§ 9. Advantages of Conventional Language. — The primary and essential procedure of the language of natural signs is to represent things and processes by imitating the broad features of their sensible appearance and especially of their appearance to the eye. But the characters which are capable of being so imitated are of a comparatively low grade of generality or abstractness. They represent an analysis of perceptual experience into universals and its reconstruction out of these universal. But the universals themselves are very far from being simple and ultimate. They in their turn are intrinsically susceptible of analysis, and constituents thus revealed are again susceptible of further analysis, and so on. Now the more advanced is this process of dissection, the more helpless is pictorial representation to express the result either within the individual consciousness or in the intercourse of different minds. But the power of mentally representing a universal is simply the power of conceiving it. Where the mental representation does not exist, the conception does not exist. Hence a mind whose discursive thinking could only find expression in selfinterpreting signs, would be incapable of the higher reaches of abstraction. Broadly speaking, natural signs are capable of fixing attention on universals which are constitutive characters of particular objects as presented in perceptual experience; but they can only to a very limited extent fix attention on universals which are constitutive characters of other universals. The thinking which depends on the imitative gesture generates concepts; but it can hardly generate a conceptual system, in which there is an ascending scale of generalisation, passing from species to genus, and from genus to higher genus, and so on through a series of gradations till the highest genus is reached. It seems beyond the unaided powers of the thought which works through natural signs to frame a system of classification.

This impotence of the imitative gesture to express higher universals is easily illustrated. "To make," says Tylor, "is too abstract an idea for the deafmute; to show that the tailor makes the coat, or that the carpenter makes the table, he would represent the tailor sewing the coat, and the carpenter sawing and planing the table."* According to Schmalz, "The more general determinations of magnitude such as broad, narrow; long, short; thick, thin; high, low; cannot be accurately expressed; the most that can be done is to teach the deafmute signs which are suitable to the largest proportion of cases."*

* Ueber die Taubstummen, p. 275.

Often a general concept is capable of pictorial expression; but only in a way which is cumbrous and circuitous when compared to the conventional. A series of imitative gestures may be needed, where a single word would effect the same purpose with greater precision and certainty. Thus an Indian who wished to convey to a deafmute that he had travelled in the train, could only convey the idea of train by three successive bits of pantomime, one representing the conception of something covered in, another that of wheels, and the third that of smoke. Now this mode of expression may at first sight appear more analytic than the use of the single word train. It resolves into three universals and reconstructs out of them what the conventional language expresses in its totality by one sign. But we must remember that we could use many signs if there were need for them. If there were occasion to give an analytic description of a railway train, that description could be given with far more fineness, precision, and adequacy in words than in imitative gestures. The conventional language uses one word because one word is enough. The language of imitative gesture uses three separate bits of pantomime because it cannot do with less. It is forced to describe because it cannot directly designate. Now why is it unable to express by a single appropriate sign the general conception of a railway train? The reason is that the concept of a railway train possesses too high a degree of universality. It gathers up into unity a great multiplicity of special features, functions, and relations. Among these are included some which it would be difficult to express concisely, or even to express at all, in the gesture language, as for instance the principle or mechanism of its locomotion, its function as a means of communication and traffic, and so on. Now the comprehensive unity which embraces within itself all these particular determinations is not capable of being directly expressed in natural signs, without the aid of convention. Hence in a communication by imitative gestures, where there has been no previous convention, the only course possible is to select certain particular characteristics of the object, which are at once important and easily presented to the eye, and to exhibit as many of them as appear sufficient to enable the intended meaning to be divined. The same deficiency also makes it difficult to refer to an individual person by a selfexpressive sign. The imitative gesture as such and apart from convention is incapable of directly expressing the universality which belongs to the individual and is associated with the Proper Name. The depiction of some special characteristic or peculiarity may or may not be successful in directing attention to the person intended. All depends on context and circumstances. If representation of this or that characteristic proves insufficient, others may be added until understanding is reached, as in the case of the railway train.