When it is admitted that simple sensations remind us of one another by direct resemblance, many of the complex cases of suggestion by resemblance may be analysed into this elementary case of association by resemblance, combined with an association by contiguity. A flower, for instance, may remind us of a former flower resembling it, because the present flower exhibits to us certain qualities, that is, excites in us certain sensations, resembling and recalling to our remembrance those we had from the former flower, and these recall the entire image of the flower by the law of association by contiguity. But this explanation, though it serves for many cases of complex phenomena suggesting one another by resemblance, does not suffice for all. For, the resemblance of complex facts often consists, not solely, or principally, in likeness between the simple sensations, but far more in likeness of the manner of their combination, and it is often by this, rather than by the single features, that they recall one another. After we had seen, and well observed, a single triangle, when we afterwards saw a second there can be little doubt that it would at once remind us of the first by mere resemblance. But the suggestion would not depend on the sides or on the angles, any or all of them; for we might have seen such sides and such angles uncombined, or combined into some other figure. The resemblance by which one triangle recalls the idea of another is not resemblance in the parts, but principally and emphatically in the manner in which the parts are put together. I am unable to see any mode in which this case of suggestion can be accounted for by contiguity; any mode, at least, which would fit all cases of the kind.--_Ed._]
{ 112} Mr. Hume makes contrast a principle of association, but not a separate one, as he thinks it is compounded { 113} of Resemblance and Causation. It is not necessary for us to show that this is an unsatisfactory account { 114} of contrast. It is only necessary to observe, that, as a case of association, it is not distinct from those which we have above explained.
A dwarf suggests the idea of a giant. How? We call a dwarf a dwarf, because he departs from a certain standard. We call a giant a giant, because he departs from the same standard. This is a case, therefore, of resemblance, that is, of frequency.