76 TWO THEORIES OF KNOWLEDGE 



It is noteworthy that the same thing necessarily 

 happens in the case of any new system of spelling. 



The most prominent advocates of phonetic spell- 

 ing have been also the authors of a system of 

 phonetic shorthand. 



Like the written and printed alphabet of Europe, 

 the alphabet of Phonography was made phonetic. 

 Indeed it started off as a more nearly perfect 

 phonetic system than the ordinary European alpha- 

 bet. But as its use advances its employment under- 

 goes the same change. The phonetic symbols are 

 abbreviated by grammalogues and contractions, and 

 this proceeds in accordance with a principle un- 

 consciously recognised but which really depends 

 on the same inherent necessity to preserve in a 

 consistent form the expression of the radical 

 vocables of Speech. Finally, in the hands of the 

 expert stenographer the system of phonetic short- 

 hand (though he still uses the sound as the guide 

 and indicator of his actions) is as far removed from 

 a pure phonetic representation as the ordinary 

 method of spelling. Indeed, unless some such supra- 

 sensible and unifying principle were available, 

 phonetic spelling would speedily perish in an 

 infinity of degenerate variations. 



We adduce this illustration as one which very 

 well confirms our main argument. We have no 



