Having said this, let us compare the psychological law and that of nature. Are they identical? We shall be told that they are not, since, as a fact, errors are committed at every moment by the sudden failures of human reason. This is the first idea which arises. Human error, it would seem, is the best proof that the two laws in question are not alike, and we will readily add that a falling stone does not mistake its way, that the crystal, in the course of formation does not miss taking the crystalline shape, because they form part of physical nature, and are subject in consequence to its determinism. But this is faulty reasoning, and a moment of reflection demonstrates it in the clearest possible manner; for adaptation may miss its aim without the being who adapts himself and his surroundings necessarily obeying different laws. When the heat of a too early spring causes buds to burst forth prematurely which are afterwards destroyed by frost, there is produced a fault of adjustment which resembles an error of adaptation, and the bringing forward of this error does not necessarily imply that the tree and the whole of physical nature are obeying different laws. Moreover, the difference between the laws of nature and those of the understanding does not need deduction by reasoning from an abstract principle; it is better to say that it is directly observable, and this is how I find that it presents itself to us.
The essential law of nature is relatively easy to formulate, as it is comprised in the very definition of law. It simply consists in the sentence: uniformity under similar conditions. We might also say: a constant relation between two or several phenomena, which can also be expressed in a more abstract way by declaring that the law of nature rests on the combination of two notions, identity and constancy.
On the other hand, the laws of our psychical activity partly correspond to the same tendencies, and it would be easy to demonstrate that the microcosm of our thoughts is governed by laws which are also an expression of these two combined notions of constancy and identity. It is, above all, in the working of the intellectual machine, the best known and the most clearly analysed up till now, that we see the application of this mental law which resembles, as we say, on certain sides, the physical law: and the best we can do for our demonstration will doubtless be to dissect our reasoning powers. Reason, a process essential to thought in action, is developed in accordance with a law which resembles in the most curious manner a physical law. It resembles it enough to imitate it, to conform to it, and, so to speak, to mould itself on it.
Now, the reason does not follow the caprices of thought, it is subject to rules; it results from the properties of the images, those properties which we have above referred to, the material character of which we have recognised, and which are two in number—similarity and contiguity, as they are termed in the jargon of the schools. They are properties which have for their aim to bring things together, to unite, and to synthetise. They are unceasingly at work, and so apparent in their labour that they have long been known. We know, since the time of Aristotle, that two facts perceived at the same time reproduce themselves together in the memory—this is the law of contiguity; and that two facts perceived separately, but which are similar, are brought together in our mind—this is the law of similarity.
Now, similarity and contiguity form by combination the essential part of all kinds of reasoning, and this reasoning, thus understood, works in a fashion which much resembles (we shall see exactly in what degree) a physical law. I wish to show this in a few words. What renders my demonstration difficult and perhaps obscure is, that we shall be obliged to bring together rather unexpectedly categories of phenomena which are generally considered separate.
The distinctive attribute of the reason consists, as I have said, in the setting to work of these two elementary properties, similarity and contiguity. It consists, in fact, in extending continuity by similarity; in endowing with identical properties and similar accompaniments things which resemble each other; in other words, it consists in impliedly asserting that the moment two things are identical in one point they are so for all the rest. This will be fairly well understood by imagining what takes place when mental images having the above-mentioned properties meet. Suppose that B is associated with C, and that A resembles B. In consequence of their resemblance the passing from A to B is easy; and then B suggesting C by contiguity, it happens that this C is connected with A connected, though, in reality, they have never been tried together. I say they are associated on the basis of their relation to B, which is the rallying point. It is thus that, on seeing a piece of red-hot iron (A), I conclude it is hot (C), because I recollect distinctly or unconsciously another piece of red-hot iron (B), of which I once experienced the heat. It is this recollection B which logicians, in their analysis of logical, verbal, and formal argument, call the middle term. Our representation of the process of reasoning is not special to argument. It also expresses the process of invention, and every kind of progress from the known to the unknown. It is an activity which creates relations, which assembles and binds together, and the connections made between different representations are due to their partial identities, which act as solder to two pieces of metal.
It will now be understood that these relations between the images curiously resemble the external order of things, the order of our sensations, the order of nature, the physical law. This is because this physical law also has the same character and expresses itself similarly. We might say "all things which resemble each other have the same properties," or "all things alike on one point resemble each other on all other points." But immediately we do so, the difference between the physical and the mental law becomes apparent. The formula we have given is only true on condition that many restrictions and distinctions are made.
The process of nature is so to do that the same phenomenon always unfolds itself in the same order. But this process is not always comprehended in real life, for it is hidden from our eyes by the manifold combinations of chance; in the reality that we perceive there is a crowd of phenomena which resemble each other but are not really the same. There are a number of phenomena which co-exist or follow each other without this order of co-existence or succession being necessary or constant. In other words, there are resemblances which are the marks of something, as a logician would say, and others which are not the marks of anything; there are relations of time and space which are the expression of a law; there are some which are accidental, and may possibly never be reproduced.
It would be a wonderful advantage if every scientific specialist would make out a list of the non-significant properties that he recognises in matter. The chemist, for example, would show us that specific weight has hardly any value in diagnosis, that the crystalline form of a salt is often not its own, that its colour especially is almost negligible because an immense number of crystals are white or colourless, that precipitation by a given substance does not ordinarily suffice to characterise a body, and so on. The botanist, on his part, would show us that, in determining plants, absolute dimension is less important than proportion, colour less important than form, certain structures of organs less important than others. The pathologist would teach us that most pathological symptoms have but a trivial value; the cries, the enervation, the agitation of a patient, even the delirium which so affects the bystanders, are less characteristic of fever than the rate of his pulse, and the latter less than the temperature of the armpit or the dryness of the tongue, &c. At every moment the study of science reveals resemblances of facts and contiguities of facts which must be neglected for the sake of others. And if we pass from this profound knowledge of the objects to the empirical knowledge, to the external perception of bodies, it is in immense number that one espies around one traps laid by nature. The sound we hear resembles several others, all produced by different causes; many of our visual sensations likewise lend themselves to the most varied interpretations; by the side of the efficient cause of an event we find a thousand entangled contingencies which appear so important that to disentangle them we are as much perplexed as the savage, who, unable to discriminate between causes and coincidences, returns to drink at the well which has cured him, carefully keeping to the same hour, the same gestures, and the same finery.
The reason of this is that the faculty of similarity and the faculty of contiguity do not give the distinction, necessary as it is, between resemblances and co-existences which are significant and those which are not. The causal nexus between two phenomena is not perceived as something apart and sui generis; it is not even perceived at all. We perceive only their relation in time and space, and it is our mind which raises a succession to the height of a causal connection, by intercalating between cause and effect something of what we ourselves feel when we voluntarily order the execution of a movement. This is not the place to inquire what are the experimental conditions in which we subject phenomena to this anthropomorphic transformation; it will suffice for us to repeat here that, in perception, a chance relation between phenomena impresses us in the same way as when it is the expression of a law.