50 lines, interrupted by unimportant questions, shows only 4 imperfects (1 aoristic) amid over 40 perfects, historical presents, &c. Capt. 497-5151 Amph. 203-261, Bacch. 947-970, show but one case each. Stich. 539-554 shows 5 cases of erat. In Epid. 207-253 there are 10 cases. In the descriptive passages of Terence the imperfect is still far from being a favorite tense, though relatively more common than in Plautus, cf. Andr. 48 ff., 74-102, Phorm. 65-135 (containing 11 imperfects). But Eunuch. 564-608 has only 4 and Heaut. 96-150 only 3. Another very instructive passage is the well-known description by Q. Claudius Quadrigarius of the combat between Manlius and a Gaul (Peter, Hist. rom. fragg., p. 137, 10b). In this passage of 28 lines there are but 2 imperfects. The very similar passage describing the combat between Valerius and a Gaul and cited by Gellius (IX, n) probably from the same Quadrigarius contains 8 imperfects in 24 lines. Since Gellius is obviously retelling the second story, the presumption is that the passage in its original form was similar in the matter of tenses to the passage about Manlius. In other words Gellius has 'edited' the story of Valerius, and one of his improvements consists in enlivening the tenses a bit. He describes the Manlius passage thus : Q. Claudius primo annalium purissime atque illustrissime simplicique et incompta orationis antiquae suavitate descripsit. This simplex et incompta suavitas is due in large measure to the fact that 172 AMERICAN JOURNAL OF PHILOLOGY. Quadrigarius has used the simple perfect (19 times), varying it with but few (4) presents and imperfects (2). A closer com- parison of the passage with the story of Valerius reveals the difference still more clearly. Quadrigarius uses (not counting subordinate clauses) 19 perfects, 4 presents, 2 imperfects ; Gellius, 4 perfects, 9 presents, 8 imperfects. In several instances the same act is expressed by each with a different tense : Quadrigarius. Gellius. processit (bis), f procedebat, \ progrediiur, constitit, c congrediuntur, \ consistent, constituerunt, conserebantur manus, 8 perfects of acts in 5 imperfects of acts combat. of the corvus. Gellius has secured greater vividness at the expense of simplicity and directness. This choice of tenses was, as has been said, a matter of art, whether conscious or unconscious. The earlier writers seem to have preferred on the whole the barer, simpler perfect even in passages which might seem to be especially adapted to the imperfect, historical present, &c. The perfect, of course, always remained far the commoner tense in narrative, and instances are not lacking in later times of passages 1 in which there is a striking preponderance of perfects. Nevertheless the imperfect, as the language developed, with the growth of the rhetorical tendency and a consequent desire for variety in artistic prose and poetry, seems to have come more and more into vogue. 2 The fact that the function of a tense is often revealed, denned, and strengthened by the presence in the context of particles of various kinds, subordinate clauses, ablative absolutes, &c, &c, 1 E. g. Caesar, B. G. I. 55 and 124-5. s The relative infrequency of the tense in early Latin was pointed out on p. 164. Its growth as a help in artistic prose is further proved by the fact that the fragments of the later and more rhetorical annalists, e. g. Quadrigarius, Sisenna, Tubero, show relatively many more cases than the earliest annalists. This is probably not accident. When compared with the history of the same phenomenon in Greek, where the imperfect, so common in Homer, gave way to the aorist, this increase in use in Latin may be viewed as a revival of a usage popular in Indo-European times. Cf. p. 185, n. 2. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. I73 was pointed out in Trans. Am. Philol. Ass. XXX, pp. 17 ff. What was there 1 said of Plautus and Terence may here be extended to the whole period of early Latin. The words and phrases used in this way are chiefly temporal. Some of those occurring most frequemly are: modo, commodum ; turn, tunc; simul; dudum, iam dudum; iam, primo, primulum ; nunc; ilico; olim, quondam; semper, saepe; fere, plerumque ; Ha, 2 &c, &c. A rough count shows in this class about 120 cases,' accompanied by one or more particles or expressions of this sort. Some merely date the tense, e. g., turn, modo, dudum, &c. Others, as saepe, fere, primulum, have a more intimate connection with the function. Naturally the effect of the latter group is clearest in the imperfects of customary past action, the frequentative, &c, and will be illustrated under those headings. Here I will notice only a few cases with iam, primulum, &c, which illustrate very well how close the relation between particle and tense may be. The most striking cases are : Plautus, Merc. 43, amare valide coepi[t] hie meretricem. ilico Res exulatum ad illam <c>lam abibat patris. Cf. Men. 1 1 16, nam tunc dentes mihi cadebant primulum. id. Merc. 197, Equidem me iam censebam esse in terra atque in tuto loco : Verum video . . . id. Cist. 566, Iam perducebam illam ad me suadela mea, Anus ei <quom> amplexast genua . . . id. Merc. 212, credet hercle: nam credebat iam mihi. The unquestionably inceptive force of these cases arises from the combination of tense and particle. No inceptive* function can be proved for the tense alone, for I find no cases with inceptive force unaccompanied by such a particle. ■Cf. also Morris, Syntax, p. 83. 5 How far the nature of the clause in which it stands may influence the choice of a tense is a question needing investigation. That causal, explanatory, characterizing, and other similar clauses very often seem to require an im- perfect is beyond question, but the proportion of imperfects to other tenses in such clauses is unknown. Cf. p. 166, n. 1. s No introductory conjunctions are included in this total, nor are other particles included, unless they are in immediate connection with the tense. 4 In Trans. Am. Philolog. Ass. XXX, p. 21, I was inclined to take at least Merc. 43 as inceptive. This I now believe to have been an error. The inceptive idea was most commonly expressed by coepi -\- m&n. which is very common in Plautus and Varro. We have here the opposite of the phenomenon discussed on p. 177. 174 AMERICAN JOURNAL OF PHILOLOGY. There are a few cases in which the imperfect produces the same effect as the imperfect of the so-called first periphrastic conjuga- tion : Terence, Hec. 172, Interea in Imbro moritur cognatus senex. Horunc: ea ad hos redibal lege hereditas.=reditura erat, English ' was coming ', ' was about to revert ', cf. Greek pi\\a> with infinitive. Cf. Phorm. 929, Nam non est aequum me propter vos decipi, Quom ego vostri honoris causa repudium alterae Remiserim, quae dotis tantundem <fti£«/.=datura erat &c. In these cases the really future event is conceived very vividly as already being realized. Plautus, Amph. 597 seems to have the effect of the English 'could': Neque . . . mihi credebam primo mihimet Sosiae Donee Sosia . . . ille . . . But the * could ' is probably inference from what is a very vivid statement. A Roman would probably not have felt such a shading. 1 I B. The Imperfect of Customary Past Action. The imperfect may indicate some act or state at some appreci- able distance in the past as customary, usual, habitual &c. The act or state must be at some appreciable distance in the past (and is usually at a great distance) because this function of the tense depends upon the contrast between past and present, a contrast so important that in a large proportion of the cases it is enforced by the use of particles. 2 The act (or state) is conceived as repeated at longer or shorter intervals, for an act does not become customary until it has been repeated. This customary act usually takes place also as a result or necessary concomitant of certain conditions expressed or implied in the context, e. g. maiores nosiri olim &c, prepares us for a statement of what they used to do. The act may indeed be conceived as occurring only as a result of a certain expressed condition, e. g. Plautus, Men. 484 mulier quidquid dixerat, 1 Some of the grammars recognize ' could' as a translation, e. g., A. & G. § 277 g- 8 E. g. turn, tunc, olim &c. with the imperfect, and nunc &c. with the con- trasted present. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 1?$ Idem ego dicebam = my words would be uttered only as a result of hers. 1 There are 462 cases of the customary past usage of which 218 occur in independent sentences, 244 in dependent. This large total, more than one-third of all the cases, is due to the character of Varro's De lingua latina from which 289 cases come. This is veritably a ' customary past ' treatise, for it is for the most part a discussion of the customs of the old Romans in matters pertaining to speech. Accordingly nearly all the imperfects fall under this head. Plautus and Terence furnish 112. The remaining 61 are pretty well scattered. As illustrations of this usage I will cite (arranging the cases according to the classes of verbs) : I. 1. Plautus, Pseud. 1180, Noctu in vigiliam quando ibat miles, quom tu Has simul, Conveniebatne in vaginam tuam machaera militis ? Terence, Hec. 157, Ph. Quid ? interea ibatne ad Bacchidem ? Pa. Cottidie. Varro, L. L. 5. 180, qui iudicio vicerat, suum sacramentum e sacro auferebat, victi ad aerarium redibat. I. 2. Plautus, Bacch. 429, Saliendo sese exercebant magis quam scorto aut saviis. (cf. the whole passage). Hist, fragg., p. 83. 27, Cn., inquit, Flavius, patre libertino natus, scriptum faciebat (occupation) isque in eo tempore aedili curuli apparebat, . . . I. 3. Terence, Eun. 398, Vel rex semper maxumas Mihi agebal quidquid feceram : Varro, L. L., 5. 121, Mensa vinaria rotunda nominabalur Cili- bantum ut etiam nunc in castris. Cf. L. L. 7. 36, appellabant, 5. 118, 5. 167 &c. 1 This usage seemed to me formerly sufficiently distinct to deserve a special class and the name 'occasional', since it is occasioned by another act. It is at best, however, only a sub-class of the customary past usage and in the present paper I have not distinguished it in the tables. It is noteworthy that the act is here at its minimum as regards repetition and that it may occur in the immediate past, cf. Rud. 1226, whereas the customary past usage in its pure form is never used of the immediate past. The usages may be approxi- mately distinguished in English by 'used to', 'were in the habit of &c. (pure customary past), and 'would' (occasional), although 'would' is often a good rendering of the pure customary past. Good cases of the occasional usage are : Plautus, Merc. 216, 217 ; Poen. 478 S ; Terence, Hec. 804 ; Hist, fragg. p. 202. 9 (5 cases), ibid. p. 66. 128 (4 cases). 176 AMERICAN JOURNAL OF PHILOLOGY. I. 4. Plautus, Bacch. 421, Eadem ne erat haec disciplina tibi, quom tu adulescens eras ? C. I. L. I. 1011.17 Ille meo officio adsiduo florebat ad omnis. II. 1. Auctor ad Herenn. 4. 16. 23, Maiores nostri si quam unius peccati mulierem damnabant, simplici iudicio multorum rnaleficiorum convictam putabant. quo pacto ? quam inpudicam iudicarant, ea venefici quoque damnata existutnabatur. Cato, De ag., 1, amplissime laudari existimabatur qui ita lau- dabatur. II. 2. Plautus, Epid. 135, Illam amabam olim: nunc tarn alia cura impendet pectori. Varro, R. R. III. 17.8, etenim hac incuria laborare aiebat M. Lucullum ac piscinas eius despiciebat quod aestivaria idonea non haberent. III. 3. Plautus, As. 212, quod nolebant ac votueram, de industria Fugiebatis neque conari id facere audebatis prius. Cf. the whole passage. Varro, L. L. 5. 162, ubi quid conditum esse volebant, a celando Cellam appellarunt. III. Terence, Phorm. 1 90, Tonstrina erat quaedam : hie sole- bamusfere Plerumque earn opperiri, . . . Varro, L. L. 6. 8, Solstitium quod sol eo die sistere videbatur . . . The influence of particles 2 and phrases in these cases is very marked. I count about 1 10 cases, more than I of the total, with which one or more particles appear. Those expressions which emphasize the contrast are most common, e. g. turn, olim, me puero with the imperfect, and nunc, iam &c. with the contrasted present. This class also affords excellent illustrations of the reciprocal influence of verb-meaning' and tense-function. In Varro there are 50 cases, out of 289, of verbs of naming, calling, &c, which are by nature evidently adapted to the expression of the customary past. Such are appellabam, nominabam, vocabam, vocitabam, &c. But the most striking illustration is found in verbs of customary action, e. g. soleo, adsuesco, consuesco, which by their 1 Cf. Trans. Am. Philolog. Ass. XXX, p. 19. s Note as illustrations the italicized particles in the citations, pp. 175-6. 3 Cf. Morris, Syntax, p. 47, and p. 72, with note. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 1 77 meaning possess already the function supplied to other verbs by the tense and context. When a verb of this class occurs in the imperfect of customary past the function is enhanced. Naturally, however, these verbs occur but rarely in the imperfect, for in any tense they express the customary past function. It is interesting to note the struggle for existence between various expressions of the same thought. A Roman could express the customary past idea in several ways, of which the most noticeable are the imperfect tense, soleo or the like with an infinitive, or various periphrases such as mos erat. Of these possibilities all are rare save the first, the imperfect tense. There are but 12 cases of soleo, consuesco, &c, occurring in the imperfect indicative in early Latin. These are all cases of solebam, and 9 of them are imperfects of customary past action. 1 One would expect to find in common use the perfect of these verbs with an infinitive, but, although I have no exact statistics on this point, a pretty careful lookout has convinced me that such expressions are by no means common. 2 Periphrases with mos, consuetudo, &c, are also rare. Comparing these facts with the large number of cases in which the customary past function is expressed by the imperfect, we must conclude that this was the favorite mode of expression already firmly established in the earliest literature. 8 I C. The Frequentative Imperfect. In the proper context 4 the imperfect may denote repeated or insistent action in the past. Although resembling the imperfect of customary past action, in which the act is also conceived as 1 Terence, Phorm. go; Varro, R.R. 1.2. 1, and II. 7. I, L. L. 5. 126; Auctor ad Herenn. 4. 54. 67 ; Lucilius, IV. 2, &c. s A collection of perfects covering 18 plays of Plautus shows but 15 cases of solitus est, consuevit, &c. My suspicion, based on Plautus and Terence, that these periphrases would prove common has thus been proven groundless. 8 The variation between imperfect and perfect is well illustrated by Varro, L. L. 5. 162, ubi cenabant, cenaculum vocitabant, and id. R. R. I. 17. 2, iique quos obaeratos nostri vocitarunt, where the frequentative verb expresses even in the perfect the customary past function. For the variation between the customary past imperfect and the perfect of statement cf. Varro's L. L. almost anywhere, e. g. 5. 121, mensa . . . rotunda nominabatur Clibantum. 5. 36, ab usu salvo saltus nominarunt. So compare 5. 124 (appellarunt) with R. R. I. 2. 9 (appellabant). Cf. also L. L. 5. 35 qua ibant . . . iter appellarunt ; qua id auguste, semita.ut semiter dictum. 4 Cf. Herbig, Aktionsart und Zeitstufe (I. F. 1896, § 59). 178 AMERICAN JOURNAL OF PHILOLOGY. repeated, the frequentative usage differs in that there is no idea of habit or custom, and the act is depicted as repeated at intervals close together and without any conditioning circumstances or contrast with the present. I find only 13 cases of this usage, 7 of which are syntactically independent, 6 dependent. All occur in the first three classes of verbs. The cases are : Plautus, Pers. 20, miquidem tu iam eras mortuos, quia non visitabam. Ibid. 432, id tibi suscensui, Quia te negabas credere argentum mihi. Rud. 540, Tibi auscultavi : tu promittebas mihi Mi esse quaestum maxumum meretricibus : Capt. 917, Aulas . . . omnis confregit nisi quae modiales erant : Cocum percontabatur, possentne seriae fervescere : As. 938, Dicebam, pater, tibi ne matri consuleres male. Cf. Mil. Gl. 1410 (dicebaf). True. 506, Quin ubi natust machaeram et clupeum poscebat sibi ? Epid. 59, Quia cottidie ipse ad me ab legione epistulas Mittebat: cf. ibid. 132 (missiculabas). Merc. 631, Promittebas te os sublinere meo patri : ego me[t] credidi Homini docto rem mandar<e>, . . . Ennius, Ann. 43, haec ecfatu' pater, germana, repente recessit. Nee sese dedit in conspectum corde cupitus, quamquam multa manus ad caeli caerula templa iendebam lacrumans et blanda voce vocabam. Hist, fragg., p. 138. 11 (Q. Claudius Quadrigarius), Ita per sexennium vagati Apuliam atque agrum quod his per militem licebat expoliabaniur. This class is so small and many of the cases are so close to the simple progressive and the imperfect of situation that it is tempting to force the cases into those classes. 1 A careful con- 1 How close the frequentative notion may be to the imperfect of the immediate past is well illustrated by As. 938 (cited above). In this case we have virtually an imperfect of the immediate past in which, however, the frequentative coloring predominates : dicebam means not ' I've been telling ', but 'I've kept telling', &c. Cf. also Pseud. 422 (dissimulabam) for another case of the imperfect of the immediate past which is close to the frequentative. In its pure form, however, the frequentative imperfect does not hold in view the present. THE IMPERFECT INDJCA TIVE IN EARL V LA TIN. 1 79 sideration of each case has, however, convinced me that the frequentative function is here clearly predominant. In Plautus, Pers. 20, E pid. 131, Capt. 917, it is impossible to say how much of the frequentative force is due to the tense and how much to the form of the verbs themselves ; both are factors in the effect. Verbs like mitto,promitio, voco, and even dico, are also obviously adapted to the expression of the frequentative function. It is noteworthy that in this usage a certain emphasis is laid on the tense. In eight of the cases the verb occupies a very em- phatic position, in verse often the first position in the line, cf. the definition on p. 177. I D. The Conative Imperfect. The imperfect may indicate action as attempted in the past. There must be something in the context, usually the immediate context, to show that the action of the verb is fruitless. There are no certain cases of this usage in early Latin. I cite the only instances, four in number, which may be interpreted as possibly conative : Plautus, As. 931, Arg. Ego dissuadebam, mater. Art. Bellum filium. Id. Epid. 215, Turn meretricum numerus tantus quantum in urbe omni fuit Obviam ornatae occurrebant suis quaeque | amatoribus : Eos captabant. Auctor ad Herenn., 4. 55. 68, . . . cum pluribus aliis ire celerius coepit. illi praeco faciebat audientiam; hie subsellium, quod erat in foro, cake premens dextera pedem defringit et . . . Hist, fragg., p. 143. 46, Fabius de nocte coepit hostibus castra simulare oppugnare, eum hostem delectare, dum collega id caperet quod capiabat. But in the second and fourth cases the verb capto itself means to 'strive to take', 'to catch at' &c, and none of the conative force can with certainty be ascribed to the tense. In the first case, again, the verb dissuadebam means 'to advise against', not 'to succeed in advising against' (dissuade). Argyrippus says : ' I've been advising against his course, mother ', not ' I've been trying, or I tried, to dissuade him'. The imperfect is, therefore, of the common immediate past variety. 1 1 Cf. a few lines below (938) dicebam. 180 AMERICAN JOURNAL OF PHILOLOGY. In Auct. ad Herenn., 4. 55. 68, the imperfect is part of the very vivid description of the scene attending the death of Tiberius Gracchus. Indeed the whole passage is an illustration of demon- stratio or vivid description which the author has just defined. The acts of Gracchus and his followers are balanced against those of the fanatical optimates under Scipio Nasica: 'While the herald was silencing 1 the murmurs in the contio, Scipio was arming himself &c. Though it may be true that the act indi- cated by faciebat audientiam was not accomplished, this seems a remote inference and one that cannot be proved from the context. If my interpretation of these cases is correct, there are no certain 1 instances of the conative imperfect in early Latin. There is but one case of conabar (Terence, Phorm. 52) and one of temptabam (Plautus, Mil. gl. 1336). Both of these belong to the immediate past class, the conative idea being wholly in the verb. II. The Aoristic Imperfect. The imperfect of certain verbs may indicate an act or state as merely past without any idea of progression. In this usage the kind of action reaches a vanishing point and only the temporal element of the tense remains. The imperfect becomes a mere preterite, cf. the Greek aorist and the Latin aoristic perfect. The verbs to which this use of the imperfect is restricted are, in early Latin, two verbs of saying, aio and dico, and the verb sum with its compounds. There are 56 cases of the aoristic imperfect in early Latin (see Table II), 48 of which occur in syntactically independent sen- tences. Some citations follow: Plautus, Bacch. 268, Quotque innocenti ei dixit contumelias. Adulterare eum aibat rebus ceteris. Id. Most. 1027, Te velle uxorem aiebat tuo gnato dare : Ideo aedificare hoc velle aiebat in tuis. Th. Hie aedificare volui? Si. Sic dixit mihi. Id. Poen. 900, Et ille qui eas vendebat dixit se furtivas vendere: Ingenuas Carthagine aibat esse. 1 Faciebat audientiam seems a technical expression, cf. lexicon. 2 The case cited by Gildersleeve- Lodge, § 233, from Auct. ad Herenn., 2. I. 2, ostendebatur seems to me a simple imperfect and there is nothing in the context to prove a conative force, cf. 3. 15. 26 demonstrabatur. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. l8 1 In these cases note the parallel cases of dixit, cf. id. Trin. 1140, Men. 1 141 &c, &c. I note but three cases of dicebam : Terence, Eun. 701, Ph. Unde [igitur] fratrem meum esse scibas ? Do. Parmeno Dicebat eum esse. Cf. Plautus, Epid. 598 for a perfect used like this. Varro, R. R. II. 4. 11, In Hispania ulteriore in Lusitania [ulteriore] sus cum esset occisus, Atilius Hispaniensis minime mendax et multarum rerum peritus in doctrina, dicebat L. Volumnio senatori missam esse offulam cum duabus costis . . . Ibid. III. 17. 4, pisces . . . quos sacrificanti tibi, Varro, ad tibicinem [graecum] gregatim venisse dicebas ad extremum litus atque aram, quod eos capere auderet nemo, . . . In these cases the verb dico becomes as vague as is aio in the preceding citations. Plautus, Poen. 1069, Nam mihi sobrina Ampsigura tua mater fuit, Pater tuos is erat frater patruelis meus, Et is me heredem fecit, . . . Id. Mil. gl. 1430, Nam illic qui | ob oculum habebat lanam nauta non erat. Py. Quis erat igitur? Sc. Philocomasio amator. Id. Amph. 1009, Naucratem quem convenire volui in navi non erat, Neque domi neque in urbe invenio quemquam qui ilium viderit. 1 Id. Merc. 45, Leno inportunus, dominus eius mulieris, Vi sum<m>a[t] quicque utpoterat rapiebat domum. In such cases as the last the imperfect has become formulaic, cf. quam maxime poter at, &c. 1 Rodenbusch, pp. 8-10, after asserting that the imperfect of verbs of saying and the like is used in narratio like the perfect (aorist), cites a number of illustrations in which (he adds) the imperfect force may still be felt ! But a case in which the imperfect force may still be felt does not illustrate the imperfect in simple past statements, if that is what is meant by narratio. Only four of R.'s citations are preterital (aoristic), and these are all cases of aibam (Plautus, Amph. 807, As. 208, 442, Most. 1002). The same may be said of the citations on p. g, of which only Eun. 701 is aoristic. J. Schneider (De temporum apud priscos latinos usu quaestiones selectae, program, Glatr, 1888) recognizes the aoristic use of aibat, but his statement that the comic poets used perfect and imperfect indiscriminately as aorists cannot be accepted. 1 82 AMERICAN JOURNAL OF PHILOLOGY. III. The Shifted Imperfect. In a few cases the imperfect appears shifted from its function as a tense of the past, and is equivalent to (i) a mere present; or (2) an imperfect or pluperfect subjunctive. The cases equivalent to a present 1 are all in Varro, L. L., and are restricted to verbs of obligation {oportebat, debebaf) : L. L. 8. 74, neque oportebat consuetudinem notare alios dicere Bourn greges, alios Boverum, et signa alios Iovum, alios Ioverum. Ibid. 8. 47, Nempe esse oportebat vocis formas ternas ut in hoc Humanus, Humana, Humanum, sed habent quaedam binas . . . ibid. 9. 85, si esset denarii in recto casu atque infinitam multi- tudinem significaret, tunc in patrico denariorum dici oportebat. Ibid. 8. 65, Sic Graeci nostra senis casibus [quinis non] dicere debebant, quod cum non faciunt, non est analogia.* The cases equivalent to the subjunctive are confined to sat &c. + erat (6 cases), poteram (3 cases), decebat (1 case), and sequebatur (1 case). As illustrations may be cited : Plautus, Mil. gl. 755, Insanivisti hercle : nam idem hoc homini- bus sat [a] era\ti\t decern. Auct. ad Herenn. 2. 22. 34, nam hie satis erat dicere, si id modo quod esset satis, curarent poetae. = ' would have been,' cf. ibid. 4. 16. 23 (iniquom erat), Plautus, Mil. gl. 911, Bonus vates poieras esse : = ' might be ' or ' might have been '. Id. Merc. 983 b, Vacuum esse istac ted aetate his decebat noxiis. Eu. Itidem ut tempus anni, aetate<m> aliam aliud factum condecet. Varro, L. L. 9. 23, si enim usquequaque non esset analogia, turn sequebatur ut in verbis quoque non esset, non, cum esset usquequaque, ut est, non esse in verbis . . . This is a very odd case and I can find no parallel for it.* 1 Varro uses the perfect also of these verbs as equivalent to the present of general statements. Cf. L. L. 8, §§ 72-74, where debuit occurs 4 times as equivalent to debet, § 48 (debuerunt twice), § 50 (pportuit = oportet). The perfect infinitive is equivalent to the present, e. g. in 8, §61 and §66 (debuisse . . . dici). The tenses are of very little importance in such verbs. 8 Note the presents expressed in the second and fourth citations. 3 The remaining cases are: Plautus, True. 511 (poterai), id. Rud. 269 (aequitts erat), Lucilius, Sat. 5. 47 M. (sat erat), Auctor ad Herenn. 4. 16. 23 (iniquom erat), ibid. 4. 41. 53 (quae separatim dictae . . . infimae erant). THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 183 TABLE I. Authors and Functions. Total. I. True Imperfect. II. A oris tic. A. B. c. III. Shifted. Progressive. Cust. Past. Frequent. 427 287 84 10 41 5 Terence 226 187 28 10 1 24 2 22 Dramatists 3 . . 69 60 7 2 Historians .... 52 34 16 I I 12 9 3 13 11 1 1 Auctor ad Her. 79 63 11 2 3 Inscriptions . . 4 3 1 320 24 289 2 5 1226 680 462 13 56 15 1 The fragments of Cato's historical work are included in the historians. 'Including the epic fragments of Ennius and Naevius. TABLE II. Verbs and Functions. Total Cases. I. True Imperfect. Classes of Verbs. A. Progressive. B. Cust. Past. C. Frequent. II.Aoristic. III. Shifted. Ind. Dep. Ind. Dep. Ind. I 6 7 Dep. I 3 2 6 7 13 Ind. 38 IO 48 Dep. 5 3 8 48 56 Ind. I 4 7 12 Dep. I. Physical. 85 302 233 346 91 QO 19 60 40 96 46 138 46 53 9 21 449 17 45 26 69 85 30 7 12 231 449 680 9 75 64 48 13 6 3 218 17 82 46 75 7 1 3 13 244 218 462 3. Verbal commun. 4. Rest, state, &c. (tram 220) II. Psychical. 3. Will 2 III. Auxiliaries. I 1226 3 12 15 184 american journal of philology. 3. Historical and Theoretical. The original function of the imperfect seems to have been to indicate action as progressing in the past, the simple progressive imperfect. This is made probable, in the first place, by the fact that this usage is more common than all others combined, including, as it does, 680 out of a total of 1226 cases. This proportion is reduced, as we should remember, by the peculiar character of the literature under examination, which contains relatively so little narrative, and especially by the nature of Varro's De lingua latina in which the cases are chiefly of the customary past variety. 1 Moreover, the customary past usage itself, and also the frequentative and the conative, are to be regarded as offshoots of the progressive usage of which they still retain abundant traces, so that if we include in our figures all the classes in which a trace of the progressive function remains we shall find that 11 55 of 1226 cases are true imperfects (see table II). Another support for the view that the progressive function is original may be drawn from the probable derivation of the tense. Stolz 2 (after Thurneysen) derives the imperfect from the infinitive in -e and an old aoristof the root *bhu. The idea of progression was thus originally inherent in the ending -bam. Let us now establish as far as possible the relations subsisting between the various uses of the true imperfect (IA, B, C, D), turning our attention first to the simple progressive (IA) and its variations. The relation between the progressive imperfect in its pure form and the usage which has been named the imperfect of the immediate past is not far to seek. The progressive function remains essentially unchanged. The only difference lies in the extension of the time up to the immediate past (or present) in the case of the immediate past usage. The transition between: ibat exulatutn'' = ' he was going into exile ' (when l See p. 175. 2 In I. Muller's Handb. d. kl. Alt. II., 2 § 113, p. 376. Lindsay, Latin Lang., pp. 489-490, emphasizes the nominal character of the first element in the compound, and suggests a possible I. E. *-bhwam, -as, &c, as antecedent of Latin -bam, -ids, -bat. He also compares very interestingly the formation of the imperfect in Slavonic, which is exactly analogous to this inferred Latin formation, except that the ending comes from a different root. 3 Cf. Plautus, Merc. 981. THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 1 85 I saw him at a more or less definite point in the past) and ibat exulatum = ' he was going (has been going) into exile' (but we have just met him) is plain enough. The difference is one of context. In this imperfect of the immediate past the Romans possessed a sub- stitute for our English compound perfect tense, 'have been doing ', &C 1 In the imperfect of situation also the function of the tense is not altered. The tense is merely applied in a different way, its progressive function adapted to vivid description, and we have found it already in the earliest 2 literature put to this use. In its extreme form it occurs in passages which would seem to require nothing more graphic than a perfect. Indeed, we must guard against the view that the imperfect is a stronger tense than the perfect; it is as strong, but in a different way, and while the earlier writers preferred in general the perfect, 8 the imperfect grew gradually in favor until in the period marked by the highest development of style the highest art consisted in a happy combination * of the two. The imperfect of customary past action is, as we have seen, already well established in the earliest literature. A glance at Table I would seem to show that it grew to sudden prominence in Varro, but the peculiar nature of Varro's work has already been pointed out, so that the apparent discrepancy between the proportion of cases in Varro and in Plautus and Terence, for instance, means little. It should be remembered also that this discrepancy is still further increased by the nature of the drama, whose action lies chiefly in the present. While, therefore, in Plautus and Terence the proportion of customary pasts is i, 1 Latin also exhibits some similar compounds, cf. Plautus, Capt. 925, te carens dum hie fui, Poen. 1038, ut tu sis sciens, and Terence, Andr. 508, ut sis sciens. Cf. Schmalz in I. Mttller's Handb. II 2 , p. 399. s In the Greek literature, which begins not only absolutely but relatively much earlier than the Latin, the imperfect was used to narrate and describe, and Brugmann, indeed, considers this a use which goes back to Indo- European times. Later the imperfect was crowded out to a great extent by the aorist, as in Latin by the (aoristic) perfect. Cf. Brugmann in I. Mailer's Handb. II, 2 p. 183. 3 Cf. p. 171. i The power of the perfect lies in its simplicity, but when too much used this degenerates into monotony and baldness. 13 1 86 AMERICAN JOURNAL OF PHILOLOGY. and in Varro f , the historians with J probably present a juster average. The relation of this usage to the simple progressive imperfect has already been pointed out, 1 but must be repeated here for the sake of completeness. If we inject into a sentence containing a simple progressive imperfect a strong temporal contrast, e. g., if facit, sed non faciebat becomes nunc facit, olim autem non faciebat, it is at once evident how the customary past usage has developed. It has been grafted on the tense by the use of such particles and phrases, expressions which were in early Latin still so necessary that they were expressed in more than one-quarter of the cases ; or, in other words, it is the outgrowth of certain oft-recurring contexts, and is still largely dependent on the context for its full effect. Transitional cases in which the temporal contrast is to be found, but no customary past coloring, may be cited from Plautus, Rud. 1123, Dudum dimidiam petebas partum. Tr. Immo etiam nunc peto. Here the action expressed by petebas is too recent to acquire the customary past notion. 2 The progressive function caused the imperfect to lend itself more naturally than other tenses 3 to the expression of this idea. 4 Although the customary past usage was well established in the language at the period of the earliest literature, and we cannot actually trace its inception and development, I am con- vinced that it was a relatively late use of the tense by the mere fact that the language possesses such verbs as soleo, consuesco, &c, and that even as late as the period of early Latin the function seemed to need definition, cf. the frequent use of particles, &c. The small number of cases (13) which may be termed frequenta- tive indicates that this function is at once rare and in its infancy in the period of early Latin. The frequentative function is so closely related 5 to the progressive that it is but a slight step from 1 Trans. Am. Philolog. Ass., Vol. XXX, pp. 18-20. 5 Cf. Men. 729. s How strong the effect of particles on other tenses may be is to be seen in such cases as Turpilius, p. 113. I (Ribbeck), Quem olim oderat, sectabat ultro ac detinet. 4 The process was therefore analogous to that which can be actually traced in cases of the frequentative and conative uses. 5 Terence, Adel. 332-3, affords a good transitional case : iurabat . . . dicebat — (almost) ' kept swearing ' ... 'kept saying' &c, cf. p. 47 n. 1. It should THE IMPERFECT INDICA TIVE IN EARL Y LA TIN. 1 87 the latter to the former. Latin 1 seems, however, to have been unwilling to take that step. The vast number of frequentative, 2 desiderative and other secondary endings also prove that the tense was not the favorite means for the expression of the frequentative idea. Nevertheless since the progressive and fre- quentative notions are so closely related and since frequentative verbs must again and again have been used in the imperfect subject to the influence of the progressive function of particles such as saepe, etiam atgue etiam, and since finally a simple verb must often have appeared in similar situations, e. g. poscebat for poscitabat, the tense inevitably acquired at times the frequentative function. We have here, therefore, an excellent illustration of the process by which a secondary function may be grafted on a tense and the frequentative function is dependent to a greater degree than the customary past upon the influence and aid of the context. That it is of later origin is proved by its far greater rarity (see Table II). If the frequentative imperfect in early Latin is still in its infancy, the conative usage is merely foreshadowed. The fact that there are no certain instances proves that relatively too much im- portance, at least for early Latin, has been assigned to the conative imperfect by the grammars. Statistics would probably prove it rare at all periods, periphrases with conor &c, having sufficed for the expression of the conative function. The most powerful influence in moulding tense functions is context. 3 In the case of the conative function this becomes all powerful for we must be able to infer from the context that the act indicated by the tense has not been accomplished. The also be pointed out that the frequentative imperfect is very closely related to the imperfect of situation. To conceive an act as frequentative necessarily implies a vivid picture of it. (Cf. next note). It is possible, therefore, to interpret as vivid imperfects of situation such cases as Ennius, Ann. 43-4 ; Plautus, True. 506, Capt. 917, but a careful study of these has convinced me that the frequentative idea predominates. 1 In Greek, however, the imperfect was commonly used with an idea of repetition in the proper context. This use is correctly attributed by Brugmann (I. Milller's Handb. &c. II, 2 p. 184) to the similarity between the progressive and frequentative ideas as well as to the fondness for description of a re- peated act. 5 Ace. to Herbig, § 62 (after Garland?) there were probably no iterative formations in Indo-European. 8 Cf. Morris, Syntax, pp. 46, 82, &c. 1 88 AMERICAN JOURNAL OF PHILOLOGY. function thus rests upon inference from the context- The presence in the language of the verbs conor, tempto, &c, proves that the conative function, like the frequentative, was a secondary growth grafted on the tense in similar fashion, but at a later period, for we have no certain instances in early Latin. This function of the imperfect certainly originates within the period of the written language. The fact that the preponderance of the aoristic cases occurs in Plautus and Terence (see Table I) indicates that this usage was rather colloquial. This is further supported by the fact that the majority of the cases are instances of aibam, a colloquial verb, and of eram which in popular language would naturally be con- fused with/i«. In this usage, therefore, we have an instance of the colloquial weakening of a function through excessive use in certain situations, a phenomenon which is common in secondary formations, e. g. diminutives. The aoristic function is not original, but originated in the progressive usage and in that application of the progressive usage which is called the imperfect of situation. Chosen originally for graphic effect the tense was used in similar contexts so often that it lost all of this force. All the cases of aibam, for instance, are accompanied by an indirect discourse either expressed (38 cases) or understood (2 cases). The statement contained in the indirect discourse is the important thing and aibam became a colorless introductory (or inserted) formula losing all tense force. 1 If this was the case with the verb which, in colloquial Latin at least, was preeminently the mark of the indirect discourse it is natural that by analogy dicebam, when similarly employed, should have followed suit. 2 With eram the development was similar. The loss of true imperfect force, always weak in such a verb, was undoubtedly due 1 Cf. Greek iXeys, tjv <5' iyi> &c. and English (vulgar) ' sez I ' &c„ (graphic present). Brugmann (I. Muller's Handb. &c. II, 2 p. 183) denies that the Greek imperfect ever in itself denotes completion, but he cites no cases of verbs of saying. Although one might say that the tense does not denote completion, yet if there was so little difference between imperfect and aorist that in Homer metrical considerations (always a doubtful explanation) decided between them (cf. Brugmann, ibid.), Brugmann seems to go too far in dis- covering any imperfect force in his examples. The two tenses were, in such cases, practical equivalents and both were colorless pasts. 8 Rodenbusch, p. 8, assigns as a cause for the frequency of aibat in this use the impossibility of telling whether ait was present or perfect. This seems improbable. THE IMPERFECT INDICATIVE IN EARLY LATIN. 189 to the vague meaning of the verb itself. Indeed it seems probable that eram is thus but repeating a process through which the lost imperfect of the root *fu} must have passed. This lost imperfect was doubtless crowded out " by the (originally) more vivid eram which in turn has in some instances lost its force. If the aoristic usage is not original, but the product of a collo- quial weakening, we should be able to point out some transitional cases and I believe that I can cite several of this character : Plautus, Merc. 190, Eho . . . quin cavisti ne earn videret . . .? Quin,sceleste,<eam>afo/7'«dfe&w,ne earn conspiceret pater? Id. Epid. 597, Quid, ob earn rem | hanc emisti, quia tuam gnatam es ratus ? Quibus de signis agnoscebas? Pe. Nullis. Phi. Quarefiliam Credidisti nostram ?* In these cases the tense is apparently used for vivid effect (im- perfect of situation), but it is evident that the progressive function is strained and that if these same verbs were used constantly in such connections, all real imperfect force would in time be lost. This is exactly what has occurred with aibam, dicebam, and eram. The progressive function if employed in this violent fashion simply to give color to a statement, when the verbs themselves {aibam, dicebam) do not contain the statement or are vague (eram), must eventually become worn out just as the diminutive meaning has been worn out of many diminutive endings. In the shifted cases also the tense is wrenched from its proper sphere. But whereas the aoristic usage displays the tense stripped of its main characteristic, the progressive function, though still in possession of its temporal element as a tense of the past, in the shifted cases both progressive function and past time (in some instances) are taken from the tense. In those cases where the temporal element is not absolutely taken away it becomes very unimportant. This phenomenon is apparently due in the first place to the contrary-to-fact idea which is present in the context of each case, and secondly to the meaning of some of the verbs involved. In many of the cases these two reasons 1 There was no present of this root ace. to Morris, Syntax, p. 56, but cf. Lindsay, Lat. Lang., p. 490. 'Also if *bhwam <.-bam was derived from *bhu </«- in fui &c., then the fact that it was assuming a new function in composition would help to drive it out of use as an independent form, eram (originally *isom) taking its place. 3 Cf. Terence, Phorm. 298 ; Adel. 809, Eun. 700. Ennius, Fab. 339. I90 AMERICAN JOURNAL OF PHILOLOGY. are merged into one, for the verbs themselves imply a contrary- to-fact notion, e. g. debebat, oportebat, poterat (the last when representing the English might, could, &c). In Varro, L. L. 8. 65 the phrase sic Graeci . . . dicere debebant implies that the Greeks do not really so speak; so Plautus, Mil. gl., 911 Bonus vaies poteras esse implies that the person addressed is not a bonus vales. In these peculiar verbs, which in recognition of their chief function I have classified as auxiliary verbs, 1 verb- meaning coincides very closely with mode, just as in soleo, conor, &c, verb-meaning coincides closely with tense. The modal idea is all important, all other elements sink into insignificance, and the force of the tense naturally becomes elusive. 2 Let us summarize the probable history of the imperfect in early Latin. The simple, progressive imperfect represents the earliest, probably the original, usage. Of the variations of this simple usage the imperfect of the immediate past and the im- perfect of situation are most closely related to the parent use. Both of these are early variants, the latter probably Indo- European, 3 and both may be termed rather applications of the progressive function than distinct uses, since the essence of the tense remains unchanged, the immediate past usage arising from a widening of the temporal element, the imperfect of situation from a wider application of the progressive quality. Later than these two variants, but perhaps still pre-literary, arose the custom- ary past usage, the first of the wider variations from the simple progressive. This was due to the application of the tense to customary past actions, aided by the contrast between past and present. Later still and practically within the period of the earliest literature was developed the frequentative usage, due chiefly to the close resemblance between the progressive and frequentative ideas and the consequent transfer of the frequentative function to the tense. Finally appears the conative use, only foreshadowed in early Latin, its real growth falling, so far as the remains of the language permit us to infer, well within the 1 Cf. Whitney, German Grammar, § 342. 1. 8 The same power of verb-meaning has shifted, e. g., the English ought from a past to a present. Cf. idei, &c. If I understand Tobler, Uebergang zwischen Tempus und Modus (Z. f. V51kerpsych., &c, II. 47), he also con- siders the imperfect in such verbs as due to the peculiar meaning of the verbs themselves. Cf. Blase, Gesch. des Plusquamperfekts, § 3. »Cf. note. THE IMPERFECT INDICA TIVE IN EARL Y LATIN. I9I Ciceronian period. In all these uses the progressive function is more or less clearly felt, and all alike require the influence of context to bring out clearly the additional notion connected with the tense. The first real alteration in the essence of the tense appears in the aoristic usage in which the tense lost its progressive function and became a simple preterite. This usage, due to colloquial weakening, is confined in early Latin to three verbs, aidant, dicebam, and eram (with compounds). It is very early, pre- literary in fact, but later than the imperfect of situation, from which it seems to have arisen. A still greater loss of the essential features of the tense is to be seen in the shifted cases in which the temporal element, as well as the progressive, has become insignificant. This complete wrenching of the tense from its proper sphere is confined to a limited number of verbs and some phrases with eram, and is due to the influence of the pervading contrary-to-fact coloring often in combination with the meaning of the verb involved.
Subscribe to:
Post Comments (Atom)
Posts
No comments:
Post a Comment