intelligences demands elaborate préparation, even after years of 
successful performance. 
It is therefore useful to possess an art by which one can obtain 
at a moments notice any information that may be necessary. This 
art is divination. The answers to one’s questions in divination are 
not conveyed directly but through the medium of a suitable sériés 
of symbols. These symbols must be interpreted by the diviner in 
terms of his problem. It is not practicable to construct a lexicon 
in which the solution of every difficulty is given in so many words. 
It would be unwieldy 3 besides, nature does not happen to work 
on those lines. 
The theory of any process of divination may be stated in a few 
simple terms. 
1. We postulate the existence of intelligences, either within 
or without the diviner, of which he is not immediately conscious. 
(It does not matter to the theory whether the communicating spirit 
so-called is an objective entity or a concealed portion of the 
diviner’s mind.) We assume that such intelligences are able to 
reply correctly — within limits — to the questions asked. 
2. We postulate that it is possible to construct a compendium 
of hieroglyphs sufficiently elastic in meaning to include every 
possible idea, and that one or more of these may always be taken 
to represent any idea. We assume that any of these hieroglyphics 
will be understood by the intelligences with whom we wish to 
communicate in the same sense as it is by ourselves. We hâve 
therefore a sort of language. One may compare it to a lingua 
jranca which is perhaps defective in expressing fine shades of 
meaning, and so is unsuitable for literature, but which yet serves 
for the conduct of daily affairs in places where many tongues are 
spoken. Hindustani is an example of this. But better still is the 
analogy between the conventional signs and symbols employed by 
mathematicians, who can thus convey their ideas perfectly 1 without 
speaking a Word of each other’s languages. 
1. As a matter of fact, they cannot. The best qualified are the most 
diffident as to having grasped the meaning of their colleagues with 
exactitude; in criticising their writings they often make a point of apologis- 
ing for possible misunderstanding. 
— 156