such amalgamation of the auxiliary with the principal verb-no formation of an inflected future-has taken place. The result is that the Germans are left with "werden," and we are obliged to do the best we can with "shall" and "will.” I think it has been shown, however, that the English application of these two verbs, though it may be difficult to acquire and preserve incorrupt, is neither irrational in its origin, nor deficient in force and precision in its use.

APPENDIX.

66

(A.)

In the sentence of Cicero (Ep. ix. 15), “Nam mihi scito jam "à regibus allatas esse litteras, quibus mihi gratias agant, quòd se meâ sententiâ reges appellaverim," the reflexive pronoun is used, because the dependent sentence is placed in the mouth of the persons addressing Cicero, and is in fact a quotation (compare Zumpt, Lat. Gramm. § 550, n. 1). The case is, however, a remarkable one, inasmuch as the quotation is in an oblique form, and the use of the first person, “appellaverim," seems inconsistent with a reflexive pronoun in the third. In Cornelius Nepos (Themistocles, c. 8), "Domino navis, qui sit aperit, multa pollicitus si se conservâsset," we could not translate " se" by himself," because our reflexive pronoun would relate to the subject of the verb "conservâsset." The principle is in fact the same as in the passage of Cicero.

66

66

The sentence of Livy (L. i. c. 54) is as follows: (Sextus Tarquinius)" è suis unum sciscitatum Romam ad patrem "mittit quidnam se facere vellet, quandoquidem ut omnia (6 unus Gabiis posset, ei Dii dedissent." Zumpt assumes that the common usage would require "sibi;" but I believe that the proper mode of translating this passage is to refer "ei" to the father-Sextus meant, according to Livy, to imply that everything was at his father's disposal. I conceive that the father, not Sextus, is the subject of the verb "posset." If this be so, the pronouns would appear perfectly regular.

The following sentence of Cæsar is again remarkable :— "Responderunt (scil. Sicambri) Populi R. imperium Rhenum "finire: si se invito Germanos in Galliam transire non

The

"potestatis trans Rhenum postularet" (L. iv. c. 16). Sicambri, in whose mouth the sentence is placed, are supposed to be remonstrating with Cæsar himself, to whom "se" and "sui" relate, as the immediate nominative of "existi"maret 99 and "postularet," as well as the narrator who quotes the speech. I am far from pretending to explain all the anomalies and difficulties which appear to exist in certain cases with regard to the Latin reflexive pronoun.

Professor Key (Latin Gr. p. 219) refers to the speech of Ariovistus (Cæsar, de B. G. i. 36), and very justly says that in it "there is much freedom in the use of these pronouns." The whole speech is in an oblique form: in the passage about the middle, "Magnam Cæsarem injuriam facere, qui suo adventu vectigalia sibi deteriora faceret," suo points to the immediate subject of the subordinate verb, sibi to the speaker. So again towards the end, "neminem sibi nisi "sua pernicie contendisse," se and suâ refer to different perIt may be said indeed that "suâ pernicie " is a sort of adverbial formula, irrespective of person. The same kind of explanation might be applied to the "per-se" and "inter-se" quoted by Professor Key from Cicero, as well as to the phrases "suo-nomine" and " suo-jure;" but in truth the difficulty of arriving at any certain rule remains much as it was before.

sons.

In the passage of Cicero, De Orat. lib. i. c. 54, "Quod quum interrogatus Socrates esset, respondit, sese meruisse, "ut amplissimis honoribus et præmiis decoraretur, et ut ei "victus quotidianus in Prytaneo publice præberetur," "sese" and "ei" clearly refer to the same person-Socrates; nor do I find in Orelli (vol. i. p. 245) any various reading. Compare Zumpt as above; Diez, Romanische Gramm. b. iii. s. 54.

Another instance presents itself in the first Philippic, c. 10:-" Ut-hujus tamen diei vocem testem reipublicæ relinquam meæ perpetuæ erga se voluntatis." I see no reason for the use of se here.

66

(B.)

BOSWORTH (A. S. Dict. p. cxc.) calls the two ordinary A. S. tenses "the indefinite and the perfect.” Dr. Prichard (ib. p.171) gives the future form of the Welsh verb-substantive thus :— Sing. bydhav, bydhi, bŷdh.

Plur. bydhwn, bydhwch, bydhant.

Zeuss (Gramm. Celt. vol. i. p. 482) says of the Irish future, Sing. 1 & 2 pers. non obvia exempla;" the third person sing. he gives as bieid, bied, and sometimes bid. He states that "beth" was the Cornish and Armoric form for the future and subjunctive, as well as root of the verb (p. 539). From this writer's mode of speaking I should infer that in his opinion a future tense, properly so called, had originally belonged to the system of the Celtic verb generally, but I am not competent to discuss such a question (see p. 411).

The Attic use of eiμ, ibo, with its future sense, may be held to have some connection with the future sense of one form of the verb-substantive, though it is not the same (see Grimm, Gesch. der Deutschen Spr. b. ii. s. 892). At any rate, the wide-spread tendency to assign this future meaning to the form which corresponds with our “be” is very remarkable, and must go back to remote times in the history of all these kindred languages. It appears to give great additional probability to the conjecture that the syllable "bo" in the Latin futures of the 1st and 2nd conjugations, as well as in those of some other verbs, was in its origin only an application of this very root of the verb-substantive as a suffix. Fui, fueram, forem, fuere, or fore, are of course all derived from this root, and it is curious that the infinitive “fore” still retains its future sense as equivalent to "futurum esse." I think this theory at any rate more probable than Professor Key's conjecture that in the Latin conjugation the suffixes ēba and eb may have some connection "with the verb habe, "have, which is so common an auxiliary in all languages " (see Latin Grammar, p. 64, note).

We know from such forms as "scibo" ("Nemo ex me 66 scibit," ," Terent. Phormio. vol. i. 38) and "ibo," that the ter

mination "bo," for the future, probably extended much further than it appears to do in our ordinary Latin Grammars. Compare Zumpt, Lat. Gramm. § 215; Facciolati, in vv. scio et eo. Bopp (Comparative Grammar, Transl. p. 889, § 662) assumes it as certain that the 3rd and 4th conjugations in Latin did originally form their futures in "bo." The ordinary futures in "am" are evidently allied to the subjunctive (compare Philolog. Museum, vol. ii. p. 218). It will be observed from what is said in the text that the subjunctive was used for the future occasionally by Ulfilas, and I believe that the same relation is to be traced in Sanscrit. See Bopp, Compar. Gramm. Transl. vol. ii. pp. 887, 891; Zeuss, Gramm. Celt. vol. i. p. 539; Grimm, Deutsche Gramm. b. iv. s. 177, n. 2; Trans. of Philological Society, 1845, No. 38; 1846, No. 44. Mr. Guest, in these Transactions (vol. ii. p. 223), tells us that the verb "be" was long retained for the expression of future time in English, more particularly in the North, and he quotes examples from Lyndsay and other writers in support of this view.

On the verb "be" in general, and its equivalents, the reader may consult Mr. Francis Newman's paper in the Classical Museum, No. xxv. p. 254.

(C.)

THE Scotch and North country forms are aw," "awin," ain," "awingis" (debts), and approach still nearer than our own to the Gothic, of which the first person indicative was "aih" (see Jameson's Scottish Dict. in vv.). Grimm (Gesch. der D. Spr. b. ii. s. 905) considers "aih" is the præterite of 66 eigan,' ""to labour" or "make"-" schaffen." The word, therefore, which originally meant "I have made" or quired by my own labour," assumed, like Kékтnμai, the present sense of "I possess," or "have as my own."

66

ac

This verb, according to Grimm's view, is thus what he calls a verschobenes præteritum," or, as Dr. Latham denominates it, “a transformed præterite," of which I have had