Essays in Experimental Logic

by John Dewey

By Aristotle's time the interest was not so much in the existence of standards of decision in cases of doubt and dispute as in the technique of their use. The judge was firmly seated on the bench. The parties in controversy recognized his jurisdiction, and their respective claims were submitted for adjudicature. The need was for rules of procedure by which the judge might, in an obvious and impartial way, bring the recognized universal or decisive law to bear upon particular matters. Hence the elaboration of those rules of evidence, those canons of demonstrative force, which are the backbone of the Aristotelian logic. There was a code by which to decide upon the admissibility and value of proffered testimony—the rules of the syllogism. The figures and terms of the syllogism provided a scheme for deciding upon the exact bearing of every statement propounded. The plan of arrangement of major and minor premises, of major, minor, and middle terms, furnished a manifesto of the exact procedure to be followed in determining the probative force of each element in reasoning. The judge knew what testimony to permit, when and how it should be introduced, how it could be impeached or have its competence lessened, and how the evidence was to be arranged so that a summary would also be an exhibit of its value in establishing a conclusion.

This means that there now is a distinctive type of thinking marked off from mere discussion and reflection. It may be called either reasoning or proof. It is reasoning when we think of the regularity of the method for getting at and employing the unquestioned grounds which give validity to other statements. It is proof as regards the degree of logical desert thereby measured out to such propositions. Proof is the acceptance or rejection justified through the reasoning. To quote from Mill: "To give credence to a proposition as a conclusion from something else is to reason in the most extensive sense of the term. We say of a fact or statement, it is proved, when we believe its truth by reason of some other fact or statement from which it is said to follow."[47] Reasoning is marshaling a series of terms and propositions until we can bind some doubtful fact firmly to an unquestioned, although remote, truth; it is the regular way in which a certain proposition is brought to bear on a precarious one, clothing the latter with something of the peremptory quality of the former. So far as we reach this result, and so far as we can exhibit each step in the nexus and be sure it has been rightly performed, we have proof.

But questions still face us. How about that truth upon which we fall back as guaranteeing the credibility of other statements—how about our major premise? Whence does it derive its guaranty? Quis custodes custodiet?

We may, of course, in turn subsume it under some further major premise, but an infinite regress is impossible, and on this track we are finally left hanging in the air. For practical purposes the unquestioned principle may be taken as signifying mutual concession or agreement—it denotes that as a matter of fact its truth is not called in question by the parties concerned. This does admirably for settling arguments and controversies. It is a good way of amicably arranging matters among those already friends and fellow-citizens. But scientifically the widespread acceptance of an idea seems to testify to custom rather than to truth; prejudice is strengthened in influence, but hardly in value, by the number who share it; conceit is none the less self-conceit because it turns the heads of many.

Great interest was indeed afterward taken in the range of persons who hold truths in common. The quod semper ubique omnibus became of great importance. This, however, was not, in theory at least, because common agreement was supposed to constitute the major premise, but because it afforded confirmatory evidence of its self-evident and universal character.

Hence the Aristotelian logic necessarily assumes certain first or fundamental truths unquestioned and unquestionable, self-evident and self-evidencing, neither established nor modified by thought, but standing firm in their own right. This assumption was not, as modern dealers in formal logic would sometimes have it, an external psychological or metaphysical attachment to the theory of reasoning, to be omitted at will from logic as such. It was an essential factor of knowledge that there should be necessary propositions directly apprehended by reason and particular ones directly apprehended by sense. Reasoning could then join them. Without the truths we have only the play of subjective, arbitrary, futile opinion. Judgment has not taken place, and assertion is without warrant. Hence the scheduling of first truths is an organic part of any reasoning which is occupied with securing demonstration, surety of assent, or valid conviction. To deny the necessary place of ultimate truths in the logical system of Aristotle and his followers is to make them players in a game of social convention. It is to overlook, to invert, the fact that they were sincerely concerned with the question of attaining the grounds and process of assurance. Hence they were obliged to assume primary intuitions, metaphysical, physical, moral, and mathematical axioms, in order to get the pegs of certainty to which to tie the bundles of otherwise contingent propositions.

It would be going too far to claim that the regard for the authority of the church, of the fathers, of the Scriptures, of ancient writers, of Aristotle himself, so characteristic of the Middle Ages, was the direct outcome of this presupposition of truths fixed and unquestionable in themselves. But the logical connection is sure. The supply of absolute premises that Aristotle was able to proffer was scant. In his own generation and situation this paucity made comparatively little difference; for to the mass of men the great bulk of values was still carried by custom, by religious belief, and social institution. It was only in the comparatively small sphere of persons who had come under the philosophic influence that need for the logical mode of confirmation was felt. In the mediaeval period, however, all important beliefs required to be concentrated by some fixed principle giving them stay and power, for they were contrary to obvious common-sense and natural tradition. The situation was exactly such as to call into active use the Aristotelian scheme of thought. Authority supplemented the meagerness of the store of universals known by direct intuition, the Aristotelian plan of reasoning afforded the precise instrumentality through which the vague and chaotic details of life could be reduced to order by subjecting them to authoritative rules.

It is not enough, however, to account for the ultimate major premises, for the unconditioned grounds upon which credibility is assigned. We have also to report where the other side comes from: matters so uncertain in themselves as to require that they have their grounds supplied from outside. The answer in the Aristotelian scheme is an obvious one. It is the very nature of sense, of ordinary experience, to supply us with matters which in themselves are only contingent. There is a certain portion of the intellectual sphere, that derived from experience, which is infected throughout by its unworthy origin. It stands forever condemned to be merely empirical—particular, more or less accidental, inherently irrational. You cannot make gold from dross, and the best that can be done for and with material of this sort is to bring it under the protection of truth which has warrant and weight in itself.

We may now characterize this stage of thinking with reference to our original remark that different stages denote various degrees in the evolution of the doubt-inquiry function. As compared with the period of fixed ideas, doubt is awake, and inquiry is active, but in itself it is rigidly limited. On one side it is bounded by fixed ultimate truths, whose very nature is that they cannot be doubted, which are not products or functions in inquiry, but bases that investigation fortunately rests upon. In the other direction all "matters of fact," all "empirical truths" belong to a particular sphere or kind of existence, and one intrinsically open to suspicion. The region is condemned in a wholesale way. In itself it exhales doubt; it cannot be reformed; it is to be shunned, or, if this is not possible, to be escaped from by climbing up a ladder of intermediate terms until we lay hold on the universal. The very way in which doubt is objectified, taken all in a piece, marks its lack of vitality. It is arrested and cooped up in a particular place. As with any doubtful character, the less of its company the better. Uncertainty is not realized as a necessary instrument in compelling experienced matters to reveal their meaning and inherent order.

This limitation upon inquiry settles the interpretation to be given thought at this stage—it is of necessity merely connective, merely mediating. It goes between the first principles—themselves, as to their validity, outside the province of thought—and the particulars of sense—also, as to their status and worth, beyond the dominion of thought. Thinking is subsumption—just placing a particular proposition under its universal. It is inclusion, finding a place for some questioned matter within a region taken as more certain. It is use of general truths to afford support to things otherwise shaky—an application that improves their standing, while leaving their content unchanged. This means that thought has only a formal value. It is of service in exhibiting and arranging grounds upon which any particular proposition may be acquitted or condemned, upon which anything already current may be assented to, or upon which belief may reasonably be withheld.