I: B. F. Skinner on Private Events

D. A. Begelman, Ph.D.

 

Introduction
 

            J. L. Austin once remarked that the current state of philosophy is like the surface of the sun: a pretty fair mess. United efforts by philosophers to resolve the other minds problem (Ayer, 1946; Wisdom, 1956: Thomson, 1951; Sellars, 1953; Putnam, 1957; Austin, 1961; Aune, 1961; Kirby, 1966), or its offshoots, like the private language problem (Wittgenstein, 1958), have likewise kept pace with the spirit of disorder. They sometimes appear to be on the verge of abandoning interest in the issue altogether, in the hopes of tackling problems distinguished by at least an occasional sense of achievement. It is surprising in the light of their frustrations that radical behaviorists like Skinner believe they can resolve the problem by ingenious—if questionable—adjustments in their wider account of how variables come to control behavior. They appear to believe that privacy can be packaged in ways uniquely suited to relieving the subject of any tendency to become as worrisome for them as it is for philosophers. Accordingly, a question arises over whether Skinner’s analysis of private events illuminates something new or insightful in speculation on the subject. In the opinion of the present author, the truth is quite the opposite: the radical behavioristic approach to private events enshrouds the topic in paradox and counterintuitive implications.

            Any discussion of Skinner’s contribution to our understanding of private events—not to mention his psychological approach in general—tends to run up against a disconcerting road block. While there may be ambiguity about what he explicitly believed about the topics he discussed, confusion is often attributed to those outside the fold. If one were to poll his sympathizers in psychology about what they felt was a distinguishing feature of his contribution to psychology, an answer might be its misunderstanding by critics. The conviction among his followers that his thought has been misconstrued regularly is everywhere in evidence. For example, in a commemorative issue on Skinner in the November, 1992 issue of American Psychologist, a majority of the contributors cite routine misinterpretations of the psychologist’s writings. Among the culprits referenced were nativists like Chomsky (Palmer and Donahoe, 1992), cognitivists like Mahoney (Rakos, 1992; Todd & Morris, 1992), literary critics like Krutch, Koestler, Jessup, and Hacker (Dinsmoor, 1992), and philosophers like Dennett (Baum & Heath, 1992) Taylor and Dretske (Rachlin, 1992), and Scriven and Malcolm. Todd and Morris (1992) in the same issue of the magazine draw attention to what they have called a tradition of “academic folklore” ostensibly manifesting a “steady misrepresentation” of Skinner’s views. As if to document what appears to be a contrarian tradition, two loyalists referred to the article by Skinner in an issue of Brain and Behavior Sciences in which he was finally driven to lament: “I am sorry if so many of my replies must consist of a series of corrections, but nothing else seems to serve” (Todd & Morris, 1992, p. 1441).

Few contributors in the same issue of American Psychologist considered the possibility that a problem arises from ambiguities in the published writings of Skinner himself. There are, however, notable exceptions to this. Lee (1992) argued that Skinner’s concept of behavior is problematic, and Glenn, Ellis, & Greenspoon (1992) averred that Skinner’s definition of an operant as a class of responses fails to distinguish it relative to other salient categories. Despite these exceptions, the response to criticisms of Skinner on the part of those sympathetic to his point of view has been stalwart. It has also over time exhibited signs of parochialism. Day has referenced what he calls a “messianic commitment” to radical behaviorism on the part of its proponents (Day, 1980, p. 169). The complacency does a disservice to the issues Skinner addressed. Those are momentous ones, and not the kind a thinker however celebrated should be expected to resolve—even during the course of a lifetime. It is imperative, therefore, that inquiry into them remain unencumbered by a response-set wedded to the conviction that disagreeing with Skinner necessarily involves misunderstanding him. In the spirit of the kind of debate he himself would have endorsed, the following is intended as a tentative contribution to the rubric of private events in a science of behavior.                                                                                                                          

Radical Behaviorism and Private Events

Skinner has been forthright in his insistence that a scientific account of private events is the “heart of behaviorism” (Skinner, 1974, p.212). The characterization was meant to emphasize the focus as not only a legitimate area of inquiry, but an enterprise distinguishing radical behaviorism from other behavioristic approaches. The latter presumably have either rejected the study of private events altogether, or posit intervening variables eschewed by Skinner as either redundant way-stations or mentalistic fictions. There is little reason to suppose, however, that the relatively exclusive focus on public events would shift appreciably under radical behaviorism or behavior analysis, even if the interest in private events within its ranks were to undergo an unparalleled surge of activity. What form this might take is difficult to envision. Radical behaviorists, despite their commitment to the analysis of events under the skin, have in the past dealt with the subject in largely ceremonial ways, or as place-holders in broader discussions about the proper purview of scientific scrutiny. Despite their commitment to the study of private events, ostensibly dictated by a philosophy of science, there has been little indication of notable contributions to the subject on their part. It is as though those who proclaim no fear of water are reluctant to get their feet wet.

Causal Efficacy

It might be conjectured Skinner’s analysis takes back with the left hand what it grants with the right. On one level, it seeks to compensate for the neglect of private events in “methodological behaviorism” (Skinner, 1945; Moore, 2001) by reestablishing the rightful place of the topic in a complete scientific formulation. Yet Skinner’s contretemps  is the relegation of private events to a limbo of causal inefficacy as the mere by-products of behavior (Martin, 1978; Schnaitter, 1978). The move seems to clash dramatically with the spirit of philosophical largesse enshrined in the Skinnerian conviction that events under the skin are no different metaphysically from public behaviors. If cut from the same ontological cloth, wherefore the causal disenfranchisement? In this connection, Skinner has declared:

The private event is at best no more than a link in a causal chain, and it is usually not even that. We may think before we act in the sense that we may behave covertly before we behave overtly, but our action is not an “expression” of the covert response or the consequence of it…to say that a man strikes another because he feels angry still leaves the feeling of anger unexplained. When we have once identified the relevant variables, we find the feeling of anger much less important by way of explanation (Skinner, 1953, p. 279). 

Despite the fact that the foregoing passage seems to reflect a philosophical allegiance to a form of epiphenomenalism (Creel, 1980)—usually identified as the   doctrine that “mental” events have no efficacy in causal chains—its  ambiguity is evident. We cannot be sure whether “the feeling of anger” is causally disenfranchised by Skinner because it is a sheer “mentalism” defying a scientific approach as a presumed event in a mysterious, hence elusive medium, or, as a covert physical response, it lacks causal status because the independent variables of which it is a function remain to be explained. The two slants are quite different ones. The former is sometimes suggested by Skinner’s allusions to the mentalistic trappings into which talk of private events is conventionally cast, as in his remark: “One is still free, of course, to assume that there are events of a nonphysical nature accessible to the experiencing organism and therefore wholly private” (Skinner, 1953, pp. 279-280). 

Infinite Regress

The second interpretation, that causality is lacking because independent variables of which the covert physical event is a function have not been explicated, is an argument that would appear to establish a set of double standards for covert and overt behaviors. If the covert event does not enjoy parity of causal status with external independent variables because its determining conditions have not been specified, the same can be said for any independent variable, public or private. Thus, a physical blow to the head cannot explain the covert sensation of pain in the victim because the independent variables that in turn control the blow have not been analyzed (i.e., supplied with a functional analysis). Skinner’s argument actually courts a problem of the infinite regress of causal relations, in which no independent variable or set of them counts as explanatory because it too is a function of some prior set of conditions. Thus, if Smith’s environment, not Smith, is responsible for his creative achievements, what is responsible for the environment that created the environment that created Smith’s achievement? Likewise, if Smith’s private sensations (even as covert physical events) are causally disenfranchised, so is the environment construed as puppet master of his private life!

Skinner’s colloquies on the illusion of “freedom” in other of his writings (Skinner, 1971) may also court the same difficulty. If Smith cannot be said to be “free” or to personally achieve anything because his actions are environmentally determined, then neither are environments causally responsible for anything, since they too cannot be explained without recourse to the preexisting independent variables of which they too are a function. Ironically, Skinner himself seems to have appreciated the point: 

How far back? When a person says that he acted “because he felt like acting, “ we can put little faith in the “because” until we have explained why he had the feeling, but it has been objected that we must stop somewhere in following a causal chain into the past and may as well stop at the psychic level. Clearly that is what is done most of the time in mentalistic discussions, and that is why they block further inquiry. It is true that we can trace human behavior not only to the physical conditions which shape and maintain it but also to the causes of those conditions and the causes of those causes, almost ad infinitum, but there is no point in going back beyond the point at which effective action can be taken. That point is not to be found in the psyche, and the explanatory force of mental life has steadily declined as the promise of the environment has come to be more clearly understood (Skinner, 1974, p.210). 

In the above passage, Skinner assumes that the rationale for stopping an infinite regress is the point at which “effective action” can be taken—although he does not explain why the idea of “explanatory force” is necessarily linked to effective action. The latter, on its face, would appear to be an advantage in any approach aimed at controlling, predicting, or modifying behavior, although it is not exactly clear why this capacity should be a criterion of explanatory power. Even so, it remains a mystery why Skinner should assume that mentalistic explanations lack the capacity to promote effective action. For example, a Cartesian dualist who witnesses a colleague having an accident occasions his rushing to the victim’s side in order to provide assistance. His action is focused on relieving his friend’s pain, although his philosophical persuasion has him believing the friend is in a state of mental anguish (as evidenced by his groans), and is, in fact, a human being with a purely and inseparably mental and physical side to his nature. The dualist may harbor a philosophically questionable account of the mind-body quandary, but how does such a philosophy prevent effective action in any sense endorsed by Skinner? Why would a radical behaviorist have an edge on effective action in the context under discussion? Is there some cryptic sense of “effective action” coming into play in behavior analysis noticeably lacking in mentalistic accounts of behavior?

In a similar vein, Skinner has declared throughout his writings that inner events construed as mentalistic cannot be affected (i.e., controlled) directly: “No one has ever directly modified any of the mental activities or traits...There is no way in which one can make contact with them” (Skinner, 1974, p. 208). Among traits that cannot presumably be modified directly Skinner lists, inter alia, “ideas” and “beliefs.” Anyone other than a radical behaviorist might find the claim that ideas and beliefs cannot be modified directly counterintuitive, if only for the reason that he or she believes they are so modified all the time. If they weren’t, it is thought, experience could teach us nothing, while education, even in the broad sense of the term, would be a waste of time.

However, there may be another reason for Skinner’s claim. It is possible that ideas and beliefs, on the mentalistic assumption they are occurrences in an immaterial medium, cannot because of this be directly modified, because we cannot, as it were, “get our hands” on such elusive stuff. It is as though the challenge were akin to trying to tangle with a ghost. The problem is: if an immaterialistic medium is a scientific fiction, what is the allegedly “scientific” basis for assuming beliefs and ideas couldn’t be modified if it weren’t?

Canonical Exposition

Unquestionably, the analysis of private events within the behavior-analytic tradition held a special meaning for Skinner. His most detailed discussion of the subject can be found in his essay On the Operational Analysis of Psychological Terms (1945). Subsequent, although shorter expositions may be found in chapter seventeen of Science and Human Behavior (1953), Verbal Behavior (1957), Contingencies of Reinforcement (1969), Beyond Freedom and Dignity (1971), and About Behaviorism (1974). Briefly, Skinner’s treatment of the problem confronts the issue of how we come to discriminate among events occurring under the skin, since we must learn such an achievement, and the resultant personal knowledge depends upon how we are “reinforced” by the verbal community for describing our private lives accurately. However, according to Skinner external reinforcement cannot be made contingent on the property of privacy. Reliance by others on observable or public events (presumably the correlates of our internal states) is the route by which they are able to reinforce our accurate self-descriptions. While Skinner outlines several distinguishable ways private events may come to control behavior, the thrust of his analysis is that the verbal community relies on the public accompaniments of private states to reinforce what he terms “tacts,” or descriptive statements about them (Skinner, 1957). For example, I am reinforced for saying I am in pain when the verbal community notices I wince or grab my jaw while experiencing a toothache.

In Verbal Behavior, Skinner distinguishes between tacts and mands. The former are verbal responses evoked by external events or certain descriptive properties of them, whereas the latter are operants which correspond to verbal responses reinforced by a characteristic consequence when the speaker is in a state of deprivation or under aversive control. There is nothing about the distinction that prohibits given verbal operants sharing characteristics of both types of response. For example, I am in pain is a tact under the control of a private event, although it is routinely emitted in contexts in which the speaker is manding reinforcement from an internally aversive state of affairs. Consequently, there would appear to be a place in the verbal repertoire for tands and macts, since many of us who mand reinforcement need not be completely tactless! 

Skinner’s writings about privacy have two principal emphases: (1) how radical behaviorism, unlike other philosophies of science, provides for the study of private events, and (2) how his treatment of privacy addresses itself to and resolves certain paradoxes in the acquisition of self-descriptive verbal repertoires. The second of these emphases is the more intriguing one. The first seems to have been repeated so often, it probably represents little more than a reminder to those who got the point decades ago.

Radical Behaviorism and Philosophy

There are, generally speaking, several areas of Skinner’s treatment of private events which can be contrasted with positions in contemporary philosophy. One of these concerns what mental predicates can be properly construed as actually referring to private events or events under the skin. Another area involves the analysis of concepts thus localized. Radical behaviorists on occasion suggest that Skinner’s approach is (1) confused with older forms of behaviorism denying that private events are a proper subject matter for psychology; (2) confused with logical positivism, which in its earlier Carnapian form provided for the translation of mentalistic concepts into physicalistic reduction sentences (Carnap, 1932/1933; 1953); (3) confused with logical behaviorism which declares that references to seemingly private events are in reality references to behavioral dispositions of a predominantly public kind; (4) confused as to the causal role of private events in public behavior; and (5) similar to Wittgenstein’s analysis of privacy.

Wittgenstein and Skinner

With respect to (5), Day (1959) and Costall (1980) have drawn attention to the alleged affinities between Wittgenstein and Skinner. However, the disparities are also noteworthy, and speak to fundamental differences between the views of both men. The similarity of viewpoint has been stressed because both were skeptical about the possibility of first-person knowledge of private events like sensations. But Wittgenstein’s skepticism sprang from his view that that the concept of “knowledge”—entailing as it does the possibility of being wrong—was not applicable to one’s experiencing sensations like pains. In the case of my sensations, claiming that I know I am in pain is tantamount to forcing the verb to “go on holiday” from its accustomed usage in utterly different linguistic contexts, since the grammar of first-person sensation statements is different than it is for statements about public objects: 

246. In what sense are my sensations private?—Well, only I can know whether I am really in pain; another person can only surmise it.—In one way this is false, and in another nonsense. If we are using the word  “to know” as it is normally used (and how else are we to use it?), then other people very often know when I am in pain.—Yes but all the same not with the certainty with which I know it myself!—It can’t be said of me at all (except as a joke) that I know I am in pain. What is that supposed to mean—except perhaps that I am in pain?

Other people cannot be said to learn of my sensations only from my behaviour,--for I cannot be said to learn of them. I have them.

The truth is: it makes sense to say about other people that they doubt whether I am in pain; but not to say it about myself. (Wittgenstein, 1958, p. 89e).

 Skinner, on the other hand, felt that self-knowledge about sensations was indeed achievable, but only as a by-product of the way the verbal community reinforced self-descriptive repertoires in the presence of the requisite private event. The handicap here is that the verbal community is in a disadvantaged position with respect to determining whether the covert event is actually occurring, because it cannot make reinforcement contingent upon the property of privacy. For Skinner, skepticism about private events was born of the shakiness of first-person “knowledge” about them before the verbal community created the relevant discriminations; for Wittgenstein, the concept of first-person “knowledge” of one’s sensations was always a misnomer because the notion of knowledge is inapplicable in relation to one’s private sensations. Skinner felt that “knowledge” of another’s private sensations was always inferential; Wittgenstein felt otherwise. I hear something in the other room that sounds like a groan—but I cannot be sure. In such a context, I infer someone is in pain. But when the victim of an industrial accident is howling and writhing on the floor in plain sight of co-workers, his arm mangled and bloody, I am no longer inferring his being in pain: I know this, according to Wittgenstein (i.e., knowing others are in pain is often not merely a surmise, a.k.a. ”inferential”): 

303. “I can only believe that someone else is in pain, but I know it if I am.”—Yes: one can make the decision to say “I believe he is in pain” instead of “He is in pain”. But that is all.—What looks like an explanation here, or like a statement about a mental process, is in truth an exchange of one expression for another, which while we are doing philosophy, seems like the more appropriate one.

Just try—in a real case—to doubt someone else’s fear or pain (Wittgenstein, 1958, p. 102e). 

Austin (1960, pp. 83-84) had a like-minded tack on knowledge about others’ private states, and one highlighting the fact that Skinner raises the bar quite high for third-person certainty:

One speaker…said roundly that the real crux of the matter remains still that ‘I ought not to say that I know Tom is angry, because I don’t introspect his feelings’: and this no doubt is just what many people do boggle at. The gist of what I have been trying to bring out is simply:

1. Of course I don’t introspect Tom’s feelings (we should be in a pretty predicament if I did).

2. Of course I do sometimes know Tom is angry.

Hence

3. to suppose that the question ‘How do I know that Tom is angry?’ is meant to mean ‘How do I introspect Tom’s feelings?’ (because, as we know, that’s the sort of thing that knowing is or ought to be), is simply barking our way up the wrong gum tree (Austin, 1960, pp. 83-84). 

If for Skinner knowledge about private events in others is always inferential, what data would be sufficient under his formulation to satisfy requirements for certainty about such states? Here, we face a conundrum. According to Skinner, first-person “knowledge” about private events is dependent upon reinforcement by the verbal community, which in turn is saddled with an inferential mode in assessing private states in others. If neither the verbal community nor the individual in question possesses certain knowledge about the latter’s private states (because the latter under the formulation is dependent upon the former for its discriminative powers), what hypothetical information would clinch a case for certainty? Since the very concept of inference entails the possibility of being mistaken should it prove false, does radical behaviorism imply that beliefs about private states always fall short of certainty? Is this tantamount to implying that there is a possibility, however remote, that private states in others may be totally misconstrued, or, in a worst case scenario, that there may not be private events at all—because the verbal community and, a fortiori, the individual who comes to know himself precisely because of its tutelage, have both been led down the wrong epistemic path?

Far from sharing affinities, the views of both Wittgenstein and Skinner about private events contrast sharply. To summarize: the former felt that I cannot “know” I have a toothache because the concept of self-knowledge is inapplicable to private sensations; Skinner felt that I can make a mistake about my sensations in the same way I can about other physical events, especially when I have not been taught to discriminate them properly by a verbal community. 

Wittgenstein’s discussions of privacy at Cambridge University—dutifully attended by such enraptured students as G. E. Moore—were accorded the title “the Toothache Club.” Toothaches are a convenient springboard for discussions of private events to this day. The opinion of a celebrated contemporary academic addressing the problem from a radical behavioristic perspective illustrates the point:

A toothache is a physical event, but the person with the toothache has access to it in a different way than does the dentist who is called on to treat it. Both respond to the unsound tooth, but one does so by feeling the tooth and the other by looking at it and probing it with instruments. The different contact with the tooth might be compared with the different ways a seeing person and a sightless person make contact with a geometric solid, if one is trying to teach its name to the other, the seeing person does so by sight and the sightless person by touch. One kind of contact is not necessarily more reliable than the other. For example, in the phenomenon of referred pain, a bad tooth in the lower jaw may be reported as a toothache in the upper jaw. In this case the dentist is a better judge of where the pain really is (another significant example is when a victim of a heart attack reports pain in the shoulder or in the small of the back rather than within the chest).

We often think of private events such as our feelings and our thoughts as ones to which we have privileged access and therefore of which we have special knowledge. But we learned the relevant words from others, and they all had access to in teaching them to us were the public correlates. If we can be mistaken even about the location of a toothache, what assurance do we have that any of our other reports of private events are reliable? (Catania, 1992, p. 1526).

Skepticism and Certainty

Where to begin? The author avers that a dentist can be a better judge of where my pain is than I when I have a toothache, and goes on to broaden skepticism about first-person reports of sensations by indicating that any doubts I should have about the location of my toothache has implications for my certainty about other of my private sensations. On the contrary, my dentist is not a better judge of where my pain is than I am, only a better judge of the location of the decay in the troublesome tooth in question. In fact, if, after inspecting my mouth to probe the toothsome event, he were to suddenly announce I could not be experiencing the pain where I (truthfully) said I was experiencing it, I could well challenge his understanding of the grammar of sensation reports (Wittgenstein, 1958). I might point out, for example, that the very concept of referred pain contradicts the claim that it is the dentist, not I, who is the ultimate arbiter of where my pain is—whatever the physical location of the decay giving rise to it. For consider: if my dentist were the final authority on where I feel my pain—presumably because my pain is a physical event, and he is the expert on localizing this—then the very idea of referred pain is incoherent. That is, if, according to the radical behaviorist, the locus of the pain had to be the same as the physical event giving rise to it, how does the notion of referred pain get a foothold in any account of what was going on in the dentist’s office? In other words, how can I have any such thing as “referred pain,” implying as it does a sensation localized some distance away from the putative physical event giving rise to it, when the radical behaviorist insists that I can make a mistake about the location of my pain when it does not jibe with where my dentist claims I must be feeling it? For referred pain is precisely that: a sensation localized away from the relevant physical event, a logical impossibility on the radical behavioristic account of private events. The meaningfulness of the very concept of referred pain is parasitic on a perceived location of a sensation differing from the location of the physical toothsome event. This means that an instrumental invasion of the organism—a future technology Skinner has envisioned as pinning down the actual physical characteristics of the private event—would still have to come to terms with any paradox occasioned by the perceived, or experienced location of the sensation in question. This may become a complicated affair, but the complication would not eventuate in correcting one’s impression of where he or she actually feels the sensation.   

The same logic holds for the phenomenon of phantom limb in amputees. In these cases, a patient may feel a pain that is experienced outside his body; he may feel it, for example, in the region of toes that no longer exist. Yet it would be absurd to suggest that he must be mistaking the perceived location of the pain because neurologists inform us that the physical event giving rise to it occurs elsewhere under the skin. That it does—and it surely does—hardly precludes an amputee’s experience of pain outside his body. We can, of course, undertake to convince such a patient that he is mistaking the location of perceived pain, although we do so at the risk of engaging in mere language reform, not engineering an insight more in accord with the supposed reality of the physical world.

The foregoing argument may not conclusively confound the Skinnerian thesis that if a private event like a toothache is ultimately a physical one, and that the sensation of ache must be localized in the same place as the physical event it is. A radical behaviorist may go on to insist that his thesis can be salvaged by denying that the physical event in question is actually the decaying tooth. He could hold, for example, that the relevant physical event is a happening within the organism mediating between the ache and the decaying tooth that meets a requirement of identity of locus with the sensation. But arguments aimed at establishing such spatial harmonies can at best serve to bolster a plausible case for Identity Theory (Place, 1956; Feigl, 1967; Smart, 1959). They cannot be used to defend the counterintuitive notion that when the physicalistic cards are on the table, dentists are in a better position to localize my sensations than I am.

What could have led the author of the above passage to make the claim that there are others who might instruct me how to correct my impression of where I feel my pain, because they have a better take on the physical picture of things, as dentists do? The answer is not hard to find, and some of the blame is traceable to equivocation in many of Skinner’s pronouncements about the subject. For example, in his 1945 essay Operational Analysis of Psychological Terms, he remarks:

The additional hypothesis follows quite naturally that being conscious, as a form of reacting to one’s behavior, is a social product. Verbal behavior may be distinguished, and conveniently defined, by the fact that the contingencies of reinforcement are provided by other organisms rather than mechanical action upon the environment. The hypothesis is equivalent to saying that it is only because the behavior of the individual is important to society that society in turn makes it important to the individual. The individual becomes aware of what he is doing only after society has reinforced verbal responses with respect to his behavior as the source of discriminative stimuli. The behavior to be described (the behavior of which one is to be aware) may later recede to the covert level, and (to add a crowning difficulty) so may the verbal response. It is an ironic twist…that as we develop a more effective vocabulary for the analysis of behavior we also enlarge the possibilities of awareness, so defined. The psychology of the other one is, after all, a direct approach to “knowing thyself.” (Skinner, 1945, p. 593). 

The italicized phrases in the above passage are the present author’s. They seem to imply—although in a possibly murky way—that “self-awareness” is something more than merely learning to speak correctly about oneself. They appear to suggest that the “possibilities of awareness” created for me by the verbal community encompass a broader achievement than correct naming. This is also implied by Catania’s assertion that one’s perception of pain may stand in need of correction. Other of Skinner’s remarks on private events smack of similar equivocations. For example, in About Behaviorism he remarks:

…if certain ways of teaching a person lead him to notice very small differences in his “sensations,” and if because he sees these differences he can classify colored objects correctly, then it should follow that we can use these ways of teaching him to classify objects correctly (Skinner, 1974, p.13) 

It is difficult to envision how being trained to “see” differences among one’s sensations amounts to nothing more than a species of linguistic skill (i.e., attaching the right terms to certain sensations). As if to emphasize the idea that Skinner’s externally derived notion of self-awareness encompasses more than merely a species of linguistic achievement, he remarks in Beyond Freedom and Dignity that:

And it is because such questions are asked that a person responds to himself and his behavior in the special way called knowing or being aware. Without the help of a verbal community all behavior would be unconscious. Consciousness is a social product (Skinner, 1971, p. 192).

Consciousness—at least in the ordinary sense of the use of this term—means something more than being able to attach the term pain to the appropriate sensation; it also means being aware (having, experiencing, perceiving, discriminating, harboring, sensing) the sensation when I have it, irrespective of whether or not I deploy the term pain correctly or in accordance with whatever linguistic conventions have been establishedbyaverbal community.                                                                                                                                                                  

Hidden Equivocations

 When a radical behaviorist uses terms like “discriminate,” the picture he may have of the term is that of a linguistic token in a set of verbal practices from which the vagaries of ordinary language have been pruned. He may even come to believe his verbal behavior permits more precision, being less vulnerable to the ambiguities of ordinary parlance. Perhaps this is true—some of the time. However, in the present context, it would appear that the “discriminations” trained by the verbal community in shaping self-awareness may encompass at least two achievements not systematically distinguished in the Skinnerian canon. For what is the precise nature of the alleged discrimination? If the verbal community is ultimately responsible for my awareness of my own private states, is this tantamount to training me to become aware of or distinguish experientially among my sensations, or training me to describe myself accurately when I have them, or both? The distinctions here are weighty ones, and they tend to get obscured underneath jargon about “discrimination.”  Furthermore, to the extent this jargon detracts from the ability to differentiate clearly between two rather different claims, it is the language of obfuscation, not clarity.

We are not born knowing how to use certain terms, so our dependence upon a verbal community to teach us the language of self-ascription is hardly a bold or challengeable claim. That on some basic level we must rely on others to learn how to speak about ourselves—or anything else for that matter—is a common assumption that tends to get pawned off as an epiphany when encased in terms like “discrimination,” “collateral responses,” and “reinforcement.”  That we need the assistance of others to feel a sensation, or otherwise “discriminate” among certain of our sensations is a patently absurd claim—for several reasons. Lower mammals experience pain without a trace of anything akin to a verbal community to enable this ability. Accordingly, to dispute this on the human level would appear to represent a break with the best informed Darwinian extrapolations. In addition, if we are dependent on others to accomplish such a feat, this implies we can be taught to “discriminate” our sensations improperly should the verbal community perversely decide to corrupt its instruction. On the contrary, mismanaging the task of verbal reinforcement is just that: attaching the wrong terms to whatever private states are under discussion; it is hardly akin to reordering or otherwise altering the private sensations of individuals acquiring language skills.

Imagine that as a developing and non-verbal organism I experience a toothache, and the verbal community decides to play a trick on me. It elects to reinforce me for saying I feel euphoric whenever I emit a collateral response ordinarily associated with toothache. Does the radical behaviorist imply I subsequently experience a mania in contexts in which others experience pain; or, alternatively, experience toothache but use the wrong words to describe it? Calling the effects of improper outside reinforcement breakdowns in “self-awareness” would appear to be a rather overblown way of characterizing bungled linguistic training about my vocabulary of self-description—if this is all “self-awareness” amounts to. After all, if I claim to be euphoric when I am actually in pain, my disadvantage is only that I have not mastered English well, not that I need an opiate.

The upshot of one interpretation of the radical behavioristic thesis about private events—that “self-awareness” encompasses considerably more than teaching the proper vocabulary of self-ascription—is that there are no collateral responses to pain indelibly stamped as such. In this connection, if the verbal community on occasion gets it wrong or botches up its tutelage, then the character of the private event will shift as a function of the net improper reinforcement. On this interpretation, if I experience euphoria where another experiences pain because of an incompatible history of reinforcement by the verbal community, what entitles the latter to assume there is any such thing as a reliable connection between a particular class of collateral physical responses and the name or identity of the private event that should be verbally reinforced? In this context, “collateral” responses cannot occasion characteristic reinforcements by the verbal community, since such reinforcement itself creates the character of the private event in question! It becomes, as it were, its own self-fulfilling prophecy. Even if instrumental invasion of the organism were to reveal the same physiological events underpinning pain and euphoria, the conclusion might be that dissimilar sensations are mediated by the same physical correlate, not that the wrong term has entered into the language of self-ascription.  I take this to be a reductio ad absurdum of the claim that verbal communities train individuals to “discriminate” their private states in the sense of perceiving differences among them on other than a naming, or linguistic, level.   

The radical behaviorist is on the horns of a dilemma. Either his approach to private events on one interpretation entails counterintuitive consequences, or else embodies a theory of verbal acquisition which, underneath the technical terminology, seems actually quite trivial, boiling down to the common wisdom that others teach us how to speak about ourselves. And that this interpretation is not always the chosen one is made amply clear in Catania’s assertion that a dentist knows better than I do about the perceived location of my toothache.  If I can make a mistake about where I locate a pain, why not the possibility of error about whether I actually have one or not? I can only make a mistake about this on the theory that my discriminatory abilities are out of keeping with what others know about the matter because of the initial epistemological edge they enjoy about events under my skin. All the same, one wonders whether the skepticism about first-person knowledge of private events prior to reinforcement by the verbal community is a deduction from a strained theory, and not in keeping with the natural history of the species .

Mistrusting Verbal Behavior
 

Skinner has in the past expressed views that seem to derive from a concocted way of looking at a problem, and one that is heedless of the realities of cultural practice. For example, in Science and Human Behavior he remarks:

The layman also finds the lack of a reliable subjective vocabulary inconvenient. Everyone mistrusts verbal responses which describe private events. Variables are often operating which tend to weaken the stimulus control of such descriptions, and the reinforcing community is usually powerless to prevent the resulting distortion. The individual who excuses himself from an unpleasant task by pleading a headache cannot be successfully challenged, even though the existence of the private event is doubtful. (Skinner, 1953, p. 260). 

What set of actual cultural practices do such remarks portray? Laymen are not bereft of reliable subjective vocabularies, nor does everyone mistrust verbal responses describing private events. Great emphasis is placed upon first-person reports about the location, quality, and intensity of pain in standard medical and dental examinations, and skepticism about them on the part of clinical practitioners, with the exception of factitious phenomena, is virtually non-existent. The same is true for reports of various kinds of sensations outside of clinical contexts. Skinner has overgeneralized a skepticism about verbal reports of private events from special cases like those in which, for example, liars report headaches in order to avoid onerous obligations. He has, in effect, portrayed a fictional picture of cultural practice, possibly because of deductions from his view about how private events come to control behavior. Who distrusts the private experience of pain as reported by an ingenuous youth who starts to yelp after an anvil is accidentally dropped on his foot? Who mistrusts the scream of pain from a dental patient who is having a wisdom tooth extracted without anesthesia? What cardiologist mistrusts the report of chest pain from a patient suddenly experiencing such a symptom? Negative results on an EKG or stress tests do not ordinarily occasion doubts about subjective reports of pain in such cases. At best, they only eventuate in a different diagnosis: reflux disorder or muscle strain, perhaps.

Skinner has in effect recast scenarios common to cultural practices, taking skepticism about the statements of liars as exemplifying the standard response to reports of private states. Even here, however, his depiction is faulty. We often distrust the statements of those who plead headache in the context of unpleasant tasks, not because their statements are about private events, but because we have reason to believe that they may be dissembling—especially when they have a past record of this. Moreover, we do, as a matter of cultural practice, challenge the claims of other people reporting private states or otherwise. When liars aver that they were not accomplices in vandalism or at a break-in at someone’s home, do we disbelieve them because what they report was a public event?

Metaphors and Metonymical Extension

Another example of a questionable theory of verbal acquisition is Skinner’s remarks about metaphorical extension in Verbal Behavior. In enumerating the several ways he feels verbal control over the private event is instituted, he remarks:

A third possibility is that the community may not need to appeal to private stimuli at all; it may reinforce a response in connection with a public stimulus, only to have the response transferred to a private event by virtue of common properties, as in metaphorical and metonymical extension. It has often been pointed out that most of the vocabulary of emotion is metaphorical in nature. When we describe internal states as “agitated,” “depressed,” or “ebullient,” certain geometrical, temporal, and intensive properties have produced a metaphorical extension of responses.

Not all metaphorical expressions evoked by private stimuli exemplify this principle. Although a sharp pain or a burning sensation may illustrate metaphorical extension arising from a similarity between the stimulation supplied by sharp or burning objects and certain private stimuli, another explanation is possible. The metaphorical step may have occurred before the response receded to the private world. In that case we should have no reason to look for a private stimulus having similar properties. If the response sharp is first acquired in connection with certain objects with identifiable physical properties not related to their effect upon the human organism—for example, if a needle is called sharp if it shows a certain geometrical pattern or easily penetrates paper or cloth, or if a knife is called sharp if it readily cuts wood, then the extension of the response to a certain type of painful stimulus generated by pricking or cutting is metonymical (Skinner, 1957, p. 132).

Skinner saw no problem in his notion of “metaphorical extension,” although his analysis may represent the putative confusion of drawing parallels between features of private and public stimuli and providing a plausible account about the transformative processes under discussion. For example, to show that the metaphor a stabbing pain  mirrors the physical properties of knives or their effects on us when we are stabbed is hardly a satisfactory explanation of why I am able to exclaim I have a stabbing pain. Skinner goes on to say that I have a stabbing pain is instituted when it “shares some of the properties of the stimuli produced by sharp objects” (Skinner, 1957, p. 133). What if an individual were never stabbed, or never experienced the private events on the basis of which the transference supposedly takes place; could such an individual never emit I have a stabbing pain? What knives do to paper or cloth will not provide the necessary basis for transference; in such cases there has to be a history of private stimulation to afford a basis for a transfer from a public feature to the private event.

The point may be better illustrated in connection with I have butterflies in my stomach. What public event is at the basis for such a metaphorical extension to a private event—especially since having actual butterflies in the gastrointestinal tract has little or no relation to the either the metaphor or the private event it purportedly describes.

Skinner’s reliance on metaphorical extension is seemingly in the service of relying on public stimuli to sharpen awareness of the inchoate or dubious world of private stimulation. Thus, we learn to discriminate private states by relying on external exemplars: the piercing pain, the bubbling up sensation, the hot flash, the burning feeling, and so on. Doesn’t the process also go in the other direction? Do not individuals illuminate something about public stimuli or the external world by characterizing it through metaphors based upon private states?  For example: the world is a depressing place, his remarks were painful, her boyfriend is “hot,” her words were cold, life is like a dream, he brought ill tidings, she’s a dizzy blonde, he writes nervous prose, politicians are a headache, mothers-in-law are pains in the neck, he makes one’s flesh crawl, he was itching to debate the point, glum news, languid prose, anxious times, a heartfelt exchange, a sullen invitation, a soulful poem, a raging sea, a manic exchange, a dour message, a hurtful announcement, a boring speech, a thoughtful letter, disappointing news, a downbeat day, a sentimental journey, an insane choice, an angry ocean, a disgusting remark, a consuming hobby, a reflective stance, a nauseating speech, and so on—ad infinitum, it would seem. If publicity gives privacy its discriminative character, isn’t the reverse also true?

Finally, it should be noted that Skinner’s analysis frequently begs the question. In the metaphor the tempest in my mind, the internal event is not somehow clarified by reference to the public event that shares characteristics with it on the metaphoric level. The explanatory challenge is showing precisely why the metaphoric extension in this case is apt. The description tempestuous feelings does not enter the verbal repertoire because a feeling is like a gale simpliciter. Why the feeling is like a gale in the first place is the explanatory challenge for radical behaviorism. One cannot, as it were, point to the “similarity” between feelings and storms to prove Skinner’s point since the “similarity” may exist only because of the existence of the metaphor.      

Stimulus Similarity

I have the same pain I had an hour ago depends upon the notion of stimulus similarity between past and present pains. The radical behavioristic treatment of private events should prompt its proponents to raise the disquieting question: “How does one know two sensations are identical or similar if it is possible to be mistaken about it?” There a problem here that cannot be dispatched by relegating it to a process of generalization or what Skinner has called “induction.” For how do I assess the similarity of pains unless “sameness” is reinforced like every other feature of my private experience? In other words, how, under radical behaviorism, can I ever learn that the pain I had today is the same as the one I had yesterday when I must learn to discriminate them as such with the assistance of a verbal community that reinforces verbal expressions of “sameness?” After all, if self-awareness of private events is shaped by the verbal community, my ability to tell whether consecutive events under my skin are experienced as the same or different from each other likewise ought to be subject to outside tutelage. However, the collateral responses associated with reports of similar sensations may differ considerably, not to mention the fact that private sensations can be experienced without any collateral behavior serving to occasion outside verbal reinforcement. For example, my reaction to a pain I experience today may be muted—or non-existent: a dramatically different response than the reaction to the same pain I had yesterday when I howled and grimaced. Today, I stoically brace up as I experience the same sensation. If I must learn to detect “sameness”—or “differences” among my sensations on the basis of how the verbal community enables me to discriminate them, how is this conceivably accomplished? Something about the behavior-analytic explanation of the entire process appears to be viciously circular.

In Verbal Behavior Skinner addresses the possibility of metaphorical extension by noting, “…in expressions like ebullient or dampened spirits, however, we must search for possible similarities between public and private events to explain the metaphorical extension. Something within the skin must ‘bubble up’ or ‘grow limp or cold’ in some sense (Skinner, 1957, p. 133). His example, however, begs a critical question. The problem is only stated, not resolved, by hypothesizing similarity between public and private events by way of metaphor. Clinching a solution to the problem would amount to explaining how the metaphor comes to be apt in the first place. In other words, how does one learn to attach the phrase “bubbling up” to the event it purports to describe, even by way of metaphor? What instruction has the individual received in his or her past history of reinforcement that enables the emission of “bubbling up” or “dampening down” in contrast, maybe, to “drowning” or “heating up”? In addition, claiming that the private event must share in some sense with a “bubbling up” characteristic of the covert occurrence implies an isomorphism between the metaphor and the physical event it delineates. Yet instrumental invasion of the organism might not reveal any such isomorphism. The latter, if anything, would seem to characterize the relationship between a sensation or raw feel (i. e., qualia) and a metaphor solely, not between the covert physical event identified with the sensation. A neurophysiologist from a third-person perspective could conceivably isolate an α-process that is the physical aspect or underpinning of a β-type sensation, whereas a metaphor only captures a β-aspect of the event in question. As an identity theorist, Skinner has claimed that so-called private “mental” events like sensations are physical events under the skin. He also feels that physiologists who conduct instrumental invasions of the organism are nonetheless debarred from access to that aspect of the private event experienced by the individual whose event it is. Thus, the physiologist can hypothetically observe the underlying physical event, but not its sensational “quality,” in the individual he investigates (Creel, 1980). Accordingly, the neurophysiologist who locates the physical event which is the underpinning of the sensation experienced by an individual without, however, knowing anything about the felt character of the sensation in question, could not determine what metaphorical extension would be appropriate in such a case. It is the experienced aspect of the private event, not its physicality from a third-person perspective that establishes the sympathetic connection between it and a metaphor.

Radical Behaviorism vs. Mentalism

Skinner’s radical behaviorism is presumably an approach that seeks to avoid the pitfalls of “mentalism.” The latter type of philosophical theory is said to be one that has characterized most psychological systems prior to radical behaviorism. The trend, according to Skinner, included reliance on cognitive way-stations, introspection, mediating, or indefinable variables owing their provenance to Cartesian dualism, or the theory that private states take place in a metaphysical order or medium differing from the physical world Skinner believes is the proper arena of a scientific focus. Mentalistic terms and concepts according to him thus need to be shorn from the behavioristic approach to analyzing behavior. The proposal may be hasty for two reasons: (1) the assault on mentalism is in part based upon negative existential propositions, and (2) there are mentalistic concepts which are not only not redundant in a scientific approach to behavior; they may be indispensable to it, as Hempel decades ago came to realize (Hempel, 1980; Rey, 1997).

(1) Skinner’s criticism of mentalistic explanations takes the bull by the horns: events under the skin are physical, not immaterial ones in a mentalistic medium, as advertised by dualism. How is such a philosophical position to be understood?—surely not as an empirical thesis, since Skinner has never specified what data would instantiate ghosts in the machine. In short, the proposition that private events are ultimately physical in nature loses some of its meaning when the falsifiability (Popper, 1959) of the alternative theory remains unspecifiable, unclear, or shrouded in mystery. 

 (2) In his animal laboratory, an experimenter places a pigeon on a fixed ratio schedule of reinforcement under which the latter is delivered every third response (FR3 scheduling). There nonetheless exists an undetermined—not undeterminable, only unverbalized—interval schedule necessarily coinciding with the FR3 schedule already specified. This is because for every ratio (R) schedule there exists an interval schedule of reinforcement (I) coinciding with it. Interval schedules are those programmed to deliver reinforcement based upon the time elapsing between responses. Every R-schedule is extensionally equivalent to an overlapping I-schedule and vice-versa, implying that differing R-and I-schedules are denotatively the same. (This is not to claim that all interval and ratio schedules of reinforcement are coextensive; only that for any one of them there is an extensionally identical twin of the other type in the parallel universe of denotation.) Thus, the pigeon placed on the FR3 schedule is also placed on a variable interval schedule (VI) coinciding with it. If the pigeon takes ten minutes to emit the first three responses before reinforcement, five minutes to emit the next three responses before reinforcement and two minutes to emit the next three responses before reinforcement, it is thereby also subject to a VI schedule of reinforcement. In the improbable event that the time elapsing between the three reinforcements was always the same, say, five minutes, an FR3 schedule coincides with a FI5 schedule. The question for the radical behaviorist is: what is the operative schedule programmed by the E? The answer is seldom “both,” and for good reason. The radical behaviorist must assume it is the E’s intention—a mentalism—that ordinarily determines schedule type. At least this is the implication, since what other criterion of schedule-type is relied on to explain an E’s selective verbal behavior when differing schedule types describe the same events?

In the manner of a distinction drawn by W. V. Quine (Quine, 1972), we might wish to say that both types of scheduling fit the facts, but the E’s arrangement guides the FR3 scheduling. However, “guiding” in Quine’s sense involves a causal relationship, and unless we assume that the E’s intention has causal force—a hunch that would be anathema to a radical behaviorist—we cannot restrict stimulus control to one schedule when two different ones fit the facts. True, the E is aware of intending to program the FR3 schedule, rather than the VI schedule, but “awareness” and “intention” are presumably dethroned as causally relevant in radical behavioristic formulations. Awareness is supposed to be a by-product of the training by the verbal community, while intention is a mere mentalism to be extruded from scientific accounts. Unfortunately, it will tend to rear its banished head even in relation to boilerplate distinctions of the experimental lab.

REFERENCES

Armstrong, D. M. (1968). A materialist theory of the mind. New York: Humanities Press.

Aune, B. (1961). The problem of other minds. Philosophical Review, 70, 320-339.

Austin, J. L. (1961). Philosophical papers. London: Oxford University Press.

Ayer, A. J. (1946). Other minds. Aristotelian Society Supplementary Volume, 28, 188-197.

Baum, W. M. and Heath, J. I. (1992). Behavioral explanations and intentional explanations in psychology. American Psychologist, 47, No. 11, 1312-1317.

Catania, A. C. (1992). B. F. Skinner, organism. American Psychologist, 47, No. 11, 1521-1530.

Carnap, R. (1932/33). Psychology in physical language. Erkenntnis, 3, 107-142.

Carnap, R. (1953). Testability and meaning. In H. Feigl and M. Brodbeck, Readings in the philosophy of science, (Pp. 47-92), New York: Appleton-Century-Crofts.

Costall, A. (1980). The limits of language: Wittgenstein’s later philosophy and Skinner’s radical behaviorism. Behaviorism: A Forum for Critical Discussion, 8, 123-131.

Creel, R. (1980). Radical epiphenomenalism: B. F. Skinner’s account of private events. Behaviorism: A Forum for Critical Discussion, 8, 31-53.

Day, W. F. (1969). On certain similarities between The Philosophical Investigations of Ludwig Wittgenstein and the operationism of B. F. Skinner. Journal of the Experimental Analysis of Behavior, 12, 489-506.

Day, W. F. (1980). Some comments on the book “Verbal Behavior.” Behaviorism; A Forum for Critical Discussion, 8, 165-173.

Dinsmoor, J. A. (1992). Setting the record straight: The social views of B. F. Skinner. American Psychologist, 47, No. 11, 1454-1463.

Feigl, H. (1967). The mental and the physical. Minneapolis, Minnesota: Univerity of Minnesota Press.

Glenn, S. S., Ellis, J., and Greenspoon, J. (1992). On the revolutionary nature of the operant as a unit of behavioral selsction. American Psychologist, 47, No. 11, 1329-1336.

Hempel, C. (1980).  The logical status of psychology. In N. Block (Ed.), readings in the philosophy of psychology, (Pp. 14-33), Vol I, Cambridge, Massachusetts: Harvard University Press.

Kirby, R. V. (1956). The other minds quandary. Ph.D. thesis. Berkeley, California: University of California.

Lee, V. I. (1992). Transdermal interpretation of the subject matter of behavior analysis. American Psychologist, 47, 1337-1343.

Martin, M. (1978). Interpreting Skinner. Behaviorism: A Forum For Critical Discussion, 6, 129-138.
 

Moore, J. (2001). On psychological terms that appeal to the mental. Behavior and Philosophy, 29, 167-186.

Palmer, D. C. and Donahoe, J. W. (1992). Essentialism and selectionism in cognitive science and behavior analysis. American Psychologist, 47, No. 11, 1344-1358.

Place, U. T. (1956). Is consciousness a brain process? British Journal of Psychology, 57, 44-50.

Popper, K. (1959). The logic of scientific discovery. London: Hutchinson.
 

Putnam, H. (1957). Psychological concepts, explication, and ordinary language. Journal of Philosophy, 54, 94-100.

Rachlin, H. (1992). Teleological behaviorism. American Psychologist, 47, No. 11, 1371-1382.

Rakos, R. F. (1992). Achieving the just society in the 21^st century: What can Skinner contribute? American psychologist, 47, No. 11, 1499-1506.

Rey, G. Contemporary philosophy of mind: A contentiously classical approach. Oxford: Basil Blackwell.

Schnaitter, R. (1978). Private causes. Behaviorism: A Forum For Critical Discussion, 6, 1-12.

Sellars, W. (1953). A semantical solution of the mind-body problem, Methods, 5, 45-82.

Skinner, B. F. (1945) The operational analysis of psychological terms . Psychological Review, 52, 270-294.

Skinner, B. F. (1957). Verbal behavior. New York: Appleton-Century-Crofts.

Skinner, B. F. (1969). Contingencies of reinforcement. New York: Appleton-Century-Crofts.

Skinner, B. F. (1974). About behaviorism. New york: Alfred A. Knopf.

Skinner, B. F. (1971). Beyond freedom and dignity. New York: Alfred A. Knopf.

Smart, J. C. C. (1959). Sensations and brain processes. Philosophical Review, 68,141-156.

Thomson, J. F. (1951). The argument from analogy and our knowledge of other minds. Mind, 40, 336-350.
 

Todd, J. T. and Morris, E. K. (1992). Case histories in the steady power of misrepresentatation. American Psychologist, 47, No. 11, 1441-1453.

Quine, W. V. (1972). Methodological reflections on current linguistic theory. In D. Davidson and G. Harman (Eds.), Semantics of natural language, (Pp. 442-454), Dordrecht, Netherlands: D. Reidel.

Wisdom, J. (1956). Other minds. Oxford: Basil Blackwell.