In fact, before articulate speech, for which man was adapted by his organs and physiological conditions, was formulated into words for things and words for shape, man like animals thought in images; he associated and dissociated, he composed and decomposed, he moved and removed images, which sufficed for all individual and immediate operations of his mind. The relations of things were felt, or rather seen through his inward representation of them as in a picture, expressing in a material form the respective positions of figures and objects which, since they are remote from him, can only be expressed by such words as _nearer, lower_ or _higher, faint_ or _clear_, by more vivid or paler tints, such as we see in a running stream, in the forms of clouds, in the reciprocal relations of all objects represented in painting.
In order to understand the primeval process of thought by means of images, it is necessary to conceive such a picture as living and mobile, and constantly forming a fresh combination of parts. Animals have not, and primeval man had not, the phonetic signs or words which give an individual character to the images, and so represent them that by combining these images in an articulate form, thought may be represented by signs; and in and through these a universal and objective mode of exercising the intellectual faculty of reasoning has been created.
Speech can, by means of reflex memory, produce at will the particular images already classified in the mind, and this makes the process of reasoning possible; since such a process becomes more easy by the use of signs to which the attention can revert. The relative size of objects, and the like qualities, which are at first regarded as so many different intuitions in space, are defined by words or gestures, and are thus subjected to comparative analogy; but in the early stages of language these relations were presented in an extrinsic form by phonetic signs, and became images which in some sort represented one particular state of consciousness with respect to the two things compared. Galton, speaking of the Damaras, tells us that they find great difficulty in counting more than five, since they have not another hand with which to grasp the fingers which represent the units. When they lose any of their cattle, they do not discover the loss by the diminution of the number, but by missing a familiar object. If two packets of tobacco are given to them as the regulation price of a sheep, they will be altogether at a loss to understand the receipt of four packets in exchange for two sheep. Such examples might be multiplied to any extent.