impu′nity, etc. and therefore that chastity and obsenity ought to be pronounced with the penultimate vowel short, and not as in chaste and obsene, as we frequently here them. I find too, that even u contracts itself before two consonants, as cur′vity, tacitur′nity, etc. and that scarcity and rarity (for whose irregularity good reasons may be given) are the only exceptions to this rule throughout the language. And thus we have a series of near seven hundred words, the accentuation of which, as well as the quantity of the accented vowel, are reduced to two or three simple rules.
The same uniformity of accentuation and quantity may be observed in the first syllable of those words which have the accent on the third, as dem-on-stra′tion, dim-i-nu′tion, lu-cu-bra′tion,[1] etc. where we evidently perceive a stress on the first syllable shortening every vowel but u, and this in every word throughout the language, except where two consonants follow the u, as in cur-vi-lin′e-ar; or where two vowels follow the consonant that succeeds any other vowel in the first syllable, as de-vi-a′tion; or, lastly, where the word is evidently of our own composition, as re-con-vey′: but as u in the first syllable of a word, having the accent on the third, has the same tendency to length and openness as was observable when it preceded the termination ity, I find it necessary to separate it from the consonant in bu-ty-ra′ceous, which I have never heard pronounced, as well as in lu-cu-bra′tion, which I have; and this from no pretended agreement with the quantity of the Latin words these are derived from; for, in the former word, the u is doubtful: but, from the general system of quantity I see adopted in English pronunciation; this only will direct an English ear with certainty: for, though we may sometimes place the accent on words we borrow from the Greek or Latin on the same syllable as in those languages, as acu′men, elegi′ac, etc. nay, though we sometimes adopt the accent of the original with every word of the same termination we derive from it, as assidu′ity, vi-du′ity, etc. yet the quantity of the accented vowel is so often contrary to that of the Latin and Greek, that not a shadow of a rule can be drawn, in this point, from these languages to ours.[2] Thus, in the letter in question, in the Latin accumulo, dubious, tumor, etc. the first u is every where short; but in the English words accumulate, dubious, tumour, every where long. Nuptialis, murmur, turbulentus, etc. where the u in the first syllable in Latin is long, we as constantly pronounce it short in nuptial, murmur, turbulent, etc. Nor indeed can we wonder that a different oeconomy of quantity is observable in the ancient and modern languages, as in the former, two consonants almost always lengthen the preceding vowel, and in the latter as constantly shorten it. Thus, without arguing in a vicious circle, we find, that as a division of the generality of words, as they are actually pronounced, gives us the general laws of syllabication, so these laws, once understood, direct us in the division of such words as we have never heard actually pronounced, and consequently to the true pronunciation of them. For these operations, like cause and effect, reflect mutually a light on each other, and prove, that by nicely observing the path which custom in language has once taken, we can more than guess at the line she must keep in a similar case, where her footsteps are not quite so discernible. So true is the observation of Scaliger: Ita omnibus in rebus certissima ratione sibi ipsa respondet natura. De causis Ling. Lat.
- ↑ See Principles, No. 524, 527, 530.
- ↑ See Principles, No. 544, 545.
