between natural and artificial instances, that is, between experiments and instances which are not experiments, always clearly traceable. All of the arrangements of human life and society are artificial; we learn from them to our cost, and often, in consequence, change our methods. Popular government is frequently spoken of as still an experiment; the construction of our armored battle ships is experimental. Yet instances of this kind are not arranged for the sake of learning from them, although with the expectation of learning, and improving.

The primary rule for any inductive thinking is to make sure of the observations. Starting with prejudices, guesses, or inferences, the truth never can be reached. Nothing but observation can establish a hitherto unknown fact. The explanation of the slow advance of science in ancient and medieval times may be found mainly in the neglect of this simple rule. spite of many errors in methods of thinking, the men of those times would have discovered a vast body of facts, if they had only given attention to them.

In

But the making of a precise and trustworthy observation is by no means the easy thing which at first it seems to be. Very much of what passes for observation is merely mistaken inference. An amusing illustration occurs in Charles Darwin's recollections of his father:

"He himself never drank a drop of any alcoholic fluid. This remark reminds me of a case showing how a witness under the most favorable circumstances may be utterly mistaken. A gentleman-farmer was strongly urged by my father not to drink, and was encouraged by being told that he himself never touched

spirituous liquor. Whereupon the gentleman said, 'Come, come, Doctor, this won't do though it is very kind of you to say so for my sake — for I know that you take a very large glass of hot gin and water every evening after your dinner.' So my father asked him how he knew this. The man answered, 'My cook was your kitchen-maid for two or three years, and she saw the butler every day prepare and take to you the gin and water.' The explanation was that my father had the odd habit of drinking hot water in a very tall and large glass after his dinner; and the butler used first to put some cold water in the glass, which the girl mistook for gin, and then filled it up with boiling water from the kitchen boiler.” 1

"That which is strictly matter of perception does not admit of being called in question; it is the ultimate basis of all our reasoning, and, if we are to repose any confidence whatever in the exercise of our faculties, must be taken for granted. But there are few of our perceptions, even of those which to the unphilosophical observer appear to be the simplest, which are not inextricably blended with inference. Thus, as is well known to every student of psychology, in what are familiarly called the perceptions of distance and of form, the only perception proper is that of the various tints of color acting on the retina of the eye, and it is by a combination of this with perceptions of touch, and the muscular sense, that the mind gains its power of determining form and distance. Now, a judgment of this kind, which is really due to inference, is, especially by the uneducated and unreflecting, perpetually mistaken for that which is due to direct observation; and thus what is really only an inference from facts is often emphatically asserted to be itself a matter of fact.” 2

To quote from Mr. Mill:

"One of the most celebrated examples of a universal error produced by mistaking an inference for the direct evidence of the senses, was the resistance made, on the ground of common sense, 2 Inductive Logic, p. 273.

1 Life and Letters, p. 15.

to the Copernican system. People fancied that they saw the sun rise and set, the stars revolve in circles round the pole. We know that they saw no such thing; what they really saw was a set of appearances, equally reconcilable with the theory they held and with a totally different one. It seems strange that such an instance as this of the testimony of the senses pleaded with the most entire conviction in favor of something which was a mere inference of the judgment, and, as it turned out, a false inference, should not have opened the eyes of the bigots of common sense, and inspired them with a more modest distrust of the competency of mere ignorance to judge the conclusions of cultivated thought.

"In proportion to any person's deficiency of knowledge and mental cultivation is, generally, his inability to discriminate between his inferences and the perceptions on which they were grounded. Many a marvelous tale, many a scandalous anecdote, owes its origin to this incapacity. The narrator relates, not what he saw or heard, but the impression which he derived from what he saw or heard, and of which perhaps the greater part consisted of inference, though the whole is related not as inference but as matter of fact. The difficulty of inducing witnesses to restrain within any moderate limits the intermixture of their inferences with the narrative of their perceptions, is well known to experienced cross-examiners; and still more is this the case when ignorant persons attempt to describe any natural phenomenon. 'The simplest narrative,' says Dugald Stewart, 'of the most illiterate observer involves more or less of hypothesis; nay, in general, it will be found that, in proportion to his ignorance, the greater is the number of conjectural principles involved in his statements. A village apothecary (and, if possible, in a still greater degree, an experienced nurse) is seldom able to describe the plainest case, without employing a phraseology of which every word is a theory: whereas a simple and genuine specification of the phenomena which mark a particular disease, a specification unsophisticated by fancy, or by preconceived opinions, may be regarded as unequivocal evidence of a mind trained by long and successful study to the most difficult of all arts, that of the faithful interpretation of nature.” ” 1

1 Logic, p. 545

CHAPTER IV.

PRIMARY INDUCTIONS.

AN Induction is a generalization, or an inference, based upon propositions that state observed facts. The truth inferred may be general or particular, but it must be one which we cannot perceive in a single act of observation. When we know the existence of anything by simply attending to it, we do not say that we know it inductively we know it directly. The word Induction is applied both to the proposition enunciated and to the process of mind by which that proposition is reached. That "all men are mortal," I know by induction, and the truth is itself an induction.

Inductions are based either wholly upon observations, in which case we call them Pure Inductions; or they are based partly upon observation and partly upon intuitively known truth, in which case we call them Mixed Inductions. Pure inductions are either Complete or Incomplete, according as we have or have not observed all the facts included in the statement. They are either Primary or Secondary, according as they are made directly by generalizing a number of observations, or indirectly by combining syllogistically a single new observation with a previous induction. These distinctions will become clear as we advance. The present chapter deals with Primary Inductions.

It soon becomes plain to every child, when he begins to observe the world, that there is an existing order of

things. It is perfectly easy to conceive of a world in which every object should be unique and every event a surprising novelty. Such a world would contradict no necessity of thought, although it would be hopelessly bewildering. But such is not our world. The child's earliest impression is of a certain permanence and uniformity in its environment. The same objects and experiences remain or recur.

This conviction of an existing order finds expression in language. The present tense in grammar does not denote a mere moment separating the past and the future; it denotes a considerable and indefinite expanse of time. Such a proverb as "The burnt child shuns the fire" is stated in the present tense, as formulating a fact of the existing order.

That experience falls largely into lines of uniformity is early perceived. The child learns that there are things called apples which are round and red and good to eat, and that there are things called cats which have soft fur and long tails and sharp claws, and that these things are liable to scratch. The profoundest question in the whole science of inductive logic is: How are these generalizations reached? How can we ever discover that we are upon the line of a uniformity? But this is really only a sort of metaphysical puzzle, like the question of the possibility of motion. The existence of lines of uniformity is every moment forced upon our observation, and the fact that they do extend is equally conspicuous.

A Primary Induction is the statement of an observed uniformity. Do we reach it by any process of inference? Philosophers have thought so. There is thought