THE NATURE OF SYMBOLS 73 



meaning them rather than by portraying them. It still re- 

 mains true, as Wittgenstein has insisted, that every proposi- 

 tion is a picture of a fact. But one cannot ascertain the fact 

 by a mere examination of the symbol as a physical event; 

 one must know what the symbol means. In fact one may say 

 that the difference between icons and characterizing symbols 

 is that in the former the determination of the referent must 

 be made primarily by examining the symbol itself and only 

 incidentally by examining the reference, and in the latter it 

 must be made primarily by examining the reference and only 

 incidentally by examining the symbol itself. 



Supposing this general distinction between icons and 

 characterizing symbols to be valid, is there any reason why 

 the one type of symbol should be preferred to the other? 

 The problem has been formulated in terms of the rival 

 claims of pictorial representation, or models, and conceptual 

 representation, or abstractions. Whether the icon is a mental 

 image or a physical model consisting of pulleys, wires, and 

 rubber tubes is of no importance in the present considera- 

 tions; both are types of pictorial symbolism, one somewhat 

 ephemeral and intangible, the other permanent and material. 

 The insistence upon this type of concrete imagery has 

 characterized a group of English physicists, among whom are 

 Faraday, Lord Kelvin, Lodge, and Maxwell. Kelvin, for 

 example, says in effect that the question, Do we understand a 

 certain physical subject or not? means Can we construct a 

 corresponding model? ! "I advise you all who are engaged 

 in teaching, or in thinking of these things for yourselves, to 

 make little models." 2 Lodge puts this plan into practice 

 when he explains modern theories of electricity 3 by diagrams 

 of machines involving pulleys and cords, weights and drums. 

 As long as one limits himself, he insists, to the abstract 

 schemes of mathematicians who are "able to live wholly 



1 W. T. Kelvin, Baltimore Lectures on Molecular Dynamics (Baltimore: Johns 

 Hopkins, 1884), p. 131. 



2 Ibid., 2nd ed., 1904, p. 34. 



3 Oliver Lodge, Modern Views of Electricity (2nd ed., New York: Macmillan, 

 1892), passim. 



