be more difficult, with the present philological and philosophical helps to such a work, than the first making of dictionaries and grammars, in the infancy of philology. Perhaps it may not be amiss just to hint, in this place, that the four following classes comprise all the possible kinds into which words can be distinguished, agreeably to the plan here proposed:
1. Words which have ideas, but no definitions.
2. Words which have both ideas and definitions.
3. Words which have definitions, but no ideas.
4. Words which have neither ideas nor definitions.
It is quite manifest, that words seen or heard, can raise no ideas in the mind, or vibrations in the brain, distinct from their visible and audible impressions, except as far as they get new powers from associations, either incidental ones or arising from express design, as in definitions; and therefore, that all other ways of considering words, besides what is here suggested, are either false or imperfect.
Cor. IV. As simple ideas run into complex ones by association, so complex ideas run into decomplex ones by the same. But here the varieties of the associations, which increase with the complexity, hinder particular ones from being so close and permanent, between the complex parts of decomplex ideas, as between the simple parts of complex ones: to which it is analogous, in languages, that the letters of words adhere closer together than the words of sentences, both in writing and speaking.
Cor. V. The simple ideas of sensation are not all equally and uniformly concerned in forming complex and decomplex ideas; i.e. these do not result from all the possible combinations of twos, threes, fours, &c. of all the simple ideas; but, on the contrary, some simple ideas occur in the complex and decomplex ones much oftener than others; and the same holds of particular combinations by twos, threes, &c. and innumerable combinations never occur at all in real life, and, consequently, are never associated into complex or decomplex ideas. All which corresponds to what happens in real languages; some letters, and combinations of letters, occur much more frequently than others, and some combinations never occur at all.
Cor. VI. As persons who speak the same language have, however, a different use and extent of words, so, though mankind, in all ages and nations, agree, in general, in their complex and decomplex ideas, yet there are many particular differences in them; and these differences are greater or less, according to the difference, or resemblance, in age, constitution, education, profession, country, age of the world, &c. i.e. in their impressions and associations.
Cor. VII. When a variety of ideas are associated together, the visible idea, being more glaring and distinct than the rest, performs the office of a symbol to all the rest, suggests them,
