120. Sensible objects, and particularly visible, are undoubtedly the first which exercise the power of abstraction, or se parate attention, and here the process ap pears plain. The object makes its ap propriate impression upon the organs of sense, and when withdrawn leaves in the mind an idea. Another sensation is re ceived from an object bearing strong fea tures of similarity to the former; by the laws of association it calls up the idea it produced, and becomes associated with it. Other similar objects are presented, and the features in which they agree being the most frequently called up, engage most the attention of the mind, and thus becoming, in some degree, separate from the objects which originally were con nected with them, constitute the abstract idea. The readiness with which these circumstances of resemblance recal the idea or conception of the individuals from which they were abstracted, depends upon the habits of the individual, and the number of objects from which the ab stract notion was formed. If we had seen but two or three sheep, it is probable that the circumstances of resemblance would be so connected in our minds with the individuals, that one or more of them would be constantly called up when con sidering the circumstances of resem blance ; but if the number be much greater, that •is, if the circumstances of resemblance have been frequently in the mind, particular individuals much less frequently, the notion of these circum stances of resemblance becomes some what disjointed from the objects by which it was fbrmed. And though it is proba bly impossible to have a general idea of any class of objects merely sensible, with out the idea of an individual being pre sent in the mind; yet from the causes I have mentioned, the general features of resemblance not being particularly con , nected with any individual, those features only are strong and vivid, and call the at tention of the mind, while all the other circumstances of dissimilarity have no ef fect upon it, and do not attract its atten tion.
121. The procedure of the mind ap pears to be exactly the same, though less obvious, and usually more difficult of analysis, when the general idea is more remote from sensation, when in fact, the notions of the quality, or qualities, even in the individual, may be very complex, and this in proportion, as it is more intel lectual and refined.—In the former class of general notions, and even in some in stances of the present, where the quality is definite and obvious, it is probable that language would not be requisite for the formation of ideas; indeed, if' the above account be just, it cannot be requisite. For the abstraction, so far as it is invo luntary, is solely the effect of the fre quent recurrence of some particular quali ties with which they are occasionally combined. But those abstract ideas in which the circumstances of resemblance between the composing ideas are not very obvious or very distinct, either would not have been fbrmed at all by the bulk of mankind, or at least would have been ve ry confused.—We can go very far with those who contend, that general ideas would not exist in the mind without the medium of language ; but that they could not, from any deficiency of mental capa city to form them, does by no means ap pear certain. The same faculties which
now produce them, might have produced them without the powers of communica tion ; and there appears no reason why the deaf and dumb child may not form a general idea amen, or horses, or fire, or any object'of a similar kind, as well as if capable of annexing terms to the objects of perception.
122. It can be no objection to this ac count of the procedure of the mind in generalization, that we are able to form classifications of objects from circum stances which are not calculated to strike the mind of the common observer. 'When left to itself, before habits of reflection are formed, the mind will be necessarily at tracted by the most prominent sensible features of resemblance, and the objects would become associated by that bond or union ; and in very many cases this would differ in different individuals; but it i s indubitable, that we may acquire such a command over our associations, that we may be able to combine objects in our minds which have no customary tendency to such combination, owing to the laws of association, by a more factitious connec tion, and that, by the requisite culture of the mind, certain connecting principles are either discovered or confirmed, which could not have been of any force in a more early period of mental progress. In the first of these cases the association is voluntary, and if there were not some apparent benefit resulting from it, or some circumstances calculated to pro duce it in the mind, it would soon give place to a more natural union. So far, however, as any general idea is formed, its production is accomplished agreeably to the principles we have stated. In the second, the operation of the mind is most usually involuntary ; when voluntary, the observations on the first cease to apply.
123. It is obvious that, the fewer and more distinct the circumstances which are comprehended under the ge neral notion of a class of objecti, the more clear and definite will be the gene neral notion itself. And it appears wor thy of notice, and tends to confirm the account given of the formation of our ge neral ideas, at least those of visible ob jects, that the greater the variety subsist ing among the individuals or subordinate species comprehended under the general idea, or, more properly, which Possess that quality, or combination of qualities, which compose the general idea, the less attention, other things being equal, do we pay to the peculiarities of the individu al. Thus the general notion of a trian gle is merely that of a figure having three sides ; and the varieties of triangles are innumerable : and, agreeably to the opi nion already mentioned, though we cer tainly cannot form a conception of a tri angle which shall be representative cf all others, without possessing the peculiari ties which constitute it an individual, yet the circumstances of its having three sides is so finite, and our attention is So thoroughly confined to it, that the pecu liarities of the triangle are Out unfeequent lv totally out of consideration ; and if, ow.