may nevertheless be employed as words, or roots of words, used in expressing the exercise of that faculty. The interjection eja, as has been said, is the root of ejulatio; and the onomatopoeia cuckoo! is used as a noun, in naming the bird which utters that sound. Of these roots, and such as these, there can be no doubt. But they supply a comparatively small portion of language. The difficulty is to ascertain in all other cases, how certain combinations of articulate sound came to express thoughts of the mind, or impressions of the senses: and on this point several theories have been suggested. 352. Some authors assume, that there is a power in every letter to Power of express a peculiar emotion or perception; a notion which furnished letters. the cabalistic writers with many mysterious doctrines. Mr. Whiter adopted a similar theory, but on somewhat different grounds. He argued, that as algebra is founded on the simple principle that equals being taken from equals the remainders must be equal; so a knowledge of words depends on the simple principle that the letters composing words have each a natural power of expressing some mental impression. But in the first place it is a gratuitous assumption that letters possess any such power; and secondly the analogy to algebra entirely fails; for the algebraic principle is an idea of the mind, which is necessarily universal; whereas the supposed glossological principle, were it true, could only be discovered by induction from numberless facts, and must therefore be necessarily but general. Again, if Mr. Whiter's theory were true of letters, the English alphabet of twenty-six letters must be competent to express little more than half the thoughts, which might be expressed by the Sanskrit alphabet of fifty letters. And it the principle were applied to the articulations represented by letters, a Chinese, who cannot pronounce several of our articulations, must be unable to express (though he still might conceive) many of our thoughts. We may therefore fairly deny that any such power of expression exists, either in letters or articulations, uncombined. 353. It is an ancient doctrine that the signification of words, and Contract. consequently of their roots, was established among mankind by contract; but to this the same objection lies, as to the doctrine of contract being the foundation of government-namely, that no such contract ever existed, as far as we are informed by history, or can conceive by probable conjecture. 354. Persons of no inconsiderable eminence in literature have held Divine inspiration. that the language of our first parents was inspired by the Almighty. But as this is not plainly asserted in the sacred writings, we cannot be justified in claiming their authority for such an assumption; and even were the fact admitted, there would be no reasonable ground for connecting it with any one existing tongue, and much less with the vast variety of tongues, which are or have been spoken throughout the world. 355. Upon the whole, the present state of glossological science Uncertain. does not justify us in asserting with confidence any primæval origin of How to find a root. verbal roots, except those which are supplied by interjections or 356. It remains to show how the root of any word is to be distinguished from any other part. And here it is first to be considered whether the word be native or foreign. If a word be introduced from a foreign language, it may indeed serve for a root to certain derivatives or compounds, which may be formed from it; but its own root is to be sought in the language from which it is taken, and thence perhaps in another, or others. Take, for instance, the English word Parliament, which has been used in this country for several centuries, in the sense which it still retains. Now this word may be considered as a root, in reference to the derivative Parliamentary, or the compound, a Parliament-man. They say, he the constable greatly outran, Anstey, Bath Guide. But no part of the word Parliament is its root, in the English lan- 2 tracing of roots from one language to another forms great part of the art called (however improperly) Etymology, which will hereafter be considered more at large. On the other hand, if the word, whose root is required, be of native origin, that is to say, if it belong to those which have formed the great staple of the language from its earliest ages, as those English words have, which have come down to us from the Saxon times, we must begin by depriving it of those particles, which, in the same and other words, serve the purposes of inflection, or derivation, either as prefixed, inserted, or suffixed. The remainder will be what some grammarians call the crude form of the word; and this is the root either unchanged, or subjected to some of the differences of articulation above specified.' It depends on the idiom of a language, whether a root can be involved in few or many particles. The English language admits of few involutions of a root, seldom exceeding four; as in the word unforgivingly, where the root, give, has two prefixed particles, un and for, and two affixed, ing and ly. The North American languages, as has been shown above, generally involve the root in many particles, and subject it to various changes. In Welsh a derivative may not only have particles prefixed and suffixed, but also subject the root itself to change, as in difrychenlyd, unspotted, the root frech, a spot, has not only the negative prefix di, and the affixes en and lyd, but also changes its vowel from ê to y. 357. Grammarians have adopted different parts of speech as roots. In what part of speech. Dr. LEE thinks that the noun substantive should be considered (at least in Hebrew) to be the root. M. COURT DE GEBELIN considers every primary root to be a noun substantive describing a physical object.3 Dr. DONALDSON seems to regard adjectives as the primary roots. In the Albanian language, not only nouns substantive and adjective, and verbs, but also adverbs, often show the root in its simplest form, as lir, cheaply, ilire, cheap. Of those who adopt the verb as a root, H. STEPHANUS and many others take the first person singular of the present tense indicative for that purpose; some take the third person singular of the præterite indicative. Mr. ARCHBELL states the (so-called) second person singular of the present imperative, as exhibiting most distinctly the root, in the Sechuana language. "In this situation (says he) the simple root appears, unencumbered by prefix or affix, and yet not wanting in any of its integral parts.' 6 This remark may be extended to most, if not all, languages; because the imperative expresses emotion, and therefore leads to a short mode of expression. Hence it is always either a simple root, or a root with a short vowel prefixed or affixed. In Turkish, Mr. DAVIDS says, "the imperative is formed by suppressing the termination of the infinitive, as deug! (from deugmak), strike! kork! (from korkmak), fear! But in common conversation, the sound of the (short) letters alif and ha 1 Supra, sec. 332, 337. 3 Monde Primitif, vol. iii. p. 57. 5 Leake, Researches in Greece, p. 290. Hebrew Grammar, p. 83. 4 New Cratylus. 6 Sechuana Grammar, p. 7. Conclusion. 1 are often joined to the imperative, as deuga! korkah!" So in Latin we have pende! and in Greek TÚTTE. But these are merely matters of pronunciation, affecting in a very slight degree, or not at all, the meaning or effect of the word. The same may sometimes be said of our common prefix a, as in Satan's address to the infernal hostAwake! arise! or be for ever fallen,2 where, had the metre permitted the use of the imperatives wake! rise! the signification would have been precisely the same. The prefix a before an adjective is often in like manner superfluous, as in Macbeth's mournful exclamation I 'gin to be aweary of the sun, where the prefix a serves at most to mark somewhat more strongly the feeling which weary alone would have expressed. 358. From what has been said, it may be concluded that the root of a word, though most commonly a single syllable, may, in certain cases, comprehend more than one syllable; that it may be susceptible of change both in its vowels and consonants; and that though, according to the idiom of some languages, those articulate sounds, which form the root of a word, may be also employed alone as a word, yet generally a root requires the aid of some one or more other articulate sounds, prefixed, inserted, or suffixed, to form a word, and enter into construction as part of a sentence. CHAPTER XII. OF PARTICLES. the term. 359. THE term Particle has been employed by most grammarians, Meaning of ancient and modern, to signify certain classes of words, which are said to be indeclinable, such as adverbs, prepositions, conjunctions, and interjections. This use of the term, though sanctioned by long practice, appeared to me objectionable on two grounds: first, because the indeclinable words being reckoned, equally with the declinable, as parts of speech, it seemed inconsistent to term them also particles, that is, something less than parts; and, secondly, because the grammatical systems which treat whole words as particles, furnish no specific designation for those portions of articulate sound which, combined with roots, make up the great majority of words in all languages not purely monosyllabic. For these reasons, I many years since employed, and shall continue to employ, the term particle to signify any portion of a word, unless separately cognizable as a noun or verb, which is either introduced for the mere sake of euphony, or else serves to modify the root lexically, or grammatically. In this sense, the term particle nearly answers to the Greek λεξέιδιον, derived from λέξις, λεξίος, Ionically, as ῥησείδιον, from ῥήσις, ῥήσιος. Any portion of a word, which may be recognised separately as a noun or verb, is not to be deemed a particle, but will be considered hereafter under the head of compound words. 1 360. A particle may consist of one or more articulations, and may Position. be placed at the beginning or end of a word, or in some intervening position. When placed at the beginning of a word it is called a prefix; when at the end an affix, or (perhaps more properly) a suffix; and when intermediate it may be denominated (as in fact it has been by some writers) an interfix. Thus in the Latin cecidi, ce is a prefix; in amavi, vi is a suffix. In the Greek ETETúpeLv, ɛ is a prefix, Tɛ is an interfix, and ɛ is a suffix. The interfixes have been comparatively little noticed, yet in many languages they perform important functions. Thus in Turkish the root sev with the suffix mek forms the active infinitive sevmek, "to love;" if the interfix il be added, sevilmek signifies "to be loved;" and if the further interfix me be introduced, as in sevilmemek, it signifies "to be not loved." In the Kafir language the particle ka is inserted between the negative verbal prefix 1 Etymol. Magn. voc. Xişéidiov. 2 Davids' Gram. Turke, p. 33. |