Association.—The topic of association has traditionally been treated, firstly, so as to account for "complex presentations," and secondly, to explain the sequence of ideas in the course of thought.
In regard to the first problem it is important to note that asso ciation proper presupposes the prior and more fundamental phases of development with which we have been concerned. Though the various forms of complication and association are genetically continuous, the last of these terms is conveniently restricted to the organisation of the tissue of experience in which the related terms are explicitly distinct as free ideas.
The Laws of Association.—Associative processes have tradi tionally been formulated in terms of specific "Laws." Thus it was supposed, in the early days of the history of Psychology, that as sociative revival depends upon certain relations obtaining between the items of Perception : contiguity, in time or place, similarity and contrast. Reflection, however, clearly shows that there is nothing in the relations as such, in virtue of which one of the related items should recall the other. Contiguity of objects on a table would not determine the idea a percept of one or reinstate an idea of another unless the contiguity had itself been appre hended. There is no reason, independently of experience, why thought of anything whatever should induce the idea of that with which it stands in contrast. The basis of association is the co presentation of the associated terms in some relation to each other, and any kind of relation whatever, provided that it is per ceived and thought of, may serve as an associative link.
So called "association by similarity" may not, at first sight, seem to come under this general principle. A tiger may suggest a cat to a person who has had no previous knowledge of tigers such as would enable him to notice their resemblance to cats. But this difficulty is easily met. The phrase "Association by Similarity" is misleading. The associative link may be more accurately defined as the relation of both the associated terms to some common class or kind of which they are both members or instances. When we see a cat we are aware of it as a certain sort of animal. When we see a tiger we are also aware of it as a certain sort of animal. In some interesting respects and to an interesting degree, the tiger, in spite of specific differences is referred to the same gen eral class as the cat. This is the associative link between them. A red patch may suggest a blue patch not because of any spatial or temporal contiguity between either the patches as a whole or any of their parts, but because red is apprehended as a colour and blue is apprehended as a colour too. We may speak of a par
tial identity in red and blue in virtue of their being colours, but the phrase here takes on a meaning radically distinct from that which it has when applied to overlapping areas or events.
It is not only in association by similarity that we have to take account of universals. The "items" of association are never bare particulars of sense. That this is the case is seen in a detailed analysis of the operation of the law. What is revived is never a simple photographic copy which is constantly related to the original impression. When such is specifically required, it calls for deliberate reconstruction. In general, however, the photo graphic image would be of no advantage. What is required is something much more plastic. In the reinstatement of the context of an object seen near at hand, but imaged as at a distance, such context is correspondingly reduced, and different adaptations are required in every case. Such plasticity is in fact involved in the sensory forms of reinstatements such as pre-perception or halluci nation. The principle has been formulated as that of Relative Suggestion (Stout : Analytic Psychology, vol. ii. ch. 1), and it is in this way, rather than in terms of sensory atomism that the law of association requires to be expressed.
The point is of vital importance for a theory of imaginative con struction. The vague conception that imagination works with the data of disintegrated memory, variously recombined much as a mosaic pattern might be rearranged, fails to fit even the simplest facts. Perceiving a red square and a blue triangle makes possible the immediate presentment of these figures to the imagination with their colours interchanged. With a row of such figures each of a different colour, material is provided for endless permuta , tions. This principle affords an answer to the question with regard to that "so singular" possible exception to the rule, that we can only image that which has first been given to sense, that the mind cannot form a novel "simple impression." Could we imagine a given shade of blue when this had never been presented, given acquaintance with every other shade and every other colour? No doubt is felt if the question is raised with regard to shape, but no difference in principle would seem to be involved. Subject to the limits of discrimination (which in imagery is coarser than in sense) we can produce a number of novel members to a series which is presented to sense.