phrases are at least not incorrect, whereas in most of these cases they are positively at variance with the usage of the language; and what is called elegant, has the still higher merit of being a correct, often the only correct, phrase.

The order of words in Latin, though a subject of no great difficulty, is so utterly neglected in almost every school, and at the same time is so important a part of the language, that we cannot pass it by without remark. In this, however, as in other parts of the Gymnasium, Dr. Crombie while he may claim credit for a distinct exposition of some of the main principles, must also plead guilty to the charge of repeatedly, nay almost uniformly, violating the very laws he himself has laid down. In Latin,' observes Dr. Crombie, 'the variety of termination in nouns and verbs enables the writer or speaker to place the words in whatever order his reason, his feelings, or his imagination may suggest.' Again he says: The starving wretch beholding a piece of bread would not say, Da mihi рапет, but Panem mihi da. This is the natural order-the order, in fact, which if the language permitted, the feelings of every man would irresistibly prescribe. When Nisus, in his impatient eagerness to save the life of his friend, exclaimed,

66

Me, me, adsum qui feci, in me convertite ferrum," he spoke a language consentaneous to his feelings, in a moment of the most agonizing apprehension.' The first half of page xxxi in the preface is equally distinct and important. It is somewhat singular after this to find such a sentence as the following: If the words of one clause are kept distinct from those of another, they may be interchanged among themselves as the writer may think fit. Thus, Juvenem ego vidi qui seros amores desisset, or Ego juvenem vidi, qui seros desisset amores, or Vidi juvenem ego, qui desisset seros amores.' Against any such licence we must most decidedly protest; nay, we are ready to contend, that the order of Latin words admits of being reduced to principle, as strictly as the order in our own, or any modern tongue. It is true, that occasionally one or two words may be placed indifferently one or another first; but the same liberty is also allowed in every language. Questions, however, of this kind are more easily decided by reference to examples; we will, therefore, proceed to the last division, English passages given by Dr. Crombie for translation into Latin, together with the Clavis. 'These exercises,' we quote from the preface, though chiefly extracted from the Latin classics, are not to be regarded as mere translations. He (Dr. Crombie) has abridged the original wherever it was necessary, in order to adapt the length

of the exercise to the scholar; and he has on the contrary occasionally introduced passages which might serve to illustrate the critical observations.' Had there been any system of arrangement in the critical observations of the Gymnasium, it would have been some excuse, though a very insufficient one, for the introduction of original passages, where the memory might fail in suggesting to Dr. Crombie appropriate examples from ancient writers. As it is, we have simply to condemn the presumption, we might almost call it, of giving modern Latin as a guide in composition to the student. But Dr. Crombie has gone much farther than his preface would lead one to think. We have compared a considerable number of passages in the Clavis with the text of Cicero, Livy, and Cæsar, from whose writings they have been evidently borrowed; and we have not found one single paragraph where the two agree. Nor can Dr. Crombie plead that the necessity of abridgment has occasioned the alteration, for in many of these passages the sentences in the English exercise are as full as in the original Latin. The liberty which Dr. Crombie has thus assumed, is one which ought not to be allowed to the most perfect Latin scholar that has appeared since Latin ceased to be a spoken language. Nay, it has always been thought prudent to restrict the young student to the study of the Latin language, as it appears in writers not later than the Augustan age. The result of Dr. Crombie's imprudence is such, that every page, almost every sentence of the Clavis, is in some respect or other at variance with the idiom of the Latin language; more especially with the principles which govern the order of words. In this condemnation, we mean not to pass any other censure on Dr. Crombie's latinity, than would be applicable to the Latin of any other modern whatever. All we say, and it is fully sufficient for our argument, is, that the latinity of Dr. Crombie is very inferior to that of Cicero. This general statement of the want of judgment shown by Dr. Crombie must be almost sufficient for the reader; we will, however, confirm our assertions by a few passages from the Clavis. In page 21, we have an abridgment from Livy, 2, 12, of the story of Mucius Scævola. His short address to the senate is given by Livy in these words: ‹ Transire Tiberim, inquit, Patres, et intrare si possim, castra hostium volo; non praedo nec populationum invicem ultor. Majus, si Dii juvant, in animo est facinus. Approbant Patres: abdito intra vestem ferro proficiscitur.' The rival version of Dr. Crombie is: Patres, Tiberim transire volo, et si possim hostium castra inire, attamen haud praedo sed patriae liberator. Praeda non mihi est in animo; si Dii juvant, majora

:

mente agito. Patres consilium approbant; adeoque, gladio sub veste abdito, proficiscitur.' The brevity in the first place is on the side of the original; and secondly, in the order of words how infinitely superior is Livy! The two words, 'Transire Tiberim' at once open the intentions of the youth, as far as he intended they should be known. The word majus, which alludes to the more secret part of his intentions, occupies a position worthy of it at the beginning of its clause, and still farther is it strengthened by the insertion of the parenthetic si Dii juvant, while the expression in animo gains the emphasis it requires from the insertion of the enclitic est immediately after it. Dr. Crombie on the other hand, independently of other inelegancies, has destroyed the vigour of the passage by his numerous conjunctions, attamen, sed, adeoque. The phrase too 'Praeda non mihi est in animo,' is somewhat ludicrous from the position of the negative before mihi, and of est after it, for the power thus conferred upon the pronoun is not very complimentary to the senate: 'Booty is not an object with me, as it is with you.' Before we go on, it may be observed, that Dr. Crombie in his introduction (p. xlviii.) has not fully stated the law for arranging antithetical words, when he says, that they should be as close as possible together. He should have added, or else as far apart as possible,' that is, the one at the commencement, the other at the end of a clause; for, to quote his own words, (p. xxxii.) the most conspicuous words in every sentence are the first and last. By the former our attention is excited; and on the latter it rests.' We shall now again see the superiority of Livy over Crombie. The former has (where Mucius addresses Porsena)-Nec ad mortem minus animi est, quam fuit ad caedem: while the latter gives us: Nec minus paratus est animus ad mortem ferendam; quam ad te fuit interficiendum. In the very next line, the contrast is still more unfavourable to Dr. Crombie: Et facere et pati fortia, Romanum est, says Livy; the other, dropping the all-important et, and placing fortia where it is altogether inadmissible, 'Facere et fortia pati Romanum est. Lastly, when the Roman thrusts his hand into the flame, how superior is the En tibi ut sentias of Livy to the tame Vide of Crombie!

6

In the passage (pp. 23, 24, of the Clavis) abridged from Livy 8, 7. on the death of T. Manlius, we see similar errors, but we will merely quote the phrase, Provocatus a Tusculano interfeci, et tibi spolia attuli,' to observe that the last word is both badly chosen and in the wrong tense. Livy has porto. We may be allowed also to ask upon what principle the words, especially quid and peccatum, are arranged in this

6

sentence:-Itaque, ne respublica ex tuo facto quid capiat detrimenti, peccatum capite te luere oportet. In more than one of the passages already quoted, it has been observed how unfortunate Dr. Crombie is in the position of antithetical words. The following, with many others, might be added to the list:-Clavis, p. 99, Spernere quam regnum accipere.'—p. 50. Nec certamen fuit de vita sed de imperio.' -p. 81. Anne copiae vobis imminutae sunt, an illorum auctae?'-p. 108. Ali quam eum alere.' The necessity for brevity renders it impossible to explain our meaning at length. We will merely state that regnum,' ' certamen fuit,'copiae,'' eum,' in their several sentences occupy places to which they have no claim. Another general principle which Dr. Crombie has forgotten to observe is the law regarding the possessive pronouns (and we may include with them genitives in general); we mean, that when emphatic, they precede their nouns, and the contrary when without emphasis. There are many errors, too, of a rhetorical nature which we should not have expected from Dr. Crombie, judging from the elegance of his English style. We may, in particular, point out a passage from the speech of Hannibal to Scipio, where the rhetorical beauty of the original, 'jam aetas, jam secundae res, jam adversae, &c.' is destroyed in Dr. Crombie's version by the change of the first clause into nunc senem, &c. The whole of this beautiful speech, we may add, is much disfigured by the variations introduced by Dr. Crombie. Much of this indeed may be owing to his English version, which, in many parts, is loose and inaccurate, and gives neither the spirit nor meaning of the original. In the second sentence, for instance, Hannibal, after congratulating himself on having to address such a man as Scipio, observes, that Scipio, too, has reason to be proud of having Hannibal in such a situation before him. In the original the expression tibi quoque occupies a position suited to this antithesis; but in Dr. Crombie's English translation this connexion between the two sentences disappears, and again his retranslation into Latin has the inconsistency of retaining tibi emphatically at the commencement of the sentence, and at the same time omitting quoque. But above all, the Latin of Dr. Crombie is remarkable for an inelegant, we may even say, incorrect, use of the little word et, which he often inserts when not wanted, and omits when necessary. More than one instance of such errors occur in the very speech we have been referring to. Thus, in Livy, we have ut et vos Italiae et nos Africae imperio contenti essemus; but in Dr. Crombie, ut vos Italiae, &c. In

Livy we have quod ego fui ad Trasimenum, ad Cannas, id tu hodie es. Dr. Crombie's version has et ad Cunnas. We might point out four more examples in the speech of Hannibal and Scipio's reply. The faulty insertion of et is particularly remarkable in those places where partitive phrases occur. The very title-page affords an instance of this, where partim datae, partim redditae, without any conjunction, would be more consonant with the practice of the best Latin writers. In the same way we would propose the omission of the conjunction in the following passages:-p. 13.-Aliud dicit et aliud cogitat. p. 69.-Ex Europa in Asiam et ex Asia in Europam transis. p. 81.-Flumen Dubis paene totum oppidum cingit, et reliquum spatium mons continet.

The questions that arise in the examination of the Clavis are so numerous, that we have not room to enter fully upon them; and we are afraid that our condemnation of the whole may be thought to require more evidence than we have produced. This fear, however, will not induce us to soften that condemnation, for we are sincerely of opinion, that, as an exercise-book, Dr. Crombie's Gymnasium will be found to be exceedingly injurious to a pupil's progress. But had there been no other objection, the very price of the work must always be an obstacle to an extensive use of it in schools. That this evil, however, might be remedied we have no doubt; for, to say nothing of the useless repetitions already noticed, without the omission of a single word the two volumes might easily be reduced by a more economical form of printing to the compass of one, and that not a large octavo volume. We have compared, for instance, the amount of letterpress in the two volumes of the Gymnasium with that in the History of Greece, published by the Society for the Diffusion of Useful Knowledge, and the result is, that the latter work, with its two hundred and ninety octavo pages, contains more words than Dr. Crombie's two volumes of 833 8vo. pages.

SCHOOL POETRY.

The Speaker, or Miscellaneous Pieces, selected from the best English writers, &c. by William Enfield, LL.D. Genuine Edition.

Beauties of Eminent Writers, selected and arranged for the Instruction of Youth, by William Scott. Revised by William Angus, A.M. Edinburgh. Fourteenth Edition.