signs of dissolution, they are, on the contrary, like the Tithonus of mythology, the Strulbrugs of Laputa, or, lastly, such monsters as Frankenstein, very liable to the causes of death, but utterly unable to die. Hence, in language, _whatever is, is right_; a fact which, taken by itself, gives great value to the historical method of inquiry, and leaves little to the _[`a] priori_ considerations of logic. But, on the other hand, there is a limit in logical regularity, which language _never_ oversteps: and as {xvii} long as this is the case, the study of the logical standard of what language is in its normal form must go hand in hand with the study of the processes that deflect it. The investigation of the irregularities of language--and be it remembered that almost all change implies original irregularity--is analogous to the investigation of fallacies in logic. It is the comparison between the rule and the practice, with this difference, that in language the practice can change the rule, which in logic is impossible. I am sure that these remarks are necessary in order to anticipate objections that may be raised against certain statements laid down in the syntax. I often write as if I took no account of the historical evidence, in respect to particular uses of particular words. I do so, not because I undervalue that department of philology, but because it is out of place. To show that one or more writers, generally correct, have used a particular expression is to show that they speak, in a few instances, as the vulgar speak in many. To show that the vulgar use one expression for another is to show that two ideas are sufficiently allied to be expressed in the same manner: in other words, the historical fact is accompanied by a logical explanation; and the historical deviation is measured by a logical standard. I am not desirous of sacrificing a truth to an antithesis, but so certain is language to change from logical accuracy to logical licence, and, at the same time, so certain is language, when so changed, to be {xviii} just as intelligible as before, that I venture upon asserting that, not only _whatever is, is right_, but also, that in many cases, _whatever was, was wrong_. There is an antagonism, between logic and practice; and the phenomena on both sides must be studied. * * * * * {xix} CONTENTS. PART I. GENERAL ETHNOLOGICAL RELATIONS OF THE ENGLISH LANGUAGE. CHAPTER I. GERMANIC ORIGIN OF THE ENGLISH LANGUAGE--DATE. SECTION PAGE