Chap. IX. ANTIENT METAPHYSICS. ijj 



is affirmed or denied, is called the Praedicate ; and that of which it 

 is affirmed or denied, is called the SiibjecSt. The praedicate be- 

 ing a more general idea than the fubjedt of which it is praedicated, 

 muft contain or include it, if it be an affirmative propofition ; or if 

 it be a negative propofition, it muft exclude it. This is the nature 

 of proportions * : And, as to Syllogifm, the ufe of it is to prove 

 any propofition that is not felf-evident. And this is done by find- 

 ing out what is called a middle term\ that is a term connedled with 

 both the praedicate and the fubje6t of the propofition to be proved. 

 Now, the propofition to be proved here is, that man is a fubjla?icc; 

 or, in other words, that fuhjlance can be praedicated of man : And 

 the middle term, by which this connexion is difcovered, is anima/^ 

 of which fubftance is praedicated; and this is the major propofition 

 of the fyllogifm, by which the major term of the propofition, to be 

 proved, is praedicated of the middle term. Then animal is praedi- 

 cated of man ; and this is the minor propofition of the fyllogifm, by 

 which the middle term is praedicated of the lefler term, or fubjed of 

 the propofition to be proved. The conclufion, therefore, is, that 

 as fubftance contains animal, and man is contained in animal, or is 

 part of animal, \\\qxq^oxq fubftance contains man. And the conclufion 

 is necefiiarily deduced from the axiom I have mentioned, as the foun-* 

 dation of the truth of the fyllogifm, " That the whole is greater than 

 " any of its parts, and contains them all :" So that the truth of the 

 fyllogifm is as evident as when we fay, that if A contain B, and B 

 contain C, then A contains C f . 



In this manner Ariftotle has demonftratcd the truth of the fyllo- 

 gifm. But a man, who has not ftudied his logic, can no more tell 

 why he believes the truth of the fyllogifm above mentioned, con- 

 cerning i7ian being a fubJla?icCy than a joiner, or any common me- 



VoL. V. U chanic, 



* Sec a great deal more concerning propofitions, and the inaccuracy of onr language 

 in exprefling them, in vol. i. of this work, p. ^IS' 



t See what I have faid on this fubjeft, in vol. 5. of Origin of Languag^e, p. 358. & 359. 



