cal meanings of a word. "As all epithets," he remarks, "employed to distinguish qualities perceiv"able only by intellect, were originally applied to

66

objects of sense, the primary words in all lan

guages belong to them; and are, therefore, applied transitively, though not always figuratively, "to objects of intellect or imagination.” * The distinction appears to me to be equally just and important; and as the epithet transitive expresses clearly and happily the idea which I have been attempting to convey by the preceding illustration, I shall make no scruple to adopt it in preference to figurative or metaphorical, wherever I may find it better adapted to my purpose, in the farther prosecution of this subject. It may not be altogether superfluous to add, that I use the word transitive as the generic term, and metaphorical as the specific; every metaphor being necessarily a transitive expression, although there are many transitive expressions which can, with no propriety, be said to be metaphorical.

A French author of the highest rank, both as a mathematician and as a philosopher (M. D'Alembert), had plainly the same distinction in view when he observed, that, beside the appropriate and the figurative meanings of a word, there is another (somewhat intermediate between the former two), which may be called its meaning par extension. † In the choice of this phrase, he has certainly been

+ Analyt. Inquiry, &c. p. 11. 3d edition.

†The same phrase is used by M. Du Marsais in his ingenious Treatise on Tropes. See, in particular, the second part, article Catachrese

less fortunate than Mr Knight; but, as he has enlarged upon his idea at some length, and with his usual perspicuity and precision, I shall borrow a few of his leading remarks, as the best comment I can offer on what has been already stated; taking the liberty only to substitute in my version the epithet transitive, instead of the phrase par extension, wherever the latter may occur in the original.

"Grammarians are accustomed to distinguish "two sorts of meaning in words; first, the literal, original, or primitive meaning; and, secondly, "the figurative or metaphorical meaning, in which "the former is transferred to an object to which it " is not naturally adapted. In the phrases, for ex"ample, l'éclat de la lumière, and l'éclat de la ver"tu, the word éclat is first employed literally, and "afterwards figuratively. But, besides these, there "is a sort of intermediate meaning, which

66

may be distinguished by the epithet transitive. Thus, "when I say, l'éclat de la lumière, l'éclat du son, "l'éclat de la vertu, the word éclat is applied transitively from light to noise; from the sense of sight, to which it properly belongs, to that of hearing, with which it has no original connection. It "would, at the same time, be incorrect to say, that the

66

66

66

phrase l'éclat du son is figurative; inasmuch as "this last epithet implies the application to some "intellectual notion, of a word at first appropriated "to an object of the external senses."

After illustrating this criticism by various other examples, the author proceeds thus: "There is "not, perhaps, in the French language, a single

"word susceptible of various interpretations, of which the different meanings may not all be "traced from one common root, by examining the "manner in which the radical idea has passed, by

slight gradations, into the other senses in which "the word is employed: And it would, in my opi“nion, be an undertaking equally philosophical "and useful, to mark, in a dictionary, all the pos"sible shades of signification belonging to the same expression, and to exhibit, in succession,

the easy transitions by which the mind might "have proceeded from the first to the last term of "the series." *

In addition to these excellent remarks (which I do not recollect to have seen referred to by any succeeding writer), I have to observe farther, that, among the innumerable applications of language which fall under the general title of transitive, there are many which are the result of local or of casual associations; while others have their origin in the constituent principles of human nature, or in the universal circumstances of the human race. The former seem to have been the transitions which D'Alembert had in his view in the foregoing quotation; and to trace them belongs properly to the compilers of etymological and critical dictionaries. The latter form a most interesting object of examination to all who prosecute the study of the Human Mind; more particularly to those who wish to investigate the principles of philosophical criticism.

* Eclaircissemens sur les Elémens de Philosophie, § ix.

A few slight observations on both may be useful, in preparing the way for the discussions which are to follow.

1. That new applications of words have been frequently suggested by habits of association peculiar to the individuals by whom they were first introduced, or resulting naturally from the limited variety of ideas presented to them in the course of their professional employments, is matter of obvious and common remark. The genius even of some languages has been supposed to be thus affected by the pursuits which chiefly engrossed the attention of the nations by which they were spoken; the genius of the Latin, for instance, by the habitual attention of the Romans to military operations ; * that of the Dutch by the early and universal familiarity of the inhabitants of Holland with the details connected with inland navigation, or with a seafaring life. It has been remarked by several writers, that the Latin word intervallum was evidently borrowed from the appropriate phraseology of a camp; inter vallos spatium,—the space between the stakes or palisades which strengthened the rampart. None of them, however, has taken any notice of the insensible transitions by which it came successively to be employed in a more enlarged sense; first, to express a limited portion of longitudinal extension in general; and afterwards limited portions of time as well as of space.

"Ut quoniam intervallo loco

* "Medium in agmen, in pulverem, in clamorem, in castra, atque aciem forensem."-Cic. de Oratore.

+ How remote are some of the following applications of the word from its primitive meaning

"rum et temporum disjuncti sumus, per literas te"cum quam sæpissime colloquar." The same word has passed into our language; and it is not a little remarkable, that it is now so exclusively appropriated to time, that to speak of the interval between two places would be censured as a mode of expression not agreeable to common use. Etymologies of this sort are, when satisfactory, or even plausible, amusing and instructive: but when we consider how very few the cases are, in which we have access thus to trace words to their first origin, it must appear manifest, into what absurdities the position of the Encyclopedists is likely to lead those who shall adopt it as a maxim of philosophical investigation. *

Other accidents, more capricious still, sometimes

"Numerum in cadentibus guttis, quòd intervallis distinguun"tur, notare possumus."-Cic. de Orat.

"Dolor si longus, levis: dat enim intervalla et relaxat."Cic. Acad.

"Vide quantum intervallum sit interjectum inter majorum "nostrorum consilia, et istorum dementiam."-Cic. pro Rab. "Neque quisquam hoc Scipione elegantius intervalla negotiorum otio dispunxit."-Paterc.

66

* A considerable number of the idiomatical turns of French expression have been traced to the ceremonial of Tournaments; to the sports of the field; and to the active exercises which formed the chief amusement of the feudal nobility. See a Dissertation on Gallicisms (strongly marked with the ingenuity and refined taste of the author), by M. Suard, of the French Academy. Similar remarks may be extended to the English Tongue; on examining which, however, it will be found (as might be expected a priori), that the sources of its idiomatical and proverbial phrases are incomparably more diversified than those of the French.