In any case, therefore, if we have a complete knowledge of a language--that is to say, of the words of which it consists--we possess a mass of facts by aid of which to judge of the claims to authenticity of every work in which that language is embodied. And if, in addition to a knowledge of its vocabulary, the laws of its construction also, and the nicest proprieties of its syntax and style are known; and if, moreover, the changes that have taken place from age to age in the sense of words, and in modes of expression, are understood, we then possess ample and exact _data_ with which to compare any book that pretends to antiquity. A writer who employs his native language must be expected to conform himself to its usages; and we should find him adhering, more or less strictly, to the peculiarities of the age in which he writes: his vocabulary, moreover, will include that compass of words which his subject demands, and which the language affords. It is true that such a degree of skill in a dead language may be acquired as may enable a writer to use it with so exact a propriety as shall deceive, or at least perplex, even the most accomplished scholars. But the difficulty of avoiding every phrase of later origin, and all modern senses of those words which are continually passing from a literal to a metaphorical meaning, is so great, as to leave the chances of escaping detection extremely small. Yet, as such a chance still remains within the range of possibility, this line of evidence cannot be reckoned absolutely conclusive, but must only be employed as subsidiary to those other evidences that bear upon questions of authenticity. The minute changes which are continually taking place in most languages, and the history of which, when known, serves often to ascertain the date of ancient books, are of two kinds; namely, those which result necessarily from actual changes in the objects represented by words, and those which are mere changes in the use and proprieties of language itself. Language being a mirror, reflecting all the communicable notions of the people who use it, every mutation in the condition of the people must bring with it, either new terms, or new combinations of words; and as the particular circumstances which introduce such additions or alterations are often known, their occurrence in an author may serve to fix the date of the book, almost with certainty.