RULES OF THE SYLLOGISM. 351 



two negative premises nothing can be inferred, because 

 two differences admit of no reasoning upon them, and 

 because two terms may each differ from a third one, and 

 may or may not differ from each other. That if one pre- 

 mise be negative, the conclusion must be negative ; and 

 if both the premises are affirmative, the conclusion must 

 be so. That if one premise be particular, the conclusion 

 must also be particular ; and that from two particular 

 premises no conclusion can be drawn. The rules, &c., of 

 the syllogism are, however, needlessly complex ; and we 

 do not require the syllogism in order to reason. 



The terms of a syllogism may be disposed in four 

 different ways or orders, termed the figures of the syllo- 

 gism. The first figure is the only one which has a uni- 

 versal affirmative for its conclusion, or which can prove 

 both a universal affirmative, a universal negative, a par- 

 ticular affirmative, and a particular negative. The third 

 proves only a particular affirmative or a particular nega- 

 tive. And the fourth is of but little value. According to Lam- 

 bert, a German logician, ' the first figure is best suited to 

 the disco very or proof of the properties of a thing; the second, 

 to discovery or proof of distinctions between things ; the 

 third, to discovery of instances and exceptions ; and the 

 fourth, to the discovery or exclusion of the different species 

 of genus.' The syllogism is not, however, of so much use 

 for the discovery of truth as for the purpose of arguing. 



The truth of one proposition interferes with that of 

 another having the same subject and predicate, and pro- 

 duces what is termed ' conflicting evidence,' which can 

 only be settled by rectifying the data. In some cases, 

 one proposition proves the truth of another ; in others, it 

 disproves it ; and in others, renders it doubtful. For 

 example, a universal affirmative disproves both a universal 

 negative and a particular one, but proves a particular 



