7. The seventh and last of these heads of Fallacy is, when the questioner puts two distinct questions in the same form of words, as if they were one — Fallacia Plurium Interrogationum ut Unius. In well-conducted Dialectic the respondent was assumed to reply either Yes or No to the question put; or, if it was put in the form of an alternative, he accepted distinctly one term of the alternative. Under such conditions he could not reply to one of these double-termed questions without speaking falsely or committing himself. Are the earth and the sea liquid? Is the heaven or the earth sea? The questions are improperly put, and neither admits of any one correct answer. You ought to confine yourself to one question at a time, with one subject and one predicate, making what is properly understood by one single proposition. The two questions here stated as examples ought properly to be put as four.39

Aristotle has thus distinguished and classified Fallacies under thirteen distinct heads in all — six In Dictione, and seven Extra Dictionem; among which last one is Ignoratio Elenchi. He now proceeds to show that, in another way of looking at the matter, all the Fallacies ranged under the thirteen heads, may be shown to be reducible to this single one — Ignoratio Elenchi. Every Fallacy, whatever it be, transgresses or fails to satisfy, in some way or other, the canons or conditions which go to constitute a valid Elenchus,40 or a valid Syllogism. For a true Elenchus is only one mode of a true Syllogism; namely, that of which the conclusion is contradictory to some given thesis or proposition.41 With this particular added, the definition of a valid Syllogism will also be the definition of a good Elenchus. And thus Ignoratio Elenchi — misconception or neglect of the conditions of a good Elenchus — understood in its largest meaning, is rather a characteristic common to all varieties of Fallacy, than one variety among others.42

In regard to two among the thirteen heads — Fallacia Accidentis and Fallacia Consequentis (which however ought properly to rank as only one head, since the second is merely a particular variety of the first) — Aristotle’s observations are remarkable. After having pointed out that a Syllogism embodying this fallacy will not be valid or conclusive (thus showing that it involves Ignoratio Elenchi), he affirms that even scientific men were often not aware of it, and conceived themselves to be really refuted by an unscientific opponent urging against them such an inconclusive syllogism. To take an example:— Every triangle has its three angles equal to two right angles; every triangle is a figure; therefore, every figure has its three angles equal to two right angles.43 Here we have an invalid syllogism; for it is in the Third figure, and sins against the conditions of that figure, by exhibiting an universal affirmative conclusion: it is a syllogism properly concluding in Darapti, but with conclusion improperly generalized. Yet Aristotle intimates that a scientific geometer of his day, in argument with an unscientific opponent, would admit the conclusion to be well proved, not knowing how to point out where the fallacy lay: he would, if asked, grant the premisses necessary for constructing such a syllogism; and, even if not asked, would suppose that he had already granted them, or that they ought to be granted.44

The passage affords us a curious insight into the intellectual grasp of the scientific men contemporary with Aristotle. Most of them were prepared to admit fallacious inferences (such as the above) which assumed the interchangeability of subject and predicate. They had paid little or no attention to the logical relations between one proposition and another, and between the two different terms of the same proposition. The differences of essential from accidental predication, and of each among the five Predicables from the others, must have been practically familiar to them, as to others, from the habit of correct speaking in detail; but they had not been called upon to consider correct speaking and reasoning in theory, nor to understand upon what conditions it depended whether the march of their argumentative discourse landed them in true or false results. And, if even the scientific men were thus unaware of logical fallacies, we may be sure that this must have been still more the case with unscientific men, of ordinary intelligence and education. Aristotle tells us here, in more than one passage, how widespread such illogical tendencies were: to fancy that two subjects which had one predicate the same must be the same with each other in all respects;45 to understand each predicate applied to a subject as being itself an independent subject, implying a new Hoc Aliquid or Unum;46 to treat the universal, not as a common epithet but, as a substantive and singular apart;47 to use equivocal words or phrases, even the most wide and vague, without any attempt to discriminate their various meanings.48 Such insensibility to the conditions of accurate reasoning prevailed alike among ordinary men and among the men of special science. A geometer would be imposed upon by the inconclusive syllogism stated in the last paragraph, which, as being founded on the Fallacia Accidentia (or interchangeability of subject and predicate), Aristotle numbers among Sophistical Refutations. Such a refutation, however, even when successful, would not at all prove that the geometer was deficient in knowledge of his own science;49 for it would puzzle the really scientific man as well as the pretender.

We must always recollect that Aristotle was the first author who studied the logical relations between Terms and Propositions, with a view to theory and to general rules founded thereupon. The distinctions which he brought to view were in his time novelties; even the simplest rules, such as those relating to the Conversion of propositions, or to Contraries and Contradictories, had never been stated in general terms before. Up to a certain point, indeed, acquired habit, even without these generalities, would doubtless lead to correct speech and reasoning; yet liable to be perverted in many cases by erroneous tendencies, requiring to be indicated and guarded against by a logician. When we are told that even a professed geometer was imposed upon by these fallacies, we learn at once how deep-seated were such illogical deficiencies, how useful was Aristotle’s theoretical study in marking them out, and how insufficient was his classification when he described the Fallacies as obvious frauds, broached only by dishonest professional Sophists. As he himself states, the cause of deceit turns upon a quite trifling difference; having its root in the imperfection of language and in our frequent habit of using words without much attention to logical distinctions.50

Under one or other, then, of the thirteen general heads above enumerated, all Paralogisms must be included — merely apparent syllogisms, or refutations, which are not real and valid;51 and all of them designated by Aristotle as sophistic or eristic. Besides these, moreover, he includes, as we saw, under the same designation, syllogisms or refutations valid in form, and true as to conclusion, yet founded on premisses not suited to the matter in debate; i.e., not suited to Dialectic. Now, here it is that difficulty arises. Dialectic and Rhetoric are carefully distinguished by Aristotle from all the special sciences (such as Geometry, Astronomy, Medicine, &c.); and are construed as embracing every variety of authoritative dicta, current beliefs, and matters of opinion, together with all the most general maxims and hypotheses of Ontology and Metaphysics, of Physics and Ethics, and the common Axioms assumed in all the sciences, as discriminated from what is special and peculiar to each. Construed in this way, we might imagine that the subject-matter of Dialectic was all-comprehensive, and that every thing without exception belonged to it, except the specialties of Geometry and of the other sciences; and such is the usual language of Aristotle. Yet in the treatise before us we find him exerting himself to establish another classification, and to part off Dialectic from a certain other science or art which he acknowledges under the title of Sophistic or Eristic.52 Elsewhere he describes Sophistic as occupied in the study of accidents or occasional conjunctions; and this characteristic feature parts it off from Demonstration and Science. But there is greater difficulty when he tries to part it off from Dialectic. Where are we to find a clear line of distinction between the matter of dialectic debate (gymnastic or testing) on the one hand, and the matter of debate sophistic or litigious, on the other? At the beginning of the Topica Aristotle assigned, as the distinction, that the Dialectician argues upon premisses really probable, while the litigious Sophist takes up premisses which are probable in appearance only, and not in reality; such apparent probabilia (he goes on to say) having only the most superficial semblance of truth, and being seen immediately to be manifest falsehoods by persons of very ordinary intelligence.53 But I have already pointed out that this description of apparent probabilia, if considered as applying to fallacious reasoning generally, is both untenable in itself, and contradicted by Aristotle himself elsewhere. The truth is, that there is no clear distinction between the matter of Dialectic and the matter of Sophistic. And so, indeed, Aristotle must be understood to admit, when he falls back upon an alleged distinction of aim and purpose between the practitioners of one and the other. The litigious man (he tells us) is bent upon nothing but victory in debate, per fas et nefas: the Sophist aims at passing himself off falsely for a wise or clever man, and making money thereby.54

Now, in regard to the distinction of aim or disposition drawn by Aristotle between the dialectical disputant and the litigious or sophistic disputant, we see at once, as was before suggested, that it lies apart from the critical estimate of art, science, or philosophy; and that it belongs, so far as it is well founded, to the estimate of individuals ethically and politically, as worthy men or patriotic citizens. Whether Euripides or Sophokles composed finer tragedies (as we find argued in the Ranæ of Aristophanes), must be decided by examining the tragedies themselves, not by enquiring whether one of them was vain and greedy of money, the other free from these blemishes. A theorist who is laying down general principles of Rhetoric, and illustrating them by the study of Æschines and Demosthenes, will appreciate the oration against Ktesiphon and the oration De Coronâ in their character of compositions intended for a particular purpose. For Rhetoric it is of no moment whether Æschines was venal or disinterested — a malignant rival or an honest patriot; this is an enquiry important indeed, but belonging to the historian and not to the rhetorical theorist. Whether Aristotle was or was not guided, in his animadversions on Plato, by an unworthy and captious jealousy of his master, is an interesting question in reference to his character; but our appreciation of his philosophy must proceed upon an examination, not of his motives but, of his doctrines and reasonings as we find them. A good argument is not deprived of its force when enunciated by a knave, nor is a bad argument rendered good because it proceeds from a virtuous man. Indeed, so far as the character of the speaker counts at all, in falsifying the fair logical estimate of an argument, it operates in a direction opposite to that here indicated by Aristotle. The same argument in the mouth of one who is esteemed and admired counts for more than its worth; in the mouth of a person of low character it counts for less than it is worth.55 To distribute arguments into two classes — those employed by persons of dishonourable character and those employed by honourable men — is a departure from the scientific character of Logic.

As to the other part of the case (if it is still necessary to recur to it), touching the peculiarity of the matter of sophistical arguments, the inconsistency of Aristotle is most apparent. In enumerating the Sophistical Refutations he tells us that these fallacies are indeed sometimes palpable and easily detected, but that they are often very difficult to detect and very misleading; that an unprepared hearer will generally be imposed upon by several of them, and even a scientific hearer by some; and that, even where the fallacy does not actually deceive, the proper mode of meeting and exposing it will not occur unless to one previously exercised in Dialectic.56 That Fallacies In Dictione, taken as a class (though these are what he declares to be the most usual modus operandi of the sham dialecticians called Sophists57), often passed unperceived, and were hard to solve and elucidate even when perceived — we know to have been his opinion; for it is not only in the Topica and Sophistici Elenchi, but also in the Metaphysica and other works,58 that he takes pains to analyse and discriminate the several distinct meanings borne by terms familiar to every one, such as idem, unum, pulchrum, bonum, amare, album, acutum, &c., which terms therefore, when employed in argument, were always liable to introduce a fallacy of Equivocation or Amphiboly. He tells us the like in specifying the seven Fallacies Extra Dictionem: that they also were often unnoticed, and required vigilant practice to see through and solve. The description in detail, therefore, which Aristotle gives (in Sophistici Elenchi) of the working process peculiar to the litigious Sophist, is completely at variance with the definition which he had given of the sophistic syllogism at the commencement of the Topica. That definition is indeed suitable for the type-specimens which he and other logicians give to illustrate this or that class of Fallacies: the type-specimen produced must carry absurdity on the face of it, so that the reader may at first sight recognize it as a fallacy; and he may even find difficulty in believing that any one can really be imposed upon by such trifling. But, though suitable for the type-specimen taken separately, this definition fails in the essential character which Aristotle postulates for a definition, since it is quite untrue and unsuitable for numerous instances of the class intended to be illustrated.59 Aristotle was the first who attempted to distribute Fallacies into classes, such that, while in each class there were certain specimens palpably stamped with the fallacious character, there were also in each class an indefinite multitude of analogous cases wherein the fallacious character did not reveal itself openly or easily, but required attentive consideration to detect it, often indeed remaining undetected, and producing its natural fruit of error and confusion. This was one of his many great merits in regard to Logic; and the classification of Fallacies (modified as to details) has passed to all subsequent logicians, so that we find difficulty in understanding that the contemporaries of Sokrates and Plato had no idea of it. But the value of his service to Logic would be much lessened, if all fallacies were sophistic syllogisms, intended to deceive but never really deceiving, corresponding to his definition at the beginning of the Topica; if (as he tells us in the Sophistici Elenchi) they were only impudent forgeries put in circulation by a set of professional knaves called Sophists; and if all non-sophistical dialecticians, and all the world without, could be trusted as speaking correctly by nature and as never falling into them.

The appeal made by Aristotle to a difference of character and motives as the distinction between the Dialectician and the Sophist is all the more misplaced, because he himself lays down as the essential feature of Dialectic generally, that it is a match or contention between two rivals, each anxious to obtain the victory. It is like a match at chess between two expert players, or a fencing-match between two celebrated masters at arms. Its very nature is to be an attack and defence, in which each combatant resorts to stratagem, and each outwits the other if he can. Whether the match is played for money or for nothing — whether the contentious spirit is more or less intense — does not concern the theorist on dialectical procedure. It is indispensable that both the questioner and the respondent should exert their full force, the one in thrusting, the other in parrying: if they do not, the purpose of Dialectic, which is the common business of both, will not be attained. That purpose is clearly declared by Aristotle. It is not didactic: he distinguishes it expressly from teaching,60 where one man who knows communicates such knowledge to an ignorant pupil. It is gymnastic, exercising the promptitude and invention of both parties; or peirastic, testing whether the respondent knows a given thesis in such manner as to avoid being driven into answers inconsistent with each other or notoriously false.61 Each party seeks, not to help or enlighten but, to puzzle and defeat the other. As at chess or in fencing, to mask one’s projects and deceive the adversary is essential to the work and to its purpose; each expects it from the other, and undertakes to meet and parry it. The theses debated were always such that arguments might be found both for the affirmative and for the negative.

According to Aristotle himself, therefore, the Dialectician is agonistic and eristic, just as much as the Sophist. If the one tries to entrap his opponent for the purpose of victory, so also does the other: the line which Aristotle draws between them is one not founded upon any real distinction between two purposes and modes of procedure, but is merely verbal and sentimental; putting aside under a discredited title what he himself disliked. He admits that the dialectical questioner, whenever the thesis which he undertakes to refute is true, can never refute it except by inducing the respondent to concede what is false; that, even where the thesis is false, he often can only refute it by some other incompatible falsehood, because he cannot obtain from the respondent better premisses; that, where the thesis is probable and conformable to received opinion, his only way of refuting it is by entrapping the respondent into concessions paradoxical and contrary to received opinion.62 But these ends — fallacious refutation, falsehood, and paradox — are the very same as those which Aristotle (in the Sophistici Elenchi)63 sets forth as the peculiar characteristics of the litigious Sophist. And the improving intellectual tendencies which he ascribes to Sophistic, are almost identical with those attributed to Dialectic, being declared in very similar words.64 That there were dialecticians of every degree of merit, in the time of Aristotle, cannot be doubted; some clever and ready, others stupid and destitute of invention. But that there were any two classes of dialecticians such as he describes and contrasts — one heretical class, called Sophists, who purposely and habitually employed the thirteen fallacious refutations, and another orthodox class who purposely avoided or habitually abstained from them — we may most reasonably doubt. If the argument in the Sophistici Elenchi is good at all, it is good against all Dialectic. The Sophist, as Aristotle describes him, is only the Dialectician looked at on the unfavourable side and painted by an enemy. We know that there were in Greece many enemies of Dialectic generally; the intense antipathy inspired by the cross-examining colloquy of Sokrates, and attested by his own declarations, is a sufficient proof of this. The enemies of Sokrates depicted him — as Aristotle depicts the Sophist in the Sophistici Elenchi — as a clever fabricator of fallacious contradictions and puzzles; to which Aristotle adds the farther charge (advanced by Plato before him) against the Sophist, of arguing for lucre — which is an irrelevant charge, travelling out of the region of art, and bearing on the personal character of the individual. If the sophistical stratagems were discreditable and mischievous when exhibited for money, they would be no less such if exhibited gratuitously. The sophistical discourse is not (as Aristotle would have us believe) generically distinguishable from the dialectical;65 nor is Sophistic an art distinct from Dialectic while adjoining to it, but an inseparable portion of the tissue of Dialectic itself.66 If the Sophist passed himself off as knowing what he did not really know, so also did the Dialectician; as we know from the testimony of Sokrates, the most consummate master of the art. The conflict of two minds each taking advantage of the misconceptions, short-comings, and blindness of the other, is the essential feature of Dialectic as Aristotle conceives it; to which the eight books of his Topica are adapted, with their multiplicity of distinctions and precepts both for attack and defence. There cannot be a game of chess without stratagems, nor a fencing-match without feints; the power of such aggressive deception is one characteristic mark of a good player. Those who teach or theorize on the game do not seek to exclude stratagem, but furnish precautions to prevent it from succeeding. Mastery of the art assumes skill in defence as well as in attack.