(in the 1st Fig.) incompatible with correctness. (See Chap. iii. § 4.)

Sorites.

Hypothetical A string of Conditional Syllogisms may in like manner be abridged into a Sorites; e. g. if A is B, C is D; if C is D, E is F; if E is F, G is H; but A is B, therefore G is H. "If the Scriptures are the word of God, it is important that they should be well explained; if it is important, &c. they deserve to be diligently studied: if they deserve, &c. an order of men should be set aside for that purpose; but the Scriptures are the word, &c.; therefore an order of men should be set aside for the purpose, &c.: "* in a destructive Sorites, you, of course, go back from the denial of the last consequent to the denial of the first antecedent: "G is not H; therefore A is not B."

Induction.

Example.

Those who have spoken of Induction or of Example, as a distinct kind of argument in a Logical point of view, have fallen into the common error of confounding Logical with Rhetorical distinctions, and have wandered from their subject as much as a writer on the orders of Architecture would do who should introduce the distinction between buildings of brick and of marble. Logic takes no cognizance of Induction, for instance, or of à priori reasoning, &c., as distinct Forms of argument; for when thrown into the syllogistic form, and when letters of the alphabet are substituted for the Terms (and it is thus that an argument is properly to be brought under the cognizance of Logic), there is no distinction between

*Hence it is evident how injudicious an arrangement has been adopted by former writers on Logic, who have treated of the Sorites and Enthymeme before they entered on the subject of Hypotheticals.

them; e. g. a "Property which belongs to the ox, sheep, deer, goat, and antelope, belongs to all horned animals ; rumination belongs to these; therefore to all." This, which is an inductive argument, is evidently a Syllogism in Barbara. The essence of an inductive argument (and so of the other kinds which are distinguished from it) consists not in the form of the Argument, but in the relation which the Subject-matter of the Premises bears to that of the Conclusion.*

tions.

3d. There are various other abbreviations Abbreviacommonly used, which are so obvious as hardly to call for explanation: as where one of the Premises of a Syllogism is itself the Conclusion of an Enthymeme which is expressed at the same time: e. g. "All useful studies deserve encouragement; Logic is such (since it helps us to reason accurately,) therefore it deserves encouragement;" here the minor Premiss is what is called an Enthymematic sentence. The antecedent in that minor Premiss (i. e. that which makes it Enthymematic) is called by Aristotle the Prosyllogism.

It is evident that you may, for brevity, substitute for

any term an equivalent; as in the last exam- Equivalents.

ple, "it," for "Logic;" "such," for "a use

ful study," &c. The doctrine of Conversion, laid down in the Second Chapter, furnishes many equivalent propositions, since each is equivalent to its illative converse. The division of nouns also (for which see Chap. v.) sup

* See Rhetoric, Part I. Ch. ii. § 6. Nothing probably has tended more to foster the prevailing error of considering Syllogism as a particular kind of argument, than the inaccuracy just noticed, which appears in all or most of the logical works extant. See Dissertation on the Province of Reasoning, Ch. i.

plies many equivalents; e. g. if A is the genus of B, B must be a species of A: if A is the cause of B, B must be the effect of A.

Syllogisms apparently incorrect.

66

4th. And many Syllogisms, which at first sight appear faulty, will often be found, on examination, to contain correct reasoning, and, consequently, to be reducible to a regular form; e. g. when you have, apparently, negative Premises, it may happen, that by considering one of them as affirmative, (see Chap. ii. § 4, p. 59), the Syllogism will be regular: e. g. no man is happy who is not secure : no tyrant is secure; therefore no tyrant is happy," is a Syllogism in Celarent.* Sometimes there will appear to be too many terms; and yet there will be no fault in the Reasoning, only an irregularity in the expression: e. g. "no irrational agent could produce a work which manifests design; the universe is a work which manifests design; therefore no irrational agent could have produced the universe." Strictly speaking, this Syllogism has five terms; but if you look to the meaning, you will see, that in the first Premiss (considering it as a part of this Argument) it is not, properly," an irrational agent" that you are speaking of, and of which you predicate that it could not produce a work manifesting design; but rather it is this "work," &c."

66

* If this experiment be tried on a Syllogism which has really negative Premises, the only effect will be to change that fault into another: viz. an excess of Terms, or (which is substantially the same) an undistributed middle; e. g. an enslaved people is not happy; the English are not enslaved; therefore they are happy:" if "enslaved" be regarded as one of the Terms, and "not enslaved " as another, there will manifestly be four. Hence you may see how very little difference there is in reality between the different faults which are enumerated.

of which you are speaking, and of which it is predicated that it could not be produced by an irrational agent; if, then, you state the Propositions in that form, the Syllogism will be perfectly regular. (See § 1, of this Supplement.)

66

Thus, such a Syllogism as this, " every true patriot is disinterested; few men are disinterested; therefore few men are true patriots; " might appear at first sight to be in the second Figure, and faulty; whereas it is Barbara, with the Premises transposed: for you do not really predicate of "few men," that they are "disinterested," but of "disinterested persons," that they are "few." Again, none but candid men are good reasoners; few infidels are candid; few infidels are good reasoners." In this it will be most convenient to consider the major Premiss as being, "all good reasoners are candid," (which of course is precisely equipollent to its illative converse by negation;) and the minor Premiss and Conclusion may in like manner be fairly expressed thus "most infidels are not candid; therefore most infidels are not good reasoners:" which is a regular Syllogism in Camestres.* Or, if you would state it in the first Figure, thus: "those who are not candid (or uncandid) are not good reasoners; most infidels are rot candid; most infidels are not good reasoners."

The reader is to observe that the term employed as the Subject of the minor premiss, and of the conclusion, is "mostinfidels:" he is not to suppose that "most" is a sign of distribution; it is merely a compendious expression for "the

greater part of."

CHAP. V.

SUPPLEMENT TO CHAP. I.

[This Supplement may be studied either before or after the Compendium.]

§ 1.

THE usual divisions of nouns into univocal, equivocal, and analogous, and into nouns of the first and second intention, are not, strictly speaking, divisions of words, but divisions of the manner of employing them; the same word may be employed either univocally, equivocally, or analogously; either in the first intention or in the second. The ordinary logical treatises often occasion great perplexity to the learner, by not noticing this circumstance, but rather leading him to suppose the contrary. (See Book III. § 8.) Some of those other divisions of nouns, which are the most commonly in use, though not appropriately and exclusively belonging to the Logical system, i. e. to the theory of reasoning, it may be worth while briefly to notice in this place.

Let it be observed then, that a noun expresses the view we take of an object. And its being viewed as an object, i. e. as one, or again as several, depends on our arbitrary choice; e. g. we may consider a troop of cavalry as one object; or we may make any single horse with its rider, or any separate man or horse, or any limb of either, the subject of our thoughts.

Singular and Common terms,

as

1. When then any one object is considered according to its actual existence, as numerically one, the noun denoting it is called Singular; "this tree," the "city of London," &c. When it is