Owen — Revision of Pronouns. 103 



therefore, that the operation of restrictive clauses has been suffi- 

 ciently exhibited in the preceding illustrations, I content myself 

 with the following generalizations : The purely restrictive 

 clause is an adjunct, auxiliary to some element of the principal 

 clause ; it may, at least in theory, be adjunctive to any such 

 element and in any degree of remoteness ; its function is never 

 that of a sentence, but always that of a subordinate sentence- 

 factor ; accordingly it does not make a statement.-^ 



From what has thus far been noted it appears, as many think- 

 ers indeed have conceded, that whenever thought is extended 

 beyond the minimum of the type "A = B,'' it must be by the 

 methods followed in forming that minimum. Some element of 

 '^A = B,'' as for instance ^^B," must appear in relation with 

 some new element, as in '^B> C." If the increment is to be 

 felt as structurally one with the original, the common factor of 

 the two, namely B, must be once thought only, or simultaneous. 



JSTow thought, at least as expressed in the language of cultured 

 peoples, exhibits extension as a rule, and usually very miany ex- 

 tensions. The venerable propositions, "Men are animals,'' 

 "John is a man," "John is an animal," are rarely encountered 

 outside of philosophy. It is much more common to meet with 

 statements such as "That smug Jolin Smith, who just bowed so 

 deferentially, with all his conciliatory manner, is a sly fox/* 

 What may be called the minimal sentence is, then, the linguistic 

 exception. In actual practice the sentence is almost always ex- 

 tended ; and, as above indicated, extension implies the occurrence 

 of a simultaneous factor. 



The problem of the simultaneous factor (which is the prob- 

 lem of the extended sentence) is, accordingly, far more fre- 

 quent than that of the minimal or elementary setitence. It 

 may indeed be conceded that this is the problem of speech. 

 To exhaust this problem, in the present investigation, would 

 require time and effort quite excessive. Yet, the relative clause 

 being only one of many forms in which the simultaneous factor, 



lit must be admitted, as implied on p. 87, that a given relative clause, while 

 aiming on the one hsind to aid the informational effort of another clause, may 

 occasionally be at the same time intended to give information of its own. Such 

 a clause, though restrictive, will be asserted because also informational. On 

 p. 110 will be found some indication that assertion by purely restrictive clauses 

 is apparent only. The exhaustive discussion of the assertion problem in relative 

 usage seems to belong of right to a special investigation of mode. 



