Critical Realism

B. H. Bode

WHEN The New Realism was published, nine years ago, some observers professed much surprise at the spectacle of philosophers laboring side by side in a common cause, without any discernible tendency on the part of any one of them to turn upon and rend his neighbor. Since then, however, the achievement has been duplicated in the volume entitled Creative Intelligence; so that the philosophical public is in process of becoming habituated :to the phenomenon. Whether these joint undertakings are evidence, as some seem to suppose, that philosophy is at last to enter upon an era of truly objective and rigidly impersonal inquiry, after the manner of the sciences, or merely that philosophers possess a hitherto unsuspected capacity for cooperation, is still a question upon which it is useless to look for agreement. The latest volume of this kind is the recent Essays In Critical Realism,[1] the purpose of which is to expound and defend the realistic faith which the contributors to the volume hold as a common possession.

As compared with the earlier books, this work offers a comparatively simple programme or plan of campaign, in that it is centered almost exclusively upon the nature of knowing. Five of the seven essays are devoted to this topic. As is stated in the preface, the authors have "found it entirely possible to isolate the problem of

( 69) knowledge," so that it is with reference to this problem, as contrasted with ontological problems, that the collaborators find themselves in essential agreement. In the two remaining essays the emphasis falls on the side of criticism rather than construction. The essay by Lovejoy is, in the main, a criticism of Dewey's pragmatism. and the essay by Rogers is a critical review of various theories regarding the nature of truth and error. While these essays contain much that is of interest and importance, they will be omitted from present consideration, since it is my purpose to discuss more specifically the doctrine to which the authors have applied the name of Critical Realism.

As was to be expected, the authors are at considerable pains to differentiate their position from other forms of realism. This is done by emphasizing the distinguishing feature of their doctrine of knowledge. Naive realism, so it is pointed out, made the mistake of supposing that physical objects could be intuited directly, and so found itself unable to deal with certain difficulties, particularly those arising from the relativity of sense-perception. Copy-ism escapes from this difficulty, to be sure, but it is obliged to get over to outer existence by a process of inference, which can be done only by a tour de force. Neo-realism is an attempt to rehabilitate the faith of naive realism in the identity of experience and object, but it is obliged to construct its world out of conceptual entities or essences, which gives rise to various difficulties, especially with reference to the problem of truth and error. Each of these standpoints contains something of value, which it is possible to conserve through a reinterpretation of knowledge along the lines laid down by Critical Realism. Naive realism and neo-realism are correct in insisting that physical objects are known directly and not through a process of inference. Copyism is correct in holding that experience and object are numerically distinct and not identical. The reconciliation and justification of these claims are the fruits of the new conception of knowledge which constitutes the distinctive trait of Critical Realism.

Stated briefly, the doctrine advanced by Critical Realism is about as follows: Knowledge takes places by means of a datum or "given."This datum, which is denoted variously as "quality-complex," "character-complex," and "essence," is not an existence, but something more in the nature of a meaning or what Bradley calls a "floating adjective." "By the essence of a percept I mean its what divorced from its that—its entire concrete nature, including its sensible characters, but not its existence" (p. 223). This doctrine of "essences" is the central feature of Critical Realism. It is by means of this doctrine that the position undertakes

( 70) to avoid the errors of its predecessors. Since the essence is not an existence, it can not be identified with outer reality, after the manner of neo-realism, though it can be affirmed of outer reality. Moreover, this affirmation is direct, which means that the reality of which the essence is affirmed becomes the object of knowledge, as against the assertion of copyism that the immediate object of knowledge is a mental state. To put it differently the essence is a means but not an object of knowledge (cf. pp. 97, 189, 226). It makes natural and easy the transition to an outer reality, which is so difficult for copyism; and it maintains the distinction between content and object of knowledge which is denied by naive realism and neo-realism.

If I interpret the doctrine correctly, this is the solution proposed by Critical Realism. It is a solution accepted by all the members of the group, but is elaborated particularly by Strong, who comments feelingly on the great significance of essences : "As I have elsewhere explained, I owe this precious conception to Mr. Santayana. I had long been convinced that cognition requires three categories for its adequate interpretation ; the intermediate one —between subject and objects—corresponding to the Kantian `phenomenon' or `appearance.' At one time I used to designate this category as content, since it agrees with the current conception of a content of consciousness; but in my efforts to conceive it clearly, I was continually falling off either into the category of object or into that of `psychic state.' What was my relief when at last I heard Mr. Santayana explain his conception of `essence,' and it dawned upon me that here was the absolutely correct description of the looked-for category" (p. 224, note).

At first sight the theory presents an appearance of engaging simplicity. It starts with the tri-partite division of mental existences, external existences and essences; and asserts that the essences are the meanings or contents through which the external existences become known. As long as we are careful to insist upon the status of the essences as floating or wandering adjectives, they are incapable of usurping the place of existences and offering themselves as the objects of thought. These floating adjectives seek an anchorage, which is provided by the act of affirming the external existences to which they pertain or the nature of which they reveal. This act of affirmation is much more fundamental and direct than any process of inference. "The sense of the outer existence of these essences is indistinguishably fused with their appearance" (p. 20). "We do not infer a realm of existence co-real with ourselves but, instead, affirm it through the very pressure and suggestion of our experience" (p. 195). That is, essences lead on "irre-

( 71) -sistibly" and "instinctively," to the world of existents; and so we escape both the Scylla of copyism and the Charybdis of hypostatized meanings.

It soon appears, however, that this doctrine of essences needs to be handled with care. On this point, it is intimated, there is, unfortunately, no complete agreement among the Critical Realists themselves. This issue is indeed "the one question in our inquiry upon which we have not been able fully to agree." The statement of the disagreement is relegated to a pair of footnotes (pp. 4 and 20), apparently with the commendable purpose of keeping family squabbles as much as possible out of the public eye. The disagreement, it seems, turns on the question of what constitutes a datum or essence. Three of the seven hold that the datum or essence is in every case the character of the mental existent; while the remaining four take the position that the essence may be, so to speak, composite in nature. According to the latter view, the essence may result in part from the nature of the mental state and in part from the function of the mental state or the use to which it is put. The statement of the difference is brief to the point of obscurity; but as I interpret it, the point is something like this: If I see a cushion as blue, the blue is an essence or datum. The mental state of the moment may include the quality `blue,' which is referred to the physical object. The dissenting three hold that it must be so included, since this is the only way in which essences can be obtained. But in the opinion of the second group, the datum, while it may be, need not be, a character of the mental state. The latter need not include the quality blue at all, "as, e.g., if I see the cushion in a faint light, when it is nearly black, or through tinted glasses, and yet perceive it as a blue cushion. So it is clear that the characters that make up the datum depend more upon the associations than upon the actual characters of the mental state" (p. 30). The actual datum may be constituted in part by the function performed by the mental state. On page 29, at the end of an illustration intended to show that the characters of the mental state may be very different from the datum or essence, it is said that "when a complex mental state of the sort just indicated exists, together with the readiness of the organism to act in a certain way, then we say, and feel, that a certain datum has been `given' or has `appeared.' This is all there is to `givenness'."

Perhaps we can rest content with the earnest assurance that the disagreement is not a serious matter. But, even so, a consideration of the disagreement furnishes an opportunity to gain a further insight into the meaning of Critical Realism. If we take the view

( 72) that the essence expresses the character of the mental existent, we come somewhat closer to the position of copyism. If my mental state must "consist" in part of the quality "blue" (whatever that may mean), it follows that sensible qualities are "subjective substitutes for the corresponding parts of the physical world" (p. 191), and that "the content in terms of which we think the object must have the property of reproducing the character of the object in some measure" (p. 198, italics mine). This is certainly the language of copyism. On the other hand, if we take cases in which, according to the second view, the essence is the joint product of mental states and their function, there is no justification for such language at all. The essence is not a subjective substitute and there is no process of reproduction. In appearance at least we are now much closer to the standpoint of common sense. Since in such cases "the datum as a whole (the total character given) is not the character of any existent" (p. 21, note), attention is naturally directed away from the conventional notion of reproduction and towards a consideration of function.

Whether the disagreement just mentioned has any serious consequences for their position is a question which we can afford to let the Critical Realists settle among themselves. My purpose is simply to point out the shift of emphasis towards function which the disagreement brings to light ; a shift that a pragmatically minded reader is not likely to overlook. He will not fail to notice that if "the datum as a whole is not the character of any existent," but is determined, in some measure, by the behavior of the organism, the import of this doctrine can easily be translated into his own familiar language of stimulus and response. The "datum as a whole," he finds, varies with the response; and there is little occasion to bother with the metaphysical "external object" of Critical Realism at all. "Data are directly dependent on the individual organism, not on the external object, varying in their character with the constitution of the sense organs and the way in which these are affected, and only secondarily and indirectly with the external thing" (p. 225). Moreover, these data are symbols or signs which make it possible to "rehearse and anticipate the movement of things" (cf. pp. 170-173). In other words, the data of Critical Realism can easily be induced to take the place accorded to objects in pragmatic philosophy. Datum and body vary concomitantly, and the process of experience becomes a process in which we "adjust our bodies and our beliefs" to our environment (p. 30), which seems to mean that experience is a constant quest for a more adequate stimulus. We test the adequacy of our data by observing how they work. If they stand up under the

( 73) test, they become symbols of other experiences, which is to say that they are not the objects but the means of knowing. It is all a question of further experiences. If a sense-observation requires confirmation, we appeal to the other senses, or to the observations of other persons, or to the congruity of the given observation with the whole body of our past experiences (cf. p. 106). To the adherent of pragmatic doctrine such extensive agreement is naturally a source of considerable gratification.

To the Critical Realist, however, this agreement is of minor significance, since his chief concern is for the "external object." To him the datum is not merely a symbol of other experiences, but is a warrant for the belief in an outer existence. Just how the datum functions in this connection is not altogether clear. It is stated that we pass to outer existence, as it were, "instinctively," since "the sense of the outer existence of these essences is indistinguishably fused with their appearance." "Thinghood and perception go together" (p. 197). Passages like these suggest that the reference to outer existence is somehow part and parcel of the datum. But we are also told that "when the datum is said to exist, something is added to it which it does not and can not contain—the finding of it, the assault, the strain, the emphasis, the prolongation of our life before and after it towards the not-given. These concomitant contributions of the psyche weight that datum, light it up, and make it seem at once substantial and incidental. Its imputed existence is a dignity borrowed from the momentum of the living mind, which spies out and takes alarm at that datum (or rather at the natural process that calls it forth), supposing that there is something substantial there, and something dangerous that will count and work in the world. But essences (as Berkeley said of his `ideas') are inert" (pp. 179, 180).

Contrasting statements of this sort suggest the uncomfortable suspicion that the harmony among the Critical Realists is attributable to company manners, rather than to inward disposition of mind. Unless the language is misleading, we have here another cleavage, besides the one already discussed. On the one hand we are assured that Critical Realism "looks upon the total content as empirical, and is sceptical of the Kantian theory of the constitutive understanding" (p. 211). On the other hand we are met with the assertion that existence is a "concomitant contribution" with which the psyche weights the datum. Whether these statements admit of reconciliation, we need not pause to inquire. Whether apparent or real, this disagreement likewise may be used to clarify issues. Just what are we to understand by the assertion that the affirmation of thinghood or existence must be superadded to the content of perception?

( 74)

Apparently the question raises a dilemma. If an additional element is superimposed from without upon the content of the datum by the affirmation, we get Kantianism; if nothing is super-imposed, we get an empty form. Perhaps these two alternatives have not been kept consistently apart. On the surface the statement that there is a "sense" of outer existence, which is "indistinguishably fused" with the content of the datum, appears to be intended as an alternative to Kantianism. But if so, it is necessary to ascertain just what is gained by the maneuver. The sensory qualities are already "present" by virtue of their status as experienced facts. But this "presence" is not what is meant by "existence." The "sense" aforementioned requires the affirmation of existence, but it furnishes no content or meaning for existence. It does not warrant the conclusion that "the special and invidious kind of reality opposed to appearance must mean an underlying reality, a substance; and it had better be called by that name" (p. 165), unless "substance" is taken to mean "existence" and nothing more. But bare existence adds nothing at all. A sensory fact which is merely present is not specifiably different from a fact which has the affirmation of existence added to it. As Hume says, "To reflect on anything simply and to reflect on it as existent, are nothing different from each other. That idea, when conjoined with the idea of any object, makes no addition to it."[2] The affirmation of this ontological existence is supposed to be vital to the position of Critical Realism, but an examination of it discloses, if a Yankeeism may be pardoned, that it is the little end of nothing, whittled down to a point.

But this is not the only connection in which the problem of existence arises to trouble us. Correlated with these "external objects" are "mental states." These too exist, although "their data, the appearances they yield me, are to be distinguished from the mental states themselves" (p. 21). The belief in these existences, however, seems to rest on a different basis from that of the belief in external objects. The appeal is not to a "sense of existence," as in the case of outer existence, but is rather to inference, backed up by introspection (cf. pp. 25, 26, 234.-237). The mental states must be held to exist, for they are needed as vehicles of the data or essences. Without the mental states we should be unable to account for the fact that data are sometimes given and sometimes not (pp. 26, 233). When we introspect, these states, ordinarily unnoticed, come to light. "I admit that an unfelt sensation, in the sense in which the word sensation is ordinarily used, is absurd; but

( 75) I persist in thinking that that which we feel, when we feel, i.e., distinctly attend to, a sensation, is capable of existing when it is not felt, and so does exist in all vision, hearing and touching of external realities" (p. 235).

The claim, then, that mental states are the vehicles of the data is intended to mean that the mental states give to the data that peculiar quality of "feltness" which distinguishes the given from the not-given. Data, to be sure, are not felt directly, since they are not existences. "It is well known that the chief factor in the visual perception of distance ... is convergence and accommodation of the eyes. The sense that distance is actually felt may then be due to the fact that it is brought before us by the muscular sensations of convergence and accommodation. Distance, in that case, would be felt, but not visually felt. And the instance would constitute a beautiful example of the way external objects and relations are known by means of sensations which have in them little of the characters of the external things, but are simply used as signs" (p. 236).

It will be recalled that copyism is criticized by Critical Realism for attempting to pass from the given experience to outer existence by a process of inference. The same criticism, it would seem, is applicable to the attempt to justify the belief in mental states by a process of inference. Since the existence of these states is not intuited, they are as much "outer" to the data as any physical fact. Does it become known to us in precisely the same way? For example, of an ordinary perception, in which a blue object is presented, it is said : "Blueness here belongs to both datum and mental state" (p. 30). The reference to the physical object, as we have been told, is brought about by a "sense of outer existence." Must we then resort to a parallel "sense of inner existence," which is likewise "indistinguishably fused" with the datum, or is it necessary to resort to inference? Two such "senses" mixed up in one experience would look dubious enough, in all conscience; while the other alternative is made unattractive by the horrible example of copyism. But waiving this point, we come upon a further question, What is a mental state when we finally discover it? Since the given consists exclusively of "essences," of meanings or universals, it would seem that introspection can not disclose a "sensation of blue," but merely "blue." That is, introspection comes upon the same datum that, in the original perception, was assigned to the physical object. What then can be meant by saying that it is now found to be the character of a mental state? The only difference that is introduced by introspection consists in the discovery of a different context for the blue. It is now found to be associated with

( 76) "sensation of eye strain" and similar introspective material. But this does not convert the blue into something mental, or make the blue a clue to the existence of a mental entity, unless this connotation is smuggled in with the word "sensations." While we speak, indeed, of "sensations of eye strain," they too are, by hypothesis, essences or data, and their reference to the eyes is as direct and unambiguous as is the reference of blue to a physical object. If we classify these "sensations" in turn as mental on the score of their associations, it is plain that we become involved in an endless regress of essences. Critical Realism provides no content for the notion of mental states; which is perhaps the reason why it is not scandalized by the suggestion of unconscious mental states. If we stick consistently to the doctrine that the given consists of essences, there can be no room for existences of any sort, and both external objects and mental states go by the board.

This conclusion is emphasized when we examine the function of mental states in giving concreteness or vividness to the essences which enter into experience. It is clear that, if externality is made to depend upon an empty reference of essence to existence, it becomes necessary to invest these essences with the "tang" of sensibility, by virtue of which they become transformed from plain abstractions into living experiences. They must take on "concretion for discourse and for action" (p. 22). This process is supposed to be illustrated by Strong's "beautiful example" of the muscular sensations of convergence and accommodation which give us the appearance of visual distance. "The datum is sensibly vivid because it is brought before us by a sensation" (p. 237. Unfortunately the illustration fails to illustrate. The datum being what it is, how can vividness apply to it? "A meaning here is not to be understood as a peculiar kind of feeling that can be met with introspectively in the same way that a visual sensation or a pain can, but as a function which the feeling dicharges in bringing us into mental relation to an external thing. When, having a sensation caused by an object in our minds, we are disposed (in virtue of the connected nervous arrangements) to act as with reference not to it but to the object, then that object is, in so far, before the mind as a datum" (p. 237).

The passage just quoted seems to reveal a significant inconsistency. Data are functions and so can not be met with introspectively, as it is possible to meet with a visual sensation or a pain. That is, a visual sensation or a pain is something different in kind from data or essences. If they are not essences, they must be existences, yet they can be the objects of our immediate apprehension. "There are states of our sensibility which do not bring before us

( 77) objects other than themselves—e.g., anger or pain, or in some cases, chill" (p. 233). How this squares with the doctrine that existence is never given directly, I am unable to make out. A little reflection will show, however, that the general position requires some concession in this matter of sensations. A rigid adherence to the doctrine of essences would leave no room for vividness at all. Vividness must come in, not as a meaning, but as something immediately "felt," something that "constitutes its own object." If we were to limit our consideration of sensations to essences referred to mental states, as the theory requires, the whole procedure would remain coldly logical. Since Critical Realism ignores the suggestion that givenness may be connected with the functioning of the "essence," and not of the "mental state," it can account for the warmth and intimacy of sensory experience only by lapsing into the standpoint of traditional subjectivism, and it finds itself obliged to give new life to the dismal theory of unconscious mental states, which seemed in process of dissolution. The whole situation seems to be just another phase of the historic difficulty about sensations and relations; and the best we get is the unintelligible assurance that "a datum can be so concrete as even to have sensible vividness, and yet not be an existence, but only an entirely concrete universal, a universal of the lowest order" (p. 231). How low a universal of this sort would have to be, it would perhaps be indelicate to inquire.

The foregoing criticism may be summed up by saying that the doctrine of essences, which constitutes the distinctive feature of the position and which is relied upon as an alternative to both copyism and neo-realism, works havoc in the end, because it leaves no room for existence of any kind. It is a pleasure to concede many merits to the book. In view of the nature of its topic, it is very readable. It possesses many keen and suggestive analyses, and it is undoubtedly an important contribution. But that it offers an acceptable solution as it stands, I am unable to believe. In the presence of the historic tradition which requires that mind be isolated from its objects by a gulf which can be traversed only by a claim, Critical Realism lays aside all its sophistication and shows a striking capacity for simple faith. But, as I have tried to show, the book itself furnishes certain suggestions as to the lines along which an acceptable revision might be made. And it provides additional evidence for the view that the "external object," to which Critical Realism attaches so much importance, serves no purpose whatever except to give a certain dignity or esthetic sanction to the proceedings. But the authors have succeeded in making their position as plausible as the materials at their disposal would per-

( 78) -mit, and in doing so they have done much towards the clarification of the important philosophic issues of the day.



  1. Essays in Critical Realism: A Cooperative Study of the .Problem of Knowledge. DURANT DRAKE, ARTHUR 0. LOVEJOY, JAMES BISSETT PRATT, ARTHUR K. ROGERS, GEORGE SANTAYANA, ROY WOOD SELLARS, C. A. STRONG. Macmillan & Co. 1920.
  2. Treatise of Human Nature, Part II, section VI.

Valid HTML 4.01 Strict Valid CSS2