tenses of the two sets of verbs: for while we find that s- often effects a primary variation in the consonant-verb, we observe that this insertion never takes place in the vowel-verb except in composite forms, or in those verbs which neglect the vowel characteristic in the formation of their perfects. The only tense in the consonant-verb, which can be considered as a composite form, is the imperfect; but the future does not correspond to this, as is the case in the vowel-verbs. Verbs in io partially approximate to the consonant-verbs in this respect.

§ 6. General scheme of Tenses in the Latin Verb.

The following table will show the organic formations and agglutinate additions, by which the tenses of the Latin verb are constructed from the crude form. With regard to the perfect indicative, it is necessary to premise that, in addition to the parathetic or agglutinate combination with -fui, which will be mentioned presently, there are two forms in common use: one which may be considered as a regular perfect, exactly corresponding to fui fufui, with a reduplication either expressed or implied, and with the -s or guttural of the affix represented, as in fui, by i or is; and another, which may be regarded as an aorist in -si, although the inflexions of the persons exhibit the same retention of i or is as the regular perfect, and therefore presume the addition of a repeated s or sa ra, which appears in the pluperfect.

=

§ 7. Verbs which may be regarded as Parathetic Compounds.

The

The fourteenth chapter will show that the most remarkable feature in the pathology of the Latin language is the prevalent tendency to abbreviation by which it is characterised. Among many instances of this, we may especially advert to the practice of prefixing the crude form of one verb to some complete inflexion of another. Every one knows the meaning of such compounds as vide-licet (= videre licet,) sci-licet (= scire licet), pate-facio (= patere facio), ven-eo (= venum eo, comp. venum-do, on the analogy of per-eo, per-do)', &c. There is a distinct class of verbs in -so, which are undoubtedly compounds of the same kind, as will appear from an examination of a few instances. verb si-n-o has for its perfect sivi; and it is obvious that the n in the present is only a fulcrum of the same nature as that in tem-no, root tem-; πí-vw, root πι-, &c. Now the verbs in -so, to which I refer, such as arcesso, capesso, incipesso, lacesso, petesso, quæro, &c., all form their perfect in -sivi. We might therefore suppose a priori, that the termination was nothing but the verb sino. But this is rendered almost certain by the meaning of arcesso or accerso, which is simply accedere sino2, "I cause to approach," i. e. "I send for." Similarly, capesso = capere sino, "I let myself take," i. e. "I undertake," facesso = facere sino, "I let myself make," i. e. "I set about," lacesso = lacere sino, "I let myself touch," i. e. "I provoke or irritate," &c. The infinitive of in-quam (above, p. 112) does not exist; but there can be little doubt that it is involved in quæ-ro or quæ-so, which means "I cause to speak," i. e. "I inquire." That quae-so was an actual form of quae-ro may be seen from the passages of Ennius quoted by Festus (p. 258, Müller):

1 The true orthography, ven-dico for vindico, furnishes a third illustration of ven-do, i. e.

ven-eo, "I go for sale" I am sold.

=

ven-do, or venum-do, "I give for sale"= I sell.

ven-dico, "I declare for sale" = I claim.

2 I am not aware that any other scholar has suggested this explanation. Müller (ad Fest. p. 320) thinks that arcesso is the inchoative of arceo = accieo: but, in the first place, the reading in Festus is by no means certain (Huschke's arce dantur being, I think, an almost necessary correction); and secondly, this would leave accerso unexplained.

Ostia munita est; idem loca navibu' pulchris

Munda facit, nautisque mari quæsentibu' vitam (Annal. II.). Ducit me uxorem liberorum sibi quæsendum gratia (Cresphont.). Liberum quæsendum causa familiæ matrem tux (Andromed.). These parathetic compounds with sino, so, sivi, are analogous to the Hebrew conjugations in Pihel and Hiph"hil. Sometimes the causative sense refers to the object, as in arcesso, "I cause him to come," quae-ro, "I cause him to speak." Sometimes it is reflexive, as in the conjugation Hithpa"hel; thus, we have facesso, "I let myself do it-I set about it," &c. Pihel and Hiph"hil only differ as ἐτύπην differs from ἐτύφθην, according to the explanation which I have given of these tenses (New Crat. § 382). We shall see below (§ 15), that the same explanation applies to the infinitives in -assere.

§ 8. Tenses of the Vowel-verbs which are combinations of the same kind.

Most of the tenses of the Latin vowel-verb seem to be composite forms of the same kind with those to which I have just referred; and the complete verbal inflexion, to which the crude =form of the particular verb is prefixed, is no other than a tense of the verb of existence fu-, Lithuan. bu-, Sanscrit bhú- (see Bopp, Vergl. Gram. vierte Abtheil. pp. iv. and 804). This verb, as we have seen, expresses "beginning of being," or "coming = into being," like the Greek yiyvoua. It is therefore well calculated to perform the functions of an auxiliary in the relation of time. For ama-bam ama-e-fiam "I became to love," "I was loving;" ama-bo = ama-fio = "I am coming into love," = "I am about to love;" ama-vi = ama-fui = "I have come into love," "I have loved," &c.

=

The vowel-verb has a present tense which preserves throughout the vowel of the crude form. From this is derived, with the addition of the element i, the present subjunctive, as it is called; and from that, by the insertion of s-, the imperfect of the same mood. Thus we have amêm=ama-im, amarem=amasem =ama-sa-im; moneâm=mone-yam, monerem=monesem=monesyam, &c. That i was the characteristic of the secondary or dependent mood is clear from the old forms du-im (dém), temperim, ed-im, verber-im, car-im, &c., which, however, are abbreviations from du-yam, ed-yam, &c. Comp. sim with the older

form siem, and didoui, &c. with didoinv, &c. The i is absorbed or included in moneam=mone-yam, legam-leg-yam, &c.; just as we have nav-ális for navi-alis, fin-ális for fini-alis, &c. (Benary, Römische Lautlehre, p. 95.) These are the only tenses which are formed by pronominal or organic additions to the root of the verb. Every other tense of the vowel-verb is a compound of the crude form of the verb and some tense of fuor bhu-.

The futures of the vowel-verbs end in -bo, -bis, -bit, &c., with which we may compare fio, fis, fit, &c. The imperfect, which must be considered as an indefinite tense corresponding to the future, ends in -ébam, -ébas, -ébat, &c., where the initial must be regarded as an augment; for as reg'-ébat is the imperfect of the consonant-verb reg'o, not regebat, and as audi-ébat is the imperfect of aud-io1, though audi-bit was the old future, it is clear that the suffix of the imperfect had something which did not belong to the crude form, but to the termination itself; it must therefore have been an augment, or the prefix which marks past time (see Benary, l. c. p. 29).

The perfect of the vowel-verbs is terminated by -vi or -ui. If we had any doubt as to the origin of this suffix, it would be removed by the analogy of pot-ui for pot-fui=potis-fui. Accordingly, ama-vi (=ama-ui), mon-ui, audi-vi (=audi-ui), are simply ama-fui= amare-fui, mon-fui = monere-fui, and audifui=audire-fui.

Similarly, with regard to the tenses derived from the perfect, we find that the terminations repeat all the derivatives of fui: thus, ama-uero=ama-fuero; ama-uisses=ama-fuisses, &c. It will be observed that the ƒ of fio and fui never appears in these agglutinate combinations. The explanation of this involves some facts of considerable importance.

We have seen above (p. 242) that the Latin ƒ involves a guttural as well as a labial, and that the v, which formed a part of the sound, had a tendency to pass into b (p. 240). If, then, which seems to be the case, the long vowel, which always forms the link of communication in this parathesis, absorbed and included the guttural part of the f (New Crat. § 116), the re

1 Virgil has lenibat (Æn. VI. 468) and polibant (VIII. 436); but these must be considered as poetical abbreviations.

maining labial would necessarily appear as b, except in the perfect, where it would subside into the u, just as fuvit itself became fuit. In general we observe that, with the exception of the three or four words ending in the verbal stem fer (furci-fer, luci-fer, &c.), the letter f does not appear among Latin terminations; and as the terminations -ber, -bra, -brum, -bulum are manifestly equivalent in meaning to -cer, -crum, -culum, it is reasonable to conclude that these formations begin with letters which represent the divergent articulations of the compound for F (see New Crat. § 267).

§ 9. Organic Derivation of the Tenses in the Consonant-verb. The consonant verb, on the other hand, forms all its tenses, except the imperfect1, by a regular deduction from its own root. Thus we have reg'o [old fut. reg-so], 1 aor. [e]-reg-si; subjunct. pres. or precative, regam=regyam, regas=regyas, or, in a softer form, regês=rege-is, &c.; subj. imperf. or optat. regerem=regesyam; subj. perf. reg-se-ro=reg-se-sim; subj. plup. regsissem= reg-si-se-syam. If we might draw an inference from the forms facsit, &c., which we find in old Latin, and from fefakust, &c., which appear in Oscan, we should conclude that the Italian consonant-verb originally possessed a complete establishment of definite and indefinite tenses, formed from the root by pronominal or organic addition, or by prefixing augments and reduplications after the manner of the genuine Greek and Sanscrit verbs. For example's sake, we may suppose the following scheme of tenses: root pag, pres. pa-n-go-m, impf. [e]-pangam, fut. pan-g-sim, 1 aor. e-pangsim, perf. pe-pigi-m, pluperf. pe-pige-sam, subj. pres. pangyam, subj. imp. pangesyam, subjunct. perf. pepige-sim or pangse-sim, subj. pluperf. (derived from this) pepigise-syam or pang-si-se-syam.

§ 10. Auxiliary Tenses of the Passive Voice.

In the passive voice, those tenses, which in the active depend upon fui and its derivatives, are expressed by the passive participle and the tenses of e-sum. The other tenses construct the

1 The loss of the imperfect, and the substitution of a compound tense, is accounted for by the practice of omitting the augment. Without this prefix the regular imperfect does not differ from the present.