[ 50 ] 



fludioufly technical, that even thofe who mofl commend his 

 limiles as appofite, cannot pretend that many of them are ex- 

 planatory. 



Of the peculiarities of John Ton's ftile, which 1 propofed to 

 treat of under my fecond head, as arifing from his ftudy of har- 

 mony, the principal I may call the parallelifm of his fentences ; 

 which admits no claufe, without one or two concomitants, ex- 

 aflly fimilar in order and conftrudlion. There is fcarcely a page 

 of the Rambler which does not produce abundant inftances of 

 this peculiarity : and what is the ornament, which, if intro- 

 duced fo often, can be always introduced happily ? Or what is 

 the ornament, however happily introduced, which will not dif- 

 guft by fuch frequent repetitions ? Johnfon's mind was fo com- 

 prehenfive, that no circumftance occurred to him unaccompanied 

 by many others fimilar ; no effedt, without many others depend- 

 ing on the fame or fimilar caufes. So clofe an alliance in the 

 thought naturally demanded a correfponding fimilitude in the 

 exprellion : yet furely all fimilar circumftances, all the efFeds of 

 each caufe, are not equally neceffary to be communicated ; and 

 as it is acknowledged that even a continued poem of pure 

 iambics would difguft, variety muft appear an indifpenfably ne- 

 cefiiary ingredient to harmony. Were we even to admit then, 

 that in any particular triod the conftruclion of one of its claufes 

 could not be altered without injuring the harmony of the fen- 

 tence, yet a regard to the harmony of the whole treatife will 

 occafionally make fuch an alteration neceffiiry. 



But thefe parallel fentences are not always faultlefs in them- 

 felves. S'i'-_:>i*imes, though indeed rarely, a word is ufed without 



a definitive 



