360 LYMAN— A PRACTICAL RATIONAL ALPHABET. [Oct. i, 



the original forms of the letters : as for instance the carvilium to 

 distinguish G from C and the distinction between J and I and 

 between U and V, which appear to be still struggling for complete 

 prevalence. It may, however, be borne in mind that notable addi- 

 tions to the Arabic alphabet have been made and accepted in order 

 to express additional sounds in Persian or other languages : but it 

 is noticeable that such added forms are strictly in keeping with the 

 original character of the alphabet. The Russians have also strongly 

 modified the Roman alphabet, and not always quite in keeping with 

 the rude simplicity of its general character ; yet have established its 

 use throughout a great empire. In proposing new forms of letters 

 for newly distinguished sounds, it is certainly advisable to maintain 

 some restraint upon one's fancy, to adhere to the utmost simplicity, 

 and to depart as little as possible from the general character of bare 

 simplicity of the Roman alphabet, making use, so far as possible, 

 of old devices, and putting forward as few novelties as possible, 

 to be learned and made familiar. It seems highly desirable to avoid 

 the use of altogether outlandish forms like the fully obsolete old 

 Anglo-Saxon letters, wholly out of keeping with our modern 

 alphabet ; or to offend the eye by intermixing italic letters with 

 Roman and by other tasteless similar devices, or by interspersing 

 inverted letters, though to be sure of good Roman shape. Above 

 all, however, let us avoid separate diacritical marks to distinguish 

 sounds, marks that are a nuisance to write, an obscurity to read, and 

 by their occasional forgetful omission a fruitful source of mislead- 

 ing. Especially the use of diacritical marks in a way opposed to 

 their time-honored significance, is to be reprehended ; as for example, 

 the use of an accent to indicate merely the length of a vowel. Such 

 practice has misled commonly into various errors of pronunciation 

 of some oriental words. We shall see if there be any serious 

 difficulty in getting handsomely along without any of those hastily, 

 inconsiderately adopted, tempting, shallow, easy, but terrible, make- 

 shifts. There are some restraints, or guides, which must cogently 

 influence our choice of letters or symbols to be used in indicating the 

 different sounds of the language. It is highly desirable, or ab- 

 solutely necessary, that each sound should be indicated by only one 

 letter, and that each letter should have but one sound ; and it would 



