348 THE SCIENCE OF LOGIC 



perhaps furnish somewhat more evidence than one cow ; but then, too, five 

 instances differing otherwise from each other a cow, a goat, a sheep, a 

 deer, a camel would furnish far stronger evidence than the former fifty : 

 the nature of the instances is more important than their number. &quot; Again, 

 we might meet a Privy Councillor in a light suit, and yet not be led to regard 

 the next man we met in a light suit as a Privy Councillor ; but if we met a 

 guardsman in a breastplate, we should very likely suppose the next man in a 

 breastplate to be a guardsman &quot; : 1 the nature of the attributes here deter 

 mines the difference. These considerations belong properly, as we shall 

 see, to the logic of induction. 



If we interpret the particular conclusion &quot; Some S s are (or are not) P &quot; 

 not modally, but assertorically, as referring to an unspecified portion of the 

 class 5, and ask ourselves of what portion of the class 5 is P affirmed (or 

 denied), we must answer : of those things that were found in the premisses to 

 be characterized not only by 5 but by M and by P (or the absence of P} and 

 of course the extent of this class, or, in other words, the extension of the con 

 clusion, will vary with the extent of the NTs referred to in both premisses. 

 If I learned that all horned animals have cloven feet, and also that they all 

 chew the cud, and if I infer from this that &quot; some ruminants are cloven 

 footed,&quot; I know the extent of the &quot; some &quot; in my conclusion : that it refers 

 to all the ruminants that are horned ; and if I keep this thought in my con 

 clusion, if my conclusion really is that &quot; all ruminants that have horns have 

 cloven feet &quot; or that &quot; all horned animals chew the cud and have cloven 

 feet,&quot; I am merely adding up the information contained in the two premisses, 

 and I have no syllogism at all. But if I eliminate the middle term 

 (&quot; horned &quot;) from my thought, and retain only the indefinite judgment that 

 &quot; some ruminants are cloven-footed,&quot; meaning thereby that these attributes 

 are compatible, I have a genuine syllogism. Of course, in all such syllogisms 

 the ground of the conclusion lies in the examined instances ; but the ground 

 for a judgment forms no necessary part of the meaning or import of the 

 judgment. 



We have now to ask whether the third figure gives us a type of 

 inference distinct from those already examined. And the answer will 

 be in the affirmative. In the third figure our inference manifestly turns on 

 the instances cited. It is because 5 and P are both present (or one 

 present and the other not) in the same instances, that we have a right 

 to connect 5 and P (affirmatively or negatively) in our conclusion. It 

 matters not whether &quot; all ATs &quot; [or &quot; most M s &quot;] are cited in both pre 

 misses, or &quot; all &quot; [or &quot; most &quot;] in one premiss and &quot; some &quot; on the other, 

 our conclusion turns on the relations found to obtain between the extremes 

 and the same M s : even if &quot; Some M*s &quot; were cited in both premisses, we 

 could still derive our conclusion did we know that the same &quot; some &quot; were 

 cited in both premisses. 2 It is only upon those M^s that are compared both 



1 JOSEPH, op. cit., p. 309. 



2 By giving the &quot; Some M &quot; a definite class-name, say &quot; N &quot; and substituting 

 All N &quot; in both premisses, we reduce all the moods of the third figure to Darapti 

 and Felapton. Bocardo may thus be directly reduced, through Felapton, to Fcrio. 

 Similarly Baroco of the second figure may be reduced, through Camestres, to 

 Cesart. 



