And, first, It is needless to pronounce every letter so as to distinguish it from all others. For then words, which are composed of letters, would each have as many criterions as they have letters, and even more; for the order of the letters is a criterion, as well as the sound of each letter. In like manner, sentences would have as many compound criterions as they have words, besides the criterion arising from the particular order of the words.
Secondly, Since words are formed from combinations, not according to any rule, which brings up all the combinations of twos, threes, &c. in order, but by particular associations, agreeably to the nature of each language, since also sentences are formed in the same way, the several component parts of words and sentences suggest each other, and also the whole words and sentences, by the power of association. Thus the beginning is commonly observed to suggest the whole, both in words and sentences; and the same is true, in a less degree, of the middles and ends.
Thirdly, The subject matter of the discourse, the gestures used in speaking, a familiar acquaintance with the particular voice, pronunciation, gestures, &c. of the speaker, and other associated circumstances, contribute greatly also. And therefore, on the other hand, we find it difficult to distinguish proper names, and the words of an unknown language, and to understand a person that is a stranger, or that uses no action.
We may see also, that it is chiefly by the means of associated circumstances, that the sounds uttered by ventriloqui suggest to us the words which they are supposed to pronounce; for their articulation must be very incomplete, as they do not move their lips at all.
It is by a like set of associated circumstances that we are enabled to read with so much facility the irregular hand-writing of various persons, and of some more than others, in proportion as we are better acquainted with the subject, language, handwriting, &c.
For the theory of sounds deduced from the nature of an elastic fluid, and the tremors of the particles of sounding bodies, and afterwards verified, as it has been, by numerous experiments, becomes a guide to us in all inquiries into the vibrations of other elastic mediums, such as the æther, and into the effects, which such vibrations must impart to and receive from the small particles that are surrounded by the elastic medium. And the general tendency which is found in natural bodies, either to yield a sound upon percussion, or at least to support and convey
