The remark is perhaps more cryptic than enlightening. Its intent is that reflection is an actual occurrence as much so as a thunderstorm or a growing plant, and as an actual existence it is characterized by specific existential traits uniquely belonging to it: the entities of simple data as such. It is in control of the evidential function that irreducible and independent simples or elements exist. They certainly are found there; as we have seen they are "common-sense" objects broken up into expeditious and unambiguous signs of conclusions to be drawn, conclusions about other things with which they—the elements—are continuous in some respects, although discrete[4] with respect to their sensory conditions. But there is no more reason for supposing that they exist elsewhere in the same manner than there is for supposing that centaurs coexist along with domestic horses and cows because they coexist with the material of folk-tales or rites, or for supposing that pigs of iron pre-existed as pigs in the mine. There is no falsifying in analysis because the analysis is carried on within a situation which controls it. The fallacy and falsifying is on the part of the philosopher who ignores the contextual situation and who transfers the properties which things have as dependable evidential signs over to things in other modes of behavior.
It is no reply to this position to say that the "elements" or simples were there prior to inquiry and to analysis and abstraction. Of course their subject-matter was in some sense "there"; and, being there, was found, discovered, or detected—hit upon. I am not questioning this statement; rather, I have been asserting it. But I am asking for patience and industry to consider the matter somewhat further. I would ask the man who takes the terms of logical analysis (physical resolution for the sake of getting assured evidential indications of objects as yet unknown) to be things which coexist with the things of a non-inferential situation, to inquire in what way his independent given ultimates were there prior to analysis. I would point out that in any case they did not pre-exist as signs. (a) Consequently, whatever traits or properties they possess as signs must at least be referred exclusively to the reflective situation. And they must possess some distinguishing traits as signs; otherwise they would be indistinguishable from anything else which happens to be thought of, and could not be employed as evidence: could not be, in short, what they are. If the reader will seriously ask just what traits data do possess as signs, or evidence, I shall be quite content to leave the issue to the results of his own inquiries. (b) Any inquiry as to how the data antecedently exist will, I am confident, show that they do not exist in the same purity, the same external exclusiveness and internal homogeneity, which they present within the situation of inference, any more than the iron which pre-existed in the rocks in the mountains was just the same as the fluxed and extracted ore. Hence they did not exist in the same isolated simplicity. I have not the slightest interest in exaggerating the scope of this difference. The important matter is not its extent or range, but what such a change—however small—indicates: namely, that the material is entering into a new environment, and has been subjected to the changes which will make it useful and effective in that environment. It is trivial to suppose that the sole or even the primary difficulty which an analytic realism has to face is the occurrence of error and illusions, of "secondary" qualities, etc. The difficulty resides in the contrast of the world of a naïve, say Aristotelian, realism with that of a highly intellectualized and analytic disintegration of the everyday world of things. If realism is generous enough to have a place within its world (as a res having social and temporal qualities as well as spatial ones) for data in process of construction of new objects, the outlook is radically different from the case where, in the interests of a theory, a realism insists that analytic determinations are the sole real things.[5]
If it be not only conceded but asserted that the subject-matter generating the data of scientific procedure antedates the procedure, it may be asked: what is the point of insisting so much upon the fact that data exist only within the procedure? Is not the statement either a trivial tautology or else an attempt to inject, sub rosa, a certain idealistic dependence upon thought into even brute facts? The question is a fair one. And the clew to the reply may be found in the consideration that it was not historically an easy matter to reduce the iron of the rocks to the iron which could freely and effectively be used in the manufacture of articles. It involved hitting upon a highly complicated art, but an art, nevertheless, which anyone with the necessary capital and education can command today as a matter of course, giving no thought to the fact that one is using an art constructed originally with vast pains. Similarly it is by art, by a carefully determined technique, that the things of our primary experience are resolved into unquestioned and irreducible data, lacking in inner complexity and hence unambiguous. There is no call for the scientific man in the pursuit of his calling to take account of this fact, any more than the manufacturer need reckon with the arts which are required to deliver him his material. But a logician, a philosopher, is supposed to take a somewhat broader survey; and for his purposes the fact which the scientific inquirer can leave out of account, because it is no part of his business, may be the important fact. For the logician, it would seem, is concerned not with the significance of these or those data, but with the significance of there being such things as data, with their traits of irreducibleness, bruteness, simplicity, etc. Now, as the special scientific inquirer answers the question as to the significance of his special brute facts by discovering other facts with which they are connected, so it would seem that the logician can find out the significance of the existence of data (the fact which concerns him) only by finding out the other facts with which they coexist—their significance being their factual continuities. And the first step in the search for these other facts which supply significance is the recognition that they have been extracted for a purpose—for the purpose of guiding inference. It is this purposeful situation of inquiry which supplies the other facts which give the existence of brute data their significance. And unless there is such a discovery (or some better one), the logician will inevitably fail in conceiving the import of the existence of brute data. And this misconception is, I repeat, just the defect from which an analytic presentative realism suffers. To perceive that the brute data laid bare in scientific proceedings are always traits of an extensive situation, and of that situation as one which needs control and which is to undergo modification in some respects, is to be protected from any temptation to turn logical specification into metaphysical atomism. The need for the protection is sufficiently great to justify spending some energy in pointing out that the brute objective facts of scientific discovery are discovered facts, discovered by physical manipulations which detach them from their ordinary setting.
We have stated that, strictly speaking, data (as the immediate considerations from which controlled inference proceeds) are not objects but means, instrumentalities, of knowledge: things by which we know rather than things known. It is by the color stain that we know a cellular structure; it is by marks on a page that we know what some man believes; it is by the height of the barometer that we know the probability of rain; it is by the scratches on the rock that we know that ice was once there; it is by qualities detected in chemical and microscopic examination that we know that a thing is human blood and not paint. Just what the realist asserts about so-called mental states of sensations, images, and ideas, namely, that they are not the subject-matter of knowledge but its agencies, holds of the chairs and tables to which he appeals in support of his doctrine of an immediate cognitive presentation, apart from any problem and any reflection. And there is very solid ground for instituting the comparison: the sensations, images, etc., of the idealist are nothing but the chairs, tables, etc., of the realist in their ultimate irreducible qualities.[6] The problem in which the realist appeals to the immediate apprehension of the table is the epistemological problem, and he appeals to the table not as an object of knowledge (as he thinks he does), but as evidence, as a means of knowing his conclusion—his real object of knowledge. He has only to examine his own evidence to see that it is evidence, and hence a term in a reflective inquiry, while the nature of knowledge is the object of his knowledge.
Again, the question may be asked: Since instrumentalism admits that the table is really "there," why make such a fuss about whether it is there as a means or as an object of knowledge? Is not the distinction mere hair-splitting unless it is a way of smuggling in a quasi-idealistic dependence upon thought? The reply will, I hope, clinch the significance of the distinction, whether or no it makes it acceptable. Respect for knowledge and its object is