"To this I answer: The defect was, that though they had in their possession Facts and Ideas, the Ideas were not distinct and appropriate to the facts.

"The peculiar characteristics of scientific ideas, which I have endeavored to express by speaking of them as distinct and appropriate to the facts, must be more fully and formally set forth when we come to the philosophy of the subject. In the meantime, the reader will probably have no difficulty in conceiving that for each class of Facts there is some special set of Ideas, by means of which the facts can be included in general scientific, truths; and that these Ideas which may thus be termed appropriate, must be possessed with entire distinctness and clearness, in order that they may be successfully applied. It was the want of Ideas having this reference to material phenomena which rendered the ancient philosophers, with very few exceptions, helpless and unsuccessful speculators on physical subjects."1

The point which Dr. Whewell makes here seems to us exactly provided for in the third rule given above for legitimate hypotheses. The Greeks failed, because their conjectures were not in the lines of known uniformities of nature. They sought the causes of phenomena in abstract and general conceptions.

Important as is the function of hypotheses, it may yet be exaggerated. Thus, Professor Davis says:—

"It is equally obvious that all experimental observation is likewise dependent on supposition. A mere trial of possible combinations to see what will come of them, without the further suggestions of a suggested supposition, can elicit nothing, save by chance." 2

But it is plain that a chemist may take the contents of the stomach of a murdered man, and may test successively for arsenic, strychnine, and other poisons, with

1 Whewell's History of the Inductive Sciences, vol. i, pp. 83, 87. 2 Inductive Logic, p. 159.

out any hypothesis whatever; and that he will reach the truth just as quickly without an hypothesis as with one. In every chemical laboratory, students are taught a regular system of tests, by which any questionable substance may be quickly identified without an hypothesis. Indeed, the tendency of science is to dispense with hypotheses as guides in research, to cease asking nature "leading questions," and to carry investigations forward on plans that permit the facts to speak for themselves. It is a waste of time to frame an hypothesis before all of the facts that can be ascertained are in hand.

Dr. Fowler says:

"Even though a hypothesis may ultimately be discovered to be false, it may be of great service in pointing the way to a truer theory. Thus, as already remarked, the circular theory of planetary motion, and the supplementary theory of epicycles and eccentrics, undoubtedly contributed to the formation of the hypothesis which was eventually proved true. Kepler himself tried no less than nineteen different hypotheses before he hit upon the right one, and his ultimate success was, doubtless, in no slight degree due to his unsuccessful efforts. There is hardly any branch of science in which it might not be affirmed that without a number of false guesses true theories could never have been attained." 1

The service that a false hypothesis renders is rather moral then intellectual. The belief that one has found a clue to the truth tends to keep up courage, and courage is necessary to persistent work upon the facts. But the false hypothesis, in itself considered, is purely a disadvantage and waste of time; it is, like every false scent, a diversion from the right path. In searching

1 Inductive Logic, p. 99.

for something, we are not likely to strike upon it at the first effort; and therefore our false guesses may be said to be necessary to our success. Where there are a number of equal possibilities, one must begin somewhere, and go on proving negatives, until the right one is reached. If a paper is in the desk, and there are four drawers, one as likely to contain it as another, the successive hypotheses that it is in the first, second, and third, will keep us looking, and when they are exploded we shall know that it is in the fourth. There is no absolute way to escape the tedium of testing wrong hypotheses, but we are fortunate in proportion to the fewness of those that we make, and the best rule is to delay in making any conjecture as long as possible. Grant's disastrous charge at Cold Harbor was necessary to his final victory over Lee, simply in showing that if he was ever to conquer, it must be in some other way; this is all of the intellectual value that can ever attach to a false hypothesis.

CHAPTER XIV.

INDUCTIVE ARGUMENTS.

HAVING considered the elementary steps of inductive investigation, we now advance to the construction of inductive arguments.

A very common form of argument is that from Analogy. Such an argument is based upon a primary induction of a uniformity of resemblances. Having observed a certain object to have, in many respects, the property x, we come to think that we are upon the line of one of its uniformities, and that it will be found to have, in all respects, the property x. But may stand for resemblance to some other object.

As Bishop Butler has said:

"Probable evidence is essentially distinguished from demonstrative by this, that it admits of degrees, and of all variety of them, from the highest moral certainty to the very lowest presumption. We cannot, indeed, say a thing is probably true upon one very slight presumption for it; because, as there may be probabilities on both sides of the question, there may be some against it; and though there be not, yet a slight presumption does not beget that degree of conviction which is implied in saying that a thing is probably true. But that the slightest possible presumption is of the nature of a probability, appears from hence, that such low presumption, often repeated, will amount even to moral certainty. Thus, a man's having observed the ebb and flow of the tide to-day, affords some sort of presumption, though the lowest imaginable, that it may happen again to-morrow; but the observation of this event for so many days and months, and ages together, as it has been observed by mankind, gives us a full assurance that it will." 1

1 Introduction to the Analogy of Religion.

Now a uniformity of resemblances is just like any other line of uniformity, and the argument from it is the same. If I have often found a substance white, I begin to expect to find it of that color next time; and if I have found it to resemble another substance in many respects, I expect to find more resemblances. An argument from Analogy, therefore, does not differ in any way from an argument based upon any other primary induction. A primary induction may be made that the peach trees of a certain region yield a crop three seasons out of four; and this becomes the basis of expectation. Just so the induction may be made that two objects resemble each other in three respects out of four (or according to any other ratio), and this will measure the probability of resemblance in any unexamined instance.

The following example of the use of the argument from analogy is taken from the Scientific Papers of Asa Gray:

"The most interesting ideas connected with trees are those suggested by their stability and duration. They far outlast all other living things, and form the familiar and appropriate symbols of long-protracted existence. 'As the days of a tree shall be the days of my people' is one of the most beautiful and striking figures under which a blessing can be conveyed. We are naturally led to inquire, whether there is any absolute limit to their existence. If not destroyed by accident, that is, by extrinsic causes, of whatever sort, do trees eventually perish, like ourselves, from old age? It is commonly thought, no doubt, that trees are fully exposed to the inevitable fate of all other living things. The opposite opinion seems to involve a paradox, and to be contradicted by every one's observation. But popular opinion is an unsafe guide; - the more so in this case, as our ordinary conceptions on the subject spring from a false analogy, which we have