When several qualities are observed to be constantly united in nature, a strong association is formed among our ideas of such qualities ; but if we make a careful analysis of mental phenomena, we shall find that, in pronouncing the name of any material being, certain sensible qualities, more or less vaguely conceived, are the only objects of our thoughts.
This dissertation on our ideas of substances may appear a deviation from the subject of Universal Grammar. But, since grammarians have supposed these ideas to be closely connected with the theory of nouns, it seemed necessary, in order to do justice to our subject, that we should shew the fallacy of the common doctrines from the nature of our thoughts, as well as from the structure of language. The views which we have stated lead us to no vague or per plexing conclusions. The fact of the uniformity of the definite combinations of certain material qualities, is in no degree deprived of its solidity or interest, though we de cline to admit the hypothesis of a substratum. The rejec tion of this hypothesis will assist our physical, as well as our grammatical studies. It will relieve us from the embarrassment of the understanding, which sometimes takes place when particular qualities are found in a de tached state. Those who are unpractised in the accurate exercise of thought, and have been led astray by words, have, in the outset of their physical inquiries, found it difficult to conceive that a body which is felt yet not seen, as the air, has an existence equally substantial with other matter. They have also considered light, which implies an object of sight unaccompanied by any object of touch, as on this. account more difficult to he understood than earth, stones, and other substances, which are both tangi ble and visible. The most rational proceeding is, to satisfy ourselves with such qualities of tangibility and visibility as we can ascertain in any of their peculiarities and rela tions, whether they are observed in a state of conjunction or of separation.
Thus, when we attempt to trace the supposed differen ces betwixt substances and qualities, we find no founda tion for a distinction into two kinds of objects, and much less for a distinction in the kind of terms by which such objects should be expressed. This same conclusion to which we arc led in tracing the history of nouns. We
find that the same kinds of ideas are designated by them as by adjectives.
If the distinction betwixt nouns and other parts of speech cannot be founded on the place which the objects expressed by them occupy in nature, it must rest entirely on the manlier in which they are introduced with relation to the other words with which they are conjoined. It depends on the rank which the word occupies in a sentence, and which the thought excited by it is intended to occupy in that -mental series which we wish to produce.
How, then, are we to define the noun, so as to distinguish it from the other parts of speech ? Shall we, with Mr Tooke, consider it as the " mere name of an idea ?" Shall we consider the verb as a part of speech more compli cated in its nature, by containing " some circumstance in addition to the name of an idea ?" And shall we be in duced to extend a similar character of complication, in a smaller degree, to the adjective ? This mode of proceeding might at first appear plausible. But, on closer reflection, we shall find that no word, not even a substantive noun, exists as the mere name of an idea ; that there is always a demonstration of some further definite use to which it is to be applied. This we know to be the purpose of the variations called cases. Even the nominative case has a peculiarity which does not consist in the want of any such demonstration. The syllable us in dominus, denoting the nominative case, informs its that the noun is to be connect ed with a verb of assertion. The genitive case, and all the others, in like manner point out some definite use of the noun. If we separate these terminations, and consider them as distinct signs, and regard the radical sy114bles as containing the essence of the noun, (as in this example the syllables domin.) we shall still retain the name of the idea, but we shall have nothing to distinguish the noun from the other parts of speech. If a verb is deprived of all the parts which are intended to connect the idea which it repre sents with the other ideas expressed in a sentence, we shall, in like manner, retain the mere name of an object. In do min we have the radical syllables of the verb dominor, as well as of the noun dominus.