CONCEPT and QUALITY

A WORLD HYPOTHESIS

BY

STEPHEN C. PEPPER

(The Paul Carus Lectures, series 13, 1961, published by Open Court Press, Illinois, 1967)
 

To my wife Ellen

AUTHOR’S ACKNOWLEDGMENTS

I find that the account of intellectual indebtedness mounts with the years. Many names are mentioned in the pages ahead, such as Herbert Feigl with whom most of the pivotal issues have been discussed to their roots. But many men are not named there, whose influences are reflected from years back, such as C. I. Lewis, Donald Williams, Bernard Diggs, and the members of that wonderful, compact, argumentative Berkeley department of all my active years—Adams, Loewenberg, Prall, Dennes, Marhenke, Donald Mackay and Ed Strong. Nor can I omit mention of the vigorous young group who have succeeded these.

I am most grateful for the stimulus of the invitation to lecture on the Paul Carus Foundation, and for its generosity in publishing this work. Three lectures on the material were given during the meetings of the American Philosophical Association, Pacific Division, at the University of British Columbia in September 1961. They comprised parts of Chapters 1 and 2, Chapter 3 (later published in The Monist, Vol. 47, No. 2), and Chapter 6. I owe many thanks for this opportunity to the officers of the American Philosophical Association and to its Lecture Committee.

I wish also to express my special appreciation to Professor Eugene Freeman and his wife, Ann, for their editorial work on the manuscript and for all else that goes to provide for smooth and efficient procedure in the process of publication, and for superintending the many arrangements connected with this liberal lectureship.

                                                                                                    STEPHEN C. PEPPER
                                                                                                                              BERKELEY, CALIFORNIA
 

1. About World Hypotheses and Their Categories

[1] THE PRESENT STUDY SHOULD perhaps be regarded as a sort of sequel to, or outcome to be expected from, the results of an earlier study done some twenty-five years ago, which I called World Hypotheses.¹ There I sought to determine for myself the methods used and the achievements attained in the previous development of speculative philosophy. My conclusion was that the constructive work of the past was all of an empirical nature. The method consisted in originating and testing hypotheses and refining them for the clarity, consistency, and adequacy of their categories in relation to the evidence available to them.

[2] In all essentials that earlier analysis still seems to me sound. And the present study takes the earlier for its point of departure. The earlier analysis brought me up to date, I believed, in the fruitful achievements of the past. In the present one I am trying to offer a further contribution.

[3] In the interval of a quarter of a century between these two studies, there has been considerable novel philosophical activity. There has arisen on the European continent a vigorous school of existentialists, and among English speaking peoples, a movement initiated by the Oxford linguistic analysts. Both of these schools have been critical of traditional metaphysics, though for quite different, if not contrary, reasons. I doubt if my approach to metaphysics is exactly what could be called traditional. But I have great respect for the speculative empirical hypotheses I find embedded in the stream of western thought. And I do not find that either the existentialists or the Oxford analysts have offered suitable substitutes for the perspectives and the insights of these fruitful hypotheses.

[4] However, I do not wish to give the impression that these two movements have not had beneficial results. I only wish to make it clear that I am aware of what they have been doing. In so far as they have sought to undermine speculative philosophy, I do not find them successful, though I can be duly grateful for such constructive criticism and clarification as they have afforded. This is not an occasion to offer a detailed criticism of some of their methods of criticism which often appear narrowly confined in outlook and closed to modes of cognitive procedure other than their own. The general criticism I would level at both movements is their common assumption that each occupies and completely fills the total field of philosophy to the exclusion of each other and any other movements in philosophy. The present study rejects such an assumption without making any similar assumption regarding its own procedure. The field of philosophy is ample enough to contain many movements. There is plenty of room for constructive speculation, and possibly even for a new world hypothesis.

[5] What follows is proposed as a new world hypothesis—or possibly a rather radical revision of an older one—contextualism. Some of its categories, however, are inconsistent with principles usually regarded as distinctive of contextualism. I am not trying to be a contextualist. I am simply following through the interpretations of a world hypothesis based on categories derived from a new root metaphor.

[6] By a root metaphor, I mean an area of empirical observation which is the point of origin for a world hypothesis. When anyone has a problem before him and is at a loss how to handle it, he looks about in his available experience for some analogy that might suggest a solution. This suggestive analogy gives rise to an hypothesis which he can apply towards the solution. The method of development of world hypotheses for the problem of gaining comprehension of our world follows, I find, the same procedure. The originating analogy, I have called the root metaphor of a world hypothesis. An analysis of the root metaphor generates the categories of the hypothesis. The adequacy of the hypothesis then depends on the capacity of the categories to render interpretations of the features of our world with precision and unrestricted scope. A world hypothesis differs from other hypotheses only in its unrestricted scope. Other hypotheses are implicitly, if not explicitly, limited to a local problem in hand or, as in the special sciences, to a special field of subject matter. Such hypotheses may always reject certain considerations as being outside their field of inquiry. A world hypothesis never has this way out. It is responsible for the interpretation of any item of criticism proffered. It is an unrestricted hypothesis.

[7] From such an approach to speculative philosophy there is an important principle that comes to light. This is the uselessness of dogmatism and all its devices in any serious cognitive inquiry. Stated thus baldly the principle is a truism. But immediately one displays any of the treasured devices of dogmatism, the principle begins to cut, and to hurt many well meaning arguments. For one of the commonest devices of dogmatism is the appeal to certainty. Dogmatism may be defined as a demand for belief in excess of the evidence for it, or without allowing a critical examination of the evidence for it. To many people, it may sound strange to say this, but the appeal to certainty is a typical device of dogmatism.

[8] Most metaphysical theories in the hands of their exponents have made liberal use of this device. One of the advantages of studying a number of world theories based on different categories and root metaphors is to have this device clearly exposed. The device appears in many forms. It may appear in the form of infallible authority in the voice of God or his representatives, or in sacred books, or in the established doctrine of an Aristotle or Marx. More subtly it may appear in the form of self-evident principles for which the Euclidean axioms were long the models. More subtly still it appears in the guise of indubitable data, such as sensations, sense-data, immediate experience, or the mystic experience.

[9] It is almost impossible for one immersed in any single world theory not to believe that the basic items of cognition as interpreted by the categories of that theory, are in the nature of things cognitively certain. This is the more insidious because it is a fact that if that world theory were fully adequate, these items would be certain—that is, they would indeed be cognitively ultimate. (In the world theory we are about to develop there will be some cognitively ultimate items of this sort. But we shall be careful to make no cognitive claims for them beyond the evidences for the adequacy of the world theory as a whole.) One has to have seen from sympathetic study of two or more relatively strong hypotheses how it is that items apparently certain and ultimate for one interpretation are not so for another, in order to become properly aware that an appeal to certainty is a dogmatic and cognitively illegitimate device.

[10] Moreover, it is a device that does not in any way lend further grounds for credence to the item to which the claim of certainty is given. For the feeling of certainty is notoriously fallible. It is a cognitive criterion of certainty that is being claimed. And any insistent critic is bound to ask for the evidential basis for this criterion and its superlative claim. Either the claimant will then offer the evidence, whence the claim becomes superfluous, or he will protest that he has no further evidence and demand credence without it. And then the intended purpose of the device stands out clearly. Its purpose is to forbid the critic to ask for the evidence. It is to forbid the critic from criticizing. It is to stop all questioning. And that is dogmatism.²

[11] In developing a world hypothesis, the cognitive appeal must accordingly be always to the evidence available, and none of this evidence should be offered as certain. There will in this way develop a give and take between the evidence and the categories of the hypothesis ordering the evidence. The more the evidence corroborates the hypothesis the more it also corroborates itself in the interpretation it receives through the hypothesis. The root metaphor method is the regular empirical method of hypothesis supported by the evidence to the degree that the evidence corroborates the hypothesis and renders it relatively adequate. A world hypothesis, as we said earlier, differs from other empirical hypothesis only in its characteristic of being unrestricted in its subject matter. The root metaphor I have spotted as a possibly new one is that of the purposive act as this has been intensively analyzed through recent psychological research as a selective system. However, I am not sure that this is an entirely new root metaphor. A careful study of Whitehead’s Process and Reality and his other later works leads me to believe that what he calls an 'actual occasion' is a form of this very root metaphor. The actual occasion is Whitehead's basic concrete actuality. On one interpretation-somewhat forced I must admit-this can be taken as his pivotal structure, and all the other concepts in his system taken as derivatives or as abstractions from it. So taken, his actual occasion would be the root metaphor of his system. It does act as a selective system and has a purposive structure. It is intrinsically qualitative feeling and yet can be conceptually analyzed in retrospect. In many ways it corresponds to the root metaphor I am suggesting-though more rigidly described-and develops many features of his world hypothesis similar to those demanded by the categories of a purposive act viewed as a selective system and treated as a root metaphor.

2. A Comparison of Whitehead's Theory of Categories with That of the Root Metaphor Theory

[12] Nevertheless, I do think Whitehead's theory must objectively be taken as an eclectic one. But it is perhaps an eclecticism with the germ of a quite new type of world hypothesis generating in it. By eclecticism I mean a world hypothesis which picks out features of other world hypotheses regarded by the author as significant -- a composite of the best of previous philosophies, as it is sometimes described. The weakness of an eclecticism is that the features picked out do not work together. There are gaps or congestions at the joints. This is easily understandable when it is realized that the most adequate world hypotheses so far developed are all quite surely inadequate. Their categories do not entirely fit with the full scope and precision demanded of a world hypothesis. They are only relatively adequate. Inevitably features lifted rather arbitrarily from a number of world theories generated from different root metaphors, will not go smoothly together. The interpretations given do not corroborate one another. There will be overlapping and contrary interpretations. Not that the  relatively adequate world hypotheses are entirely free from conflict of interpretations. But there is more freedom from conflicting interpretations in a world hypothesis that is guided by a single fruitful root metaphor than in an eclectic theory that lacks any such guidance. So, I shall deliberately follow the root metaphor method in the development of the world hypothesis ahead.

[13] Occasional references to Whitehead will be useful, however, in so far as he does seem to have insight into the pivotal importance of a purposive structure in working out a comprehensive philosophy. Apart from his eclecticism his mode of approach is very similar to that recommended here. A comparison of his definition of the aim of ‘speculative philosophy’ with mine for that of a ‘world hypothesis’ may accordingly be useful at this moment.

[14] "Speculative philosophy," he writes, "is the endeavor to frame a coherent, logical, necessary system of general ideas in terms of which every element of our experience can be interpreted."³ In this definition, Whitehead expresses the primary characteristic of a world hypothesis noted earlier that it have unrestricted scope.) The "system of general ideas" framed by a speculative philosophy must be such that "in terms of [it] every element of our experience can be interpreted." He goes on to say "By this notion of ‘interpretation’ I mean that every thing of which we are conscious, as enjoyed, perceived, willed, or thought, shall have the character of a particular instance of the general scheme" (PR, p. 4). Whitehead’s intention in this passage is clearly to refer to ‘everything’ literally that can be brought up as an item for interpretation.

[15] In saying "everything of which we are conscious," I am sure he does not mean to restrict the scope of his system to only those items of which we are conscious. As soon as we enter upon the exposition of his system, we discover that conscious experience covers a very small range of the actuality of things he undertakes to interpret. His point here, I think (and it is a point worth noting), is that interpretation and in fact everything going on in the construction of a world hypothesis has to be registered finally in the field of conscious activity. We do, of course, become conscious of evidences for items of which we are not conscious. Most items covered by world hypotheses are of this sort. But ultimately we are responsible for the interpretation of these within the field of conscious awareness. And the criticism of the adequacy of a world hypothesis takes place in this narrow field. This is the condition of all reflective thinking. But the restriction of reflective thinking to the field of conscious activity does not in any way entail a similar restriction upon the scope of a world hypothesis. There are plenty of evidences within this field of consciousness that activities go on outside this field. The scope of a world hypothesis extends to everything without restriction and in this context specifically without restriction to the field of consciousness.

[16] There is, however, a technical advantage in defining scope in terms of whatever may come up for interpretation within our experience. If the unrestricted reference of a world hypothesis is taken as one to the universe as a whole, it may be objected that ‘the universe as a whole’ is a meaningless object of reference.

[17] Let me interpolate at this juncture that ‘meaninglessness’ is itself a meaningless term unless one has a precise theory about the nature of meaning. Since meaning is one of the most controversial items subject to interpretation by the various schools of philosophy and their world theories, it is not a term that can legitimately be employed against any world theory other than one’s own nor against the methods by which world theories may  be developed in general. The aspersion of ‘meaninglessness’ is one of the most ancient devices of dogmatism, though enjoying at present a sort of ebullient renaissance. It is a hidden appeal to certainty—akin to an appeal to self-evidence or indubitability. For it presupposes a theory of meaning which is presented as closed to criticism. An aspersion of meaninglessness should accordingly be taken as a cue immediately to investigate the dogmas of the critic making the aspersion.

[18] So, the criticism of a reference to the world as a whole as meaningless’ is not of vital importance. But it is perhaps wiser to define the scope of a world hypothesis without suggesting such a reference. For this reference might be taken as a presumption that there is a sort of determinate totality like the Absolute of the Hegelians being implied. This would indeed be restrictive, and far from what would be intended for the scope of a world hypothesis. It is wiser, therefore, to define unrestricted scope so as to allow for everything that may come up for interpretation, rather than as a reference to a totality however indeterminate.

[19] There are some other terms in Whitehead’s definition that require comment. He says in expanding on his definition, "the philosophical scheme should be coherent, logical, and, in respect to its interpretation, applicable and adequate. Here ‘applicable’ means that some items of experience are thus interpretable, and ‘adequate’ means that there are no items incapable of such interpretation" (PR, p. 4).

[20] These qualifications seem to me unnecessarily restrictive for a set of categories. They could easily conflict with the distinctive characterization of a world hypothesis as unrestricted. For me the characterization of a world hypothesis as adequate directly signifies its applicability. A world hypothesis does not hover in a vacuum. Its purpose is its applicability to this world of ours. By definition it is an hypothesis that applies with unlimited scope. Its adequacy consists in its ability to describe or otherwise to handle in its own terms any item that comes up in any degree of detail. An adequate world hypothesis would thus have unrestricted scope and unlimited precision. Actually precision goes with scope, for lack of precision in handling a detail would automatically be one way in which there would be lack of scope. A relatively adequate world hypothesis (which is the most a man could humanly expect) would then be one that accepted responsibility for unrestricted scope and precision and fulfilled these criteria more nearly than any other world hypotheses available even though not completely. And there could be several such hypotheses on a par.

[21] So, there seems to be no need of adding applicability to adequacy as a requirement for a world hypothesis. The more so if there is any intimation in Whitehead’s distinction to the effect that ‘applicability’ means that there be some items to which this system of categories applies and no items to which this whole system does not apply. For if it is required of a system of categories that every category apply to every item presented, this again places a serious restriction on the nature of a world hypothesis, and indeed upon the sort of world it is thought to apply to.

[22] There is no reason that I can see why some of the guiding concepts of a world hypothesis should not be such as would apply to only some items of experience. The world itself for all we know may be a dualism or a pluralism, where an adequate set of categories could not all of them apply to everything. Whitehead’s system does, in fact, turn Out to be a sort of monism.⁴ Inadvertently he may be restricting world hypotheses to monisms by his own restricted demands for adequacy.

[23] His amplifications of ‘coherent’, ‘logical’, and ‘necessary’ bear this suspicion out. By "coherence" he says he "means that the fundamental ideas… presuppose each other so that in isolation they are meaningless" (PR, p. 5). He adds that "this requirement does not mean that they are definable in terms of each other; it  means that what is indefinable in one such notion cannot be abstracted from its relevance to the other notions" (PR, p. 5). It sounds as if he meant by this requirement to eliminate dualisms and pluralisms. Later he alludes to Descartes’ philosophy as an instance of incoherence in its "arbitrary disconnection of first principles" (PR, p. 9). Granted the inadequacy of Descartes’ philosophy, but was this due purely to its dualism? At the same time, when Whitehead adds further, "In other words, is it is presupposed that no entity can be conceived in complete abstraction from the system of the universe" (PR, p. 5), we definitely do have to agree. But conceiving entities "in complete abstraction from the universe" is surely not the necessary result of admitting among the categories of a world hypothesis some which can "be abstracted from.., relevance to the other notions" (PR, p. 5). Coherence in this sense is not for me a requirement for a set of categories. If the actual world is not coherent, presumably a coherent set of categories would not fit it. Of course, Whitehead’s philosophy did make out that the world was essentially coherent. But his theory is hardly entirely adequate, and he may well be mistaken about the world’s coherence.

[24] His requirement of ‘necessity’ for the categories of a world theory would also be a mistake, for it amounts to practically the same thing as coherence. "The philosophic scheme," he says, "should be ‘necessary’ in the sense of bearing in itself its own warrant of universality throughout all experience, provided that we confine ourselves to that which communicates with immediate matter of fact" (PR, p. 5). This seems to imply that all items of the universe must be coherent with immediate matter of fact. Otherwise he fears that "what does not so communicate is unknowable, and the unknowable is unknown" (PR, pp. 5-6). Again such ‘necessity’ appears to be an unnecessary restriction. Why should a set of categories be required to "bear in itself its own warrant of universality throughout all experience"? If new experience leads to the discovery of new categories why should not they be added to the system? How can one wishing to maintain this adequacy of the categories, do other than add them?

[25] And now as to the requirement that the system of categories be ‘logical’. This sounds innocent particularly if we restrict ‘logical’ to "consistency or lack of contradiction." In point of fact, any discoveries of pairs of categories that are contradictory or lead to contradictory interpretations of the same items are sure to be taken as evidences of inadequacy in the theory. But instead of requiring logical consistency of a set of categories, it would seem to me wiser to let this be discovered as an important item in the world of our experience. Non-contradiction cannot be set up as self-evident without dogmatism, like any other appeal to certainty. There is actually a philosophical school of long tradition and widely respected which does not accept this principle of non-contradiction among its categories of the real world—namely, mysticism. This way of philosophizing cannot without dogmatism be thrown out of consideration by a requirement for its categories which it does not accept. I do think an examination of any mystical world view will find it extremely inadequate. Most mysticisms make a dogmatic appeal to certainty and all of them seem to be seriously lacking in scope. I would say that this is the way to criticize them. Show them up to be philosophically inadequate admitting the possibility that the requirements of logic are illusory.

[26] Whitehead, however, means more than just non-contradiction by his requirement that a system of categories be ‘logical’. He includes "the definition of constructs in logical terms, the exemplification of general logical notions in specific instances, and the principles of inference" (PR, p. 5). With Whitehead’s own mathematical background and bias, no one could object to his making this limitation upon his own set of categories. But it is clearly restrictive to require that any set of categories for a world hypothesis incorporate the principles of formal or mathematical logic uncriticized into its system. It would be somewhat like requiring the modern theory of thermodynamics to be incorporated uncriticized into a philosophical system. These are items within human experience to be interpreted, not to be assumed uncritically, by a world hypothesis.

[27] Lastly, Whitehead’s definition of a speculative philosophy as a "system of general ideas" indicating that the categories should be general, would be restrictive for some systems. For instance, a system including a monotheistic God might need to treat God as a category. A monotheistic God would hardly be a general idea. Also a single substance theory like that of the Milesian philosophers would be excluded, and so would mysticism. To be sure, most philosophical categories are general, but not all. The conception of an unrestricted hypothesis cannot, therefore, make this restriction.

[28] All that we can say without encumbrance about the categories of an unrestricted hypothesis is that they be adequate in their scope and precision to cover in their own terms any items that may be presented to them for interpretation. I use the indeterminate term ‘cover’ in this connection so as not to introduce a restriction upon the manner in which the adequacy may be achieved. The manner of this coverage may be by descriptive correspondence, by operational reference, by coherent inclusion, or even by emotional absorption. To exclude any of these would be a way of placing a restriction upon an unrestricted hypothesis which might be just the reason it would fail to achieve adequacy.

[29] Some might be apprehensive that such complete freedom of application could lay this sort of philosophic enterprise open to fantastic flights of imagination. The history of philosophy has not borne this apprehension out. Romantic flights of philosophic imagination have not prospered, and have lasted as long as some of them have only through appeals to certainty or through the force of some imposed authority. An unrestricted hypothesis cleared of all dogmatic devices is controlled by the evidences for the facts of the world themselves demanding corroboration wherever conflicts of expectation and belief arise. Questioned evidence calls for more evidence and for the refinement of the concepts referring to it. This process goes on with wider and wider demands for corroboration, and, where there is pertinacious criticism, it goes on till it reaches the limit of unrestricted scope. The stubborn facts of the world control world hypotheses toward the development of those concepts only which offer some fair promise of corroboration. Irresponsible flights of cognitive imagination come soon to earth when it is found that only corroborative evidence will support them.

[30] Herein lies the importance of a fruitful root metaphor. It is, to begin with, a small body of factual material as it stands. It guarantees at least an empirical start. Its fruitfulness consists solely in its capacity to generate a set of categories which with careful refinement may prove relatively adequate for an unrestricted hypothesis.
 

NOTES:

1. (Berkeley: University of California Press, 1942),  hereafter referred to as WH.]

2. There has been a lot of careful recent analysis on various meanings of the word ‘certainty’ as used in common language and in much philosophical writing. Among these I find it frequently useful to speak of ‘practical certainty’ for many beliefs and anticipated actions. But this is quite different from the cognitive criterion of certainty alluded to above. The latter is all I am at pains to expose and reject on account of its illegitimate cognitive appeal. Mast other common uses of ‘certainty’ paradoxically allow for some hypothetical uncertainty. As for the ‘certainty’, attributed to logical definitions and the like, this is simply an act of stipulation, and need not conflict with our rejection of certainty as a cognitive criterion, unless logical certainty is illicitly taken to establish self-evidence besides. The principle of contradiction is not for me, for instance, self-evident, but a principle highly evidenced to the point of practical certainty. (Cf. Chapter 14, Section 4.)]

3. Alfred North Whitehead, Process and Reality (New York: Macmillan Co., 1929), p. 4, hereafter referred to as PR.

4. By this I do not mean to attribute an absolutism to Whitehead, but only to call attention to his categorial denial of unconnected or distinct sorts of being. His multiplicity of individual occasions could be considered a sort of pluralism. Cf. footnote in PR, p. 508.

contents
 
 

1. The Purposive Act

[31] THE MOST ORGANIZED TYPE of simple purpose is that of the goal seeking act. Its structure has been analyzed in detail only recently. Consequently, there has never till now been an opportunity to make full use of it as a point of departure for a world hypothesis.

[32] Many teleological world theories have, of course, been proposed from the Aristotelian to the Hegelian. Strangely enough a theory based on purposive structure is not necessarily teleological. It does not require that the totality of nature should be regarded as heading for some cosmic goal. All it requires is that the structural character of purposive activity find full embodiment in this theory and that all interpretations be made in terms derivable from this structural character. There may be processes that fall far short of a fully developed goal seeking purpose. All that is required for the adequacy of the theory is that such processes be capable of fitting in with such a purpose. They might even turn out to be elements in its structure. It would not be required that the elements of a structure repeat the form of the structure of which they are elements.

[33] There is an assumption that occasionally comes to the surface in the course of a philosophic discussion that if elements of a larger whole are discoverable these should be regarded as basic or categorial in the construction of an hypothesis. Such an assumption is open to question. There is no more reason on the face of it for regarding wholes as derivative from elements than in regarding elements as derivative from wholes. For elements in this broad sense include not only spatial parts but aspects, phases, stages, characters, and anything else that might go into a whole. Many such elements could appear in wholes or out of them, and it might well be more convenient or appropriate to refer to them primarily in their relation to certain wholes in which they function and secondarily as capable of independent functioning.

[34] Taking purposive structure as a whole for a root metaphor does not commit one to a teleological world theory beyond the fact that this theory will provide a firm place for purposive activity and not seek to reduce it to something else or explain it away. Activities may be described by the categories developed from the root metaphor which are not purposive, and there would be no reason to assume that the choice of such a root metaphor would commit one to a theory that the world process was one great purposive act. If any such results come out, these will not be because the categories derived from the root metaphor demanded them, but because the empirical evidence in developing the theory led to them.¹

[35] There are three initial reasons why the selection of the goal seeking purposive act for the root metaphor of a world hypothesis may be thought promising. One was mentioned earlier. This act is the most highly organized type of simple purpose—possibly the most highly organized activity in the world of which we have any considerable evidence. It is the act associated with intelligence. And so it entails the features of the organism which performs the act. If we concentrate attention on this act, we are not likely to miss important features in cosmic structure and process. For other activities and structures are likely to be simplifications of this. We can learn about them by a sort of subtraction and will not be mystified by the demand for the addition of features emerging beyond our original basic categories.

[36] The second reason for promise in our selection is that a purposive activity is one that may go on in the full illumination of consciousness. We can feel its whole qualitative course from initial impulse to terminal satisfaction. We can have the immediate feel of the perceptual demands of an environment in all its qualitative variety and graded intensity upon the search for the means of satisfaction. And we can feel the shock of a blocked anticipation when a wrong choice is made. We also feel here the emotions and have awareness of the values that seep into our perceptions in any living concrete act of pursuing a purpose within an environment that must be felt out, selected from, and cognized in the purposive pursuit. By starting with a root metaphor that possesses full qualitative immediacy, this should go far towards re-solving a problem that has tantalized mechanistic naturalism from Descartes down to the present—that of the relation of the mental to the physical—which cannot be dismissed as entirely a pseudo problem.

[37] And thirdly, this qualitative structure has submitted to a detailed conceptual analysis in behavioristic terms. Here, then, is an ideal opportunity to see how a set of effective and well elaborated concepts come to apply to a qualitative structure lived through in a man’s immediate experience.

[38] If we are somewhat successful in making this contact between the behavioristic conceptual description of a purposive act and the immediate qualitative experience of this act felt as it is going on, we may then be able to extend this sort of relationship much further. We may be in a position to interpret other conceptual descriptions that mesh in with the behavioristic concepts of purposive activity towards an understanding of a concrete qualitative side of nature underlying our overt purposive acts. The main idea here is that where there are well confirmed concepts for physical activities, there are qualities to which they are adapted somewhere.

[39] For the three reasons mentioned: (1) that the goal seeking purpose is so highly organized an activity as perhaps to enfold most of the simpler ones there are, and thus to present an exceptionally useful sample of the processes of nature; (2) that it is an activity open to awareness of its full qualitative immediacy throughout its whole course; and (3) that a detailed conceptual description of it in behavioristic terms is available—for these three reasons there is perhaps some justification for being hopeful that rewarding results may come from the selection of this item as a root metaphor.

2. A Description of Purposive Structure

[40] Let me now spread out a preliminary analysis of the appetitive or goal seeking purpose. We shall then lift out of this description the categories for our world hypothesis.

[41] It will be convenient to have a name for the forthcoming hypothesis. I propose selectivism. Its appropriateness will appear later as we proceed with the details.

[42] I shall use for my description of a typical purposive act an adaptation from E. C. Tolman’s analysis in his Purposive Behavior in Animals and Men.² For the complete expansion of this analysis, I refer to my treatment in The Sources of Value.³ actually this whole analysis should be taken as my root metaphor. But, in a way, the appetitive structure is the culmination of purposive activity, and for the intention of lifting out categories for a world hypothesis it is all that is needed to begin with.

[43] Should anyone wish to criticize this way of describing purposive behavior on the grounds that it is the way employed by only one school of psychology, my initial reply is that this is one of the strongest contemporary schools and but for minor issues includes a wide diversity of prominent writers from Hull to Tolman. But my broader reply is that of course this analysis like any empirical analysis is subject to correction and refinement. It seems to me the most comprehensive and promising to date. I will take my chances that further development in the subject will not eliminate some persistent dynamic agency in a purposive act, containing a reference into the future and a ‘split dynamics’. These are the essential traits for the adequacy of the categories about to be developed from this root metaphor.

[44] Suppose we have in mind some fairly simple example of a goal seeking or appetitive purpose. I wake up early in the morning and find I want a drink of water, I feel for a glass beside my bed and find there does not seem to be any. I turn on the light and see sure enough there isn’t. So I get out of bed and walk into the bathroom, switch on the bathroom light, open the door to the cabinet over the hand basin, take out a glass, turn on the cold water, fill the glass and take a long drink, then set the glass down and return to bed for another snooze.

[45] This short illustration has all the essential features of an appetitive purpose. Other appetitions are but elaborations of ones like this, sometimes producing systems of intricate complexity and great range of reference.

[46] In the foregoing illustration I am asking you to imagine it as if it were your own act. Such a purposive act is common enough for anyone to have performed it, or one very much like it. Imagine it, or bring it up in your memory in its full qualitative immediacy—the way it felt when it was going on. Also watch for its next recurrence. And think to yourself as you find yourself enacting it in its sequence of qualitative impulses and sensations, "This is what an appetitive act actually is."

[47] This is the basic root metaphor I have in mind. Now I turn to the description of it. The concepts will be rather technical, for they are refined psychological concepts. And in the treatment of the root metaphor these concepts will be also included. For it is not possible to keep the concepts of description entirely separate from the qualitative data of immediacy described by them. However, as far as possible the descriptive concepts should be kept separate in one’s thought from the qualitative data they describe.

[48] Returning to the illustration, we observe first that I woke up to realize that I had a want. The concept for a want like this is ‘drive’. This was a specific well known drive, thirst. It emerges frequently in the course of a day, and is predictable. For this reason it is classified as a cyclic drive, and also becomes conceptually labeled as a ‘disposition’ of the ‘organism’ in which it emerges.

[49] This drive evoked a reference immediately to water. This term ‘reference’ is a descriptive concept. Specifically it was an ‘anticipatory reference’, and we shall frequently call it an ‘anticipatory set’. My body assumed a set of muscular tensions, or at least a neural pattern of anticipatory references connected with the liquidity and other qualities of water. Anticipatory references or sets of this sort are called cognitive references because they have been learned and consequently may be in error. They are references which may be true or false. I expected water would satisfy my thirst. The expectation has been so often verified that I do not even think about it any more as something I had to learn. In fact, I learned that water satisfied thirst so early that I cannot remember ever having learned it. But I can be shocked occasionally into realizing I did learn it, when I lean over a pool and take a drink of salt water, and I see I can be in error about this sort of reference.

[50] Such facts lead to the realization that there are other kinds of references connected with drives like thirst which are not cognitive references. These may be called conditions of satisfaction for a drive. These are an inherent character of a drive and cannot be in error. Hence they are not cognitive references. They may be called inherent drive references.

[51] Both cognitive and inherent drive references designate purposive goals. But the inherent drive reference is the ultimate terminal goal for an appetitive purpose. This is the goal of satisfaction for the drive that brings the purpose to an end. In an appetitive purpose, the terminal goal is an act of satisfaction specific to the drive motivating the act. For a thirst drive it is the act of drinking which quenches the thirst. It may be called the ‘quiescence pattern’, or ‘consummatory act’.

[52] The cognitive references designate goals expected to bring about the terminal goal of the quiescence pattern. These subordinate goals may be called ‘goal objects’. They are the goals of anticipatory sets. If there is a wide gap between the start of the drive and the attainment of the quiescence pattern satisfying the drive, there will be a succession of goal objects with their corresponding anticipatory sets leading up to the terminal goal of satisfaction. These are generally called instrumental acts and the goal objects the means for the attainment of the ultimate goal.

[53] One more important feature. The drive motivating the instrumental acts is identical with that motivating the demand for terminal satisfaction in the quiescence pattern. the drive exhibits a split dynamics. Its charge in its references is split off into the anticipatory sets acting for their goal objects as means for reaching the final quiescence pattern. At the same time it continues to want the attainment of its quiescence pattern. The drive simultaneously motivates both the demands for the goal objects and also that for the quiescence pattern.

[54] When we want the means for an end, we want the means because we want the end. Our wanting of the end goes on all the time that we are wanting the means, and it is the continuous wanting of the end that keeps us wanting the means. So the wanting gets split between the end and the means for attaining the end.

[55] It is in consequence of this split dynamics that some instrumental acts prove to be in error and others correct. This is the basis for intelligent selection which is the crucial characteristic of purposive behavior. Add to this mode of selection the capacity for learning, by which an incorrect act is discarded and a correct act integrated into a structure that can be relied upon for future situations of a similar sort, and we find here the most flexible and efficient selective process in nature.

[56] The appetitive purpose may be schematized so as to bring out its dynamic structure as follows:

                       D = drive, Aⁿ = anticipatory set, O^g = goal object, Q^p = quiescence pattern.

[57] The left hand arrows indicate the flow of the charge of the drive, in anticipation that each goal object referred to could actually be reached and would practice the goal object ahead till the terminal quiescence pattern were attained.  The right hand arrows indicate the instrumental sequence of the correct acts.

[58] Let me explain my getting up in the night for a drink of water in terms of this diagram. I awoke with a thirst drive, D, demanding satisfaction in a specific act of drinking, which is the quiescence pattern Q^p. This demand augmented by the effects of past learning activated the cognitive reference or anticipatory set Aⁿ1 for its goal object, water, Q^g1. But water is not directly at hand. I thought it might be and felt for it, and then put on the light for a surer try. These were anticipatory sets tried out and rejected because they did not serve towards the satisfaction of the drive. The drive being unsatisfied persisted, it will be noticed, and initiated another line of action which did lead to the satisfaction of my thirst, this latter set of actions led to a successful achievement in the quiescence of my drive, so that I could go back to sleep again. This was a succession of instrumental acts with anticipatory sets and goal objects leading up to satisfaction on the anticipation Aⁿ2 of water available at the bathroom basin, a glass in the cabinet, Aⁿ3, accessible by opening the cabinet door, Aⁿ4, reached by walking into the bathroom, Aⁿ5 which I can start doing right now from my bed, so off I go following the references of my anticipations, goal object after goal object till quiescence of the drive ensues in the pleasant act of drinking the cold water. The drive charged the succession of anticipatory sets, which in turn motivated the acts in sequence for the attainment of the successive goals.

[58a] So here we have a detailed conceptual description, and an immediate qualitative act to which the description applies. The conceptual description could of course be made much more detailed, but let us keep it relatively simple as long as we can. The principle of this interpretation is what we need to get at just now. We can make the analysis as rigorous as the data will permit, once we see what this sort of descriptive analysis is up to.

3. The Two Descriptions—The Qualitative and the Conceptual

[59] Perhaps at this point, someone is asking just what the description is describing. Is it describing my inner feelings or someone’s observations of my outer behavior? It is basically describing my qualitative activity between the awakening in the night and the quenching of my thirst. Fortunately in this instance (and I chose it partly for that reason) I have access to the qualitative activity. I performed it and I can remember it and verbally describe it. And I am thinking that my readers have had similar experiences and can follow my description in their own qualitative terms. I suppose we have all read novels about qualitative acts much more intricate than this one and were not utterly mystified by the symbols.

[60] At the same time The foregoing conceptual analysis also fits, when augmented with certain congruous interpolations, as a so-called objective behavioristic account. This would be the sort of account an external observer would give of my actions watching them from the outside. The thirst drive would be accounted for in terms of the dehydration of my body as a result of a number of hours without taking in water. The conditions of quiescence for this drive would be known from many observations of the results of dehydration on my organism and other organisms resembling mine. The identification of this specific drive in my behavior would be verified on this occasion by observing the termination of this particular series of acts upon my final act of drinking the water. The distinction between this final act as the quiescence pattern and terminal goal of the sequence of acts, and of water as the terminal goal object instrumental in producing the final act of quiescence, would be observed. Likewise the subordinate goal objects sought as means to the attainment of the final goal object would be recorded. The error in reaching for something on the table by the bed would be observed and described with the assistance of past observations of the behavior of intelligent organisms in case of error. What would not be perceived by an outside observer of my behavior would be the anticipatory sets. But these would be inferred as dispositions of the organism as a result of learning. These dispositions as hypothetical concepts, which explain the operation of learning, have been abundantly confirmed by observations of many organisms and by controlled laboratory experiments. And these concepts are fitted in with physiological concepts of neural activity and the functioning of the brain.

[61] The foregoing analysis of a goal seeking purpose, an appetition, thus fits either an introspective qualitative description of this activity or a behavioristic objective description. The de-tailed content, of course, is different. It is richly qualitative in the introspective report. It is mostly what is called quantitative and relational in the objective report.

[62] The advantages and disadvantages of the two reports show up clearly in their respective treatment of the anticipatory sets. The introspective account reports directly about the qualitative feelings of anticipation which are observed in the course of action. The behavioristic description has to treat these as hypothetical explanatory concepts, or "intervening variables" as they have sometimes been called. On the other hand, these hypothetical concepts can be directly related to a great system of concepts under the headings of physiology and anatomy, supported by a mass of objective observations and interlocking hypotheses widely confirmed and possessing great predictive power. And, of course, the physiological system interlocks with the conceptual systems of chemistry and physics. In short, the behavioristic report connects an appetitive purpose with the whole system of contemporary empirical science.

[63] The introspective report, in comparison, is extremely limited in range. It is connected with a considerable amount of personal qualitative memories which are more or less available. Also indirectly (and at this stage of our inquiry rather mysteriously) it is connected with a lot of other qualitative reports communicated by other persons. In particular, it is connected with nearly the whole domain of literature and all the various arts which exploit qualitative elaboration and enhancement. Even with these indirect contributions from the arts, however, the range of material open to introspective qualitative description is extremely limited in comparison with that of the sciences which exploit so called objective methods. Their concepts reach in depth to elements below the chemical atoms, and in extent to galaxies beyond the visible stars.

[64] The relation, whatever it may be, between these two kinds of reports — the qualitative introspective and the quantitative or relational objective—will be one of our principal concerns in the study ahead. In some way they must both find a full justification.

[65] It appeals to me, therefore, that in the choice of the goal seeking purpose, or appetition, as a root metaphor we have a segment of fact in respect to which the two kinds of report are wedded to each other. The schematic analysis of the appetition can, as we saw, be taken either way—either in qualitative or in behavioristic terms. In developing the categories for a world hypothesis Out of this root metaphor, we shall take cognizance of both these ways of considering this analysis.

[66] We shall draw up two lists, a qualitative and a conceptual list. Though closely parallel, they are not exactly the same concepts. For the qualitative list does not automatically gear in with the conceptual system of the natural sciences, while the conceptual list does. It will be maintained that both of these lists are fully descriptive, in their own ways, of a purposive act. This fact will be a crucial one in the development of the world hypothesis to come. We shall take each list equally seriously as a veridical description so far as observation has gone. If there have been errors, they are open to correction. A refinement of the concepts will always be possible.

[67] Of course, the categories in both lists, the qualitative and the conceptual, are themselves concepts. But the qualitative categories are so named because they refer directly to qualities and qualitative features immediately felt. The conceptual list refers to these only indirectly by way of an external observer reporting on the behavior of an organism exhibiting purposive actions. These behavioristic concepts are framed and controlled, accordingly, by the conceptual system of the natural sciences. They are the relevant concepts for the conceptual system of the sciences in their bearing on purposive action.

[68] Let me repeat once more the unique significance of this dual description for the root metaphor of a world hypothesis. At this point in the description of a purposive action, almost alone in the whole expanse of nature, we have both a highly articulated qualitative description and a highly articulated conceptual description which refer to exactly the same actual process. The bifurcation of nature into conceptual system and qualitative experience meet here at this point. Here is where the crotch of the fork is from which the bifurcation extends.

[69] My thesis will be that there is nothing wrong in this bifurcation. It was inevitable if our knowledge of the world was to increase. Once we understand it, and can trace it from its point of bifurcation at the handle to the tips of the prongs at the other end, we shall find it to be, not a source of division in our knowledge, but the very instrument for its comprehensive unification.

4.The Categories

[70] Here are the two lists of categories referred to in the previous section:

                                            Qualitative Categories

A. Categories for a single qualitative strand (e.g., drive quality).
          1. Felt quality with dynamic urge for action.
          2. Duration of the quality yielding a continuous qualitative strand.
          3. Intensity of quality felt as dynamics of activity.
          4. Reference to goal felt in the dynamic quality.
          5. Blockage from environmental strands.
          6. Splitting of dynamic reference to charge instrumental strands with their felt references to
               instrumental goals towards attainment of goal of drive.
          7. Selection of instrumental strands towards attainment of final goal.
          8. Positive feeling of satisfaction in terminal act or quiescence pattern (and, in blockage, negative
               feeling of dissatisfaction).
B. Categories of context of qualitative strand.
         1. Simultaneity of diverse strands.
         2. Articulation of successive strands in an integrated total act.
         3. Anticipations and apprehensions as felt dispositions for action.
         4. Fusion—the merging of the qualities of diverse strands into a new distinct quality instituting a
                qualitative strand in its own right.
         5. Specious present—or field of immediacy.
C. Categories of qualitative range.
        1. (a) Actual present—consisting of whole range of specious presents in action.
            (b) Past—real but not actual, as events referred to as once actual but now outside the actual present.
            (c) Future—real so far as it is an inherent potentiality of the actual present, but not actual.
         2. Controlling environment of strands—for any qualitative strand in action—the actuality and
                reality of the situation.

                                           Conceptual Categories

A. Categories for a single complete act of purposive behavior in "objective" terms.
        1. Bodily action and tension pattern arising from internal bodily changes, or environmental
               stimulation (the drive impulse)
        2. Continuity through a period of time.
        3. Energy of measurable quantity observable as kinetic energy in overt action or conceived as
               potential energy in states of bodily tension.
        4.Vector character of bodily energy indicating along with bodily changes conditions for dissipation
               of the energy or its maintenance in a steady state.
        5. Interaction with environmental activities.
        6. Vector changes due to interaction with environment, and channeling of energy for shortest path
               through response mechanisms for final discharge of energy, or for maintenance of a steady state.
         7. Selection of response mechanisms for discharge of energy or maintenance of steady state.
         8. Quiescence pattern of responses in reduction of energy for the drive impulse.
B. Categories of physical structure.
         1. Body of organism.
         2. Articulation of behavior of organism in an integrated act.
         3. Dynamic dispositions such as physiological sets available for action upon proper stimulation.
         4.
         5.
C. Categories of physical environment.
         1. Space-time.
         2. Configurations of matter in space-time.

[71] It will be seen that these two sets of categories are closely parallel. The only marked discrepancy is the absence of categories in the conceptual scheme for B4 (fusion) and B5 specious present) in the qualitative scheme. But possibly evidences for fitting these in will come up in the course of the development of the total theory.

[72] Also the distinction between actuality and reality which appears in category C1 for qualitative range does not appear in the conceptual category C1 of space-time. This is a matter that will require considerable expansion later.

[73] The apparent absence of reference to space in the qualitative category C1 is not a matter of concern. For the qualitative space corresponding to the spatial dimensions of space-time is taken care of in terms of environmental references in A4 and As and elsewhere among the qualitative categories, and in that sense are implicit in any actual specious present constituting the qualitative category, C1 (a) .

[74] Of course, the running distinction between the two sets is that the qualitative set refers directly to felt qualities throughout, whereas the conceptual categories do so only indirectly, and might be superficially interpreted as not referring to them at all. As an initial reply to a criticism that I am confused in thinking the conceptual categories refer to felt qualities at all, let me emphasize the fact of the very close parallelism of the two sets and ask what would be its significance unless both sets were referring to the identical sequence of acts, which as clearly as evidence can testify are qualitatively felt whatever else they may be.

[75] This leads me to explain the importance of stressing felt quality. The qualitative experiences referred to in what is often called an introspective report are what I shall mean by felt qualities. I think this term is a little better than Whitehead’s term "feelings," though I believe the two terms refer to essentially the same entities.

[76] To speak of the felt quality of an emotion such as anger, remorse, compassion, or love seems natural enough. These differ in quality, and the quality is a felt quality. An unfelt quality of anger would not be anger at all. It would not be anything. If anyone started taking this idea seriously, a philosopher would have to find out what got this person in the way of conceiving an unfelt anger. What could be this person’s presuppositions for such a concept? Having discovered them, what evidence would there be in support of them? We surmise that the evidence would be slim or lacking. There would, however, be plenty of evidence for the error and the manner in which it came about. The concept would thus be explained (explained away, if you wish) as an error. It would be an hypothesis entertained but lacking in confirmation—in short, false.

[77] We find it natural to speak of pain and pleasure as felt qualities also. Similarly, no doubt, of the felt qualities of touch—pres.. sure, prick, warmth and coolness. But when we come to speak of the felt qualities of color or sound, there is likely to be a hesitation or a downright balking. One whole side of common sense thought is impelled to distinguish between a color quality—red, or yellow, or green—from the feeling of it. Red, it is believed, is the quality of the surface of certain objects. This may be felt or not. That is, people may become aware of the quality, red, or not. The quality is separable from the awareness of it.

[78] Now, it is this separation of a quality from a feeling or awareness of it that I wish to deny in the expression ‘felt quality’. I wish to emphasize by this expression that an unfelt red is as erroneous a concept as an unfelt anger. Whitehead was making the same denial in calling all such entities ‘feelings’ simply. I must admit that my expression runs the hazard of an objection that it implicitly concedes what I am explicitly denying. For does not the expression with its two words implicitly concede that there are two entities, ‘feeling’ and ‘quality’, which are here referred to as combined? Whitehead’s simple term ‘feeling’ carries no such hazard.

[79] All I can say at this point is that I hold, in conformity with the set of categories spread out above, that the distinction between a quality felt and a quality unfelt is an erroneous one. A red unfelt is as mistaken a notion as an anger unfelt. Anyone can develop a concept of an unfelt red, as indeed also of an unfelt anger. But the concepts will be found unconfirmable. There will be no ultimately acceptable evidence for the empirical references of these concepts, and plenty of evidence for the errors of thought that brought them about.

[80] Then why do I use the expression ‘felt quality’? Because I want to use the term ‘quality’ for what I am referring to. It comes closer in its range of common sense connotations to what I wish to refer to than Whitehead’s ‘feeling’. But I must take pains to shut off one side of the common sense usage of ‘quality’ which permits it to be separated from the common meaning of ‘feeling’ which Whitehead is utilizing. Having in mind the impulsions of common usage, I think the combined expression ‘felt quality’ comes as near denoting the elements referred to in the set of qualitative categories above as can be got without coining a new technical word, which has disadvantages of an opposite kind. A new technical term gives the reader nothing to hitch on to.

[81] As a matter of practice, I shall use the term ‘quality’ alone most of the time for what I am here explaining as ‘felt quality’. This use of ‘quality’, incidentally, has solid precedents in the philosophical writings of contextualists and organicists and in other schools also. Hume’s ‘impressions’, for instance, are ‘felt qualities’ in the above sense so far as they are fully qualitative and treated as actual occurrences and shown to be entirely free from dependence on any additional entity such as a ‘spirit’ or an ‘awareness’ to hold them ‘in mind’.

[82] Speaking of Hume’s impressions as precedents of a sort for what the qualitative categories refer to as qualities, I must make it clear that ‘felt qualities’ are not conceived as limited to substantive contents like sensations and emotions. There are felt qualities of relations and references and fusions and patterns and perhaps other things. There are ‘felt qualities’ for any sort of occurrence.

[83] I should also perhaps guard against an unintended interpretation of an opposite kind. It is not implied that all predicates of descriptive sentences refer to ‘felt qualities’. There is no restriction set by the above categories on the sort of references that may be made in sentences or other instruments of thought. Even true sentences may have predicates which do not refer to specific qualities. If a predicate is a determinable like the concept, color, it presumably does not refer to a specific felt quality.⁴ A sentence such as ‘This flower is brightly colored’ would nevertheless be true. For the predicate has a set of references that could be verified or confirmed in terms of felt qualities. In more usual terms, this is much the same as saying that though color is not a datum, it is a concept that can be verified in terms of data.

[84] The term ‘property’ will help in avoiding some purely linguistic ambiguities that could arise here. Let ‘property’ mean a descriptive reference to entities which may be either conceptual or qualitative. For a property to have a veridical reference it must ultimately reach its terminus in one or a group of felt qualities. But a sentence may be true without following its references all the way through. Also properties may be referred to that are entirely fictitious. And finally, felt qualities are not necessarily restricted to those available for human introspective reports. Felt qualities may be inferred as a blind man may correctly infer the existence of visual qualities for men of normal vision.

[85] With the foregoing distinctions in mind, it can be seen that the qualitative categories and the conceptual categories for the world hypothesis being here developed are distinct sorts of predicates. The qualitative categories refer to qualitative properties of the world, the conceptual categories to conceptual properties. A full description of the world in terms of either the qualitative or the conceptual properties would be true, so far as our world hypothesis may be adequate, but the truth of the conceptual description would ultimately depend on the truth of the qualitative one.⁵

contents

NOTES:

1. I have mentioned Whitehead as a possible precursor in the use of this goal seeking purpose as a root metaphor. And I may yet regret mentioning this idea. Here is a case in point. Whitehead does develop a kind of cosmic purpose generated through the activity of God analogous in God’s terms to the "concrescence" of an "actual occasion." This latter is the pivotal actuality in his system described in a way to suggest roughly a purposive goal seeking act. But Whitehead’s development of God in these terms does not follow from the purposive structure of an act. It is largely due to the organistic (Hegelian) portion of his eclecticism. The organistic feature even enters into his description of an actual occasion somewhat distorting the empirical data.

2. (Berkeley and Los Angeles: University of California Press, 1932).

3. (Berkeley and Los Angeles: University of California Press, 1958), Chapters I-IX.

4.  I do not wish, however, by this illustration to eliminate the possibility that often by the agency of fusion determinables may well appear as simple felt qualities. A felt quality may be ‘vague’. For instance, considering the many discriminable reds there are, the predicate red is properly to be regarded as a determinable. Yet a person may truly report a felt quality, red, having actually not tried to discriminate a more specific color quality. However, I think it is safe to say that the extremely indeterminate determinable, color, would never be introspectively reported as a simple felt quality. The action of fusion balks at producing a felt quality of indiscriminate color. It even balks at an indiscriminate blue-red. What comes out is a felt quality of a fairly discriminate purple.

5. This statement should not be taken as a commitment to a verifiability theory of meaning in which verification is limited to the termination of references in sensations or sense-data. Felt qualities for our world theory (as will become clear later) are not limited to sensations or to the experiences of members of the human species.