188 



Mr DE morgan, ON THE SYLLOGISM, No. Ill, 



The Port Royal Logic is said to be the first modern work in which the distinction is in- 

 sisted on ; the use made of it is not very extensive. But it is correctly conceived. It is said, 

 for example, that the attribute (predicate) of an affirmative proposition is affirmed according to 

 its whole comprehension : and also that the negative proposition does not separate from the 

 subject all the parts contained in the comprehension of the attribute: meaning, for instance, 

 that to say Bucephalus is not man does not separate animal from Bucephalus. I do not 

 think that many works carry on this distinction : for, had there been many, I suppose my own 

 research, limited as it is, would have detected more than one. But I have only* found it in 

 the Institutiones Philosophicce of J. Bouvier of Mans (third edition, Mans, 1830, 12mo). 



The logicians who have recently contended for the revival, or rather the full introduction, 

 of the distinction of extension and comprehension, have completely passed over the change of 

 the quantity, and, so far as I see, make it only the distinction of number of objects in a class, 

 and number of marks of class, or concrete qualities, one to each. So that, with them, 

 quantity of comprehension is the same thing as quantity of extension : six of one, half-a-dozen 

 of the other. For example, when one of them reads in extension ' all metals are some shining 

 things ' he converts it into a proposition of comprehension as follows — ' The notion of some 

 shining things attaches to the notion of all metals.'' It ought to be said that the whole notion 

 of shining, with all its components, badness of radiation, hurtfulness to the eye if too long 

 continued, fitness for mirrors, difficulty of good imitation in painting, &c. &c. all form a part, 

 and, as it happens (but this is not necessarily contained in the proposition), a part only, of 

 the notion metal : which contains besides, hardness, conduction of electricity, &c. &c. 



Another powerful writer sets out a table of descent from the summutn genus A, which con- 

 tains species E, which contains I, which contains 0, which contains U, which contains Y, which 

 contains the individuals z, z', z", &c. As in animal, mammal, man, European, Greek, Athenian 

 (j. e. Socrates, Plato, &c. &c.) Beginning with z, we are told this subject is (contains in it the 

 inherent attribute) some Y. Here Plato contains in himself as an attribute, some Athenian. 

 Now he is contained in the class as some (one) Athenian : but he contains in himself all attri- 

 butes essential to an Athenian. As an individual, he helps to make up Athenian (class of men): 

 but all that goes to the composition of an Athenian man is in him, the whole of every attribute 

 must be awarded to him as a quality. If an Athenian must talk Attic Greek, so must Plato; 

 if an Athenian must look down on a Spartan as a fellow not fit to talk to nor to dine with, so 

 must Plato. The next step is ' All Y is some U,' or ' All Athenian is some Greek.' But it 

 should be : — All the attribute Greek is part of what we think of under the attribute Athenian. 

 Any one would suppose my quotations were meant for class readings ; but it is not so : the 

 readings in extension are given as ' Some Y is z ' and ' Some U is all Y,' &c. 



It seems that by a strong bias in favour of the mathematical side of logic, the decompo- 

 sition of the attribute has been discarded in favour of the disaggregation of a class of 

 qualities. But whether this be done avowedly, in opposition to the ancient distinction of 



* Since this was written I find — or rather refind, for I must 

 have peen it — that a learned reviewer of" myself and others has 

 hinted an objection to the growing theory of his day, in a 

 manner which implies that he inclines to the old view. 1 find 



also that he admits it is the old view. But liis objection 

 is hesitatingly given, and with ah avowed expectation of 

 refutation ; though this I suspect to be rather deference than 

 doubt. 



