form of our language was still Saxon, the matter was in a great measure French." These opinions indeed relate to such different periods, that they are not, strictly speaking, capable of being opposed to each other; but it is nearly evident that both are exaggerated: Dr. Hickes having probably imagined that he saw traces of a Gothic etymology in words which were, in fact, purely French; while Mr. Tyrwhitt, being misled by his own glossary of obsolete words (in which the two languages are pretty nearly balanced), has neglected to observe that the greater part of his author's text, which required no explanation, was almost solely derived from the Saxon. But, be the proportion what it may, it should seem that we ought to possess in the various existing glossaries of the Gothic and Romance dialects the means of recovering nearly all the original materials of our language.

It is true that these materials, in passing from the parent tongues into English, are likely to have undergone considerable changes in their appearance it may therefore be worth while to examine for a moment the probable nature and extent of these alterations.

Dr. Adam Smith, in his very ingenious essay on the formation of languages, has observed, that the order in which the several kinds of words (or parts

of speech) were invented, may fairly be inferred from the degree of reasoning and abstraction which was necessary to their invention: that it was a much simpler expedient to represent what grammarians call the cases of nouns, and the moods and tenses of verbs, by varying their terminations and inflections, than to invent prepositions expressive of relation in general, or auxiliary verbs conveying the very abstract ideas of existence, possession, &c. and, consequently, that all original languages will be found to be very complicated in their mechanism, and full of varieties of termination and grammatical intricacy, but extremely limited in the number of their elementary and radical words.

But although the speech of any nation, in which the paucity of its distinct words is thus supplied by the number of their inflections, may become perfectly applicable to every purpose; it is evident that two such languages cannot easily be amalgamated, because the radical words in each, having been arbitrarily chosen, will probably be very different; their respective schemes of grammar will have been formed on different analogies; and, consequently, the number of declensions and conjugations resulting from a mixture of the two would be almost infinite. When, therefore, a very close intercourse takes place between the natives of two countries, in

consequence of their commercial pursuits, or the operations of war and conquest, it is likely that they will be under the necessity of forming an intermediate language, whose grammatical construction shall be so simple as to be capable of admitting indifferently, from either of the component parts, as many words as it may from time to time become convenient to adopt. And observation will soon teach them, that this simplicity is easily attainable by means of the prepositions and auxiliary verbs, which are capable of being substituted for all the varieties of the ancient declensions and conjuga tions.

Whether this theory be universally true or not, it is perfectly evident that the expedient here mentioned has been adopted in the formation of all the mixed European languages; from the Latin (which is supposed to be a compound of the Greek and ancient Tuscan), to that lingua-franca, of which the various dialects are spoken along both coasts of the Mediterranean: and that in Italy, France, and England, the scheme and mechanism of grammar has become progressively more simple, in proportion to the number of heterogeneous parts of which the respective languages have been composed.

It is remarkable that Dr. Johnson, though he has noticed, and even accurately described the

gradations by which the Saxon was insensibly melted into the English language, has considered the cause of these changes as inexplicable. "The adulteration of the Saxon tongue (says he), by a mixture of the Norman, becomes apparent; yet it is not so much changed by the admixture of new words, which might be imputed to commerce with the Continent, as by changes of its own form and terminations ; for which no reason can be given." The reader, however, who shall take even a cursory survey of the extracts which gave rise to this remark, will pro bably be convinced, that these changes in the Saxon consist solely in the extinction of its ancient grammatical inflections, and that they are exactly similar to the alterations by which the Latin was gradually transformed into the several Romance dialects.

But it is evident that, although the new scheme of grammar was perfectly simple, and composed of few elements, yet the precise and definite use of those elements could not be suddenly established. In employing our prepositions, for instance, though we are seldom aware of the nice shades of discrimination which we observe, till the remark is forced upon us by some striking violation of the usual practice, it is certain that mere reasoning and analogy would prove very insufficient guides. When

our neighbours the Scots talk of going till instead of to a place, or of asking a question at rather than of a man, we are immediately startled; without reflecting that our own practice is only founded on convention and habit. Amongst our elder writers the use of the prepositions was, as might be expected, extremely vague and indefinite.

With the auxiliary verbs there was less difficulty; indeed the Normans, having only two words of this class, were accustomed to apply them to a greater variety of purposes than was usual with the Saxons. Hence perhaps arose the transitive use of the verb do, which is so frequent in our early writers; as in do make (faire faire) &c. ; and the old Scotish poets carry their imitation of the French still farther, so as to use doing make; done make; &c. an employ. ment of the verb which I do not recollect to have seen in English.

It is unnecessary to pursue these remarks any farther, because the reader will find in Mr. Tyrwhitt's "Essay on the Language and Versification of Chaucer," a complete analysis of our grammar, as it subsisted during the thirteenth and fourteenth centuries. Indeed, from what has been already premised, we are perhaps authorized to conclude that, notwithstanding the pretended fluctuation of speech, a fluctuation which has been oftener supposed