|Eric LaRock, PhD
Assistant Professor of Philosophy
348 O’Dowd Hall
Rochester, MI 48309
Consciousness is not a Self-Scanning Mechanism: A Response to Armstrong
Several contemporary philosophers think the best way to uncover the nature of mind is through the empirical sciences. That is because, for many, it is not obvious that a priori conceptual analyses alone could provide sufficient grounds to conclude in favor of either materialism or dualism.1 We need to supplement our conceptual analyses with empirical investigations to make headway on the reality of mind and its place in nature. Although there are wide disagreements between theorists working on the nature of mind, especially regarding the nature of consciousness, David Armstrong believes that a materialist view of mind is gaining converts and will eventually become “scientific doctrine.”2 What does Armstrong mean by a materialist view of mind? It is “the view that we can give a complete account of man in purely physico-chemical terms.”3 Armstrong goes on to assert that scientists and philosophers who resist this materialist view of mind “do so primarily for philosophical, or moral or religious reasons, and only secondarily, and half-heartedly, for reasons of scientific detail.”4 Perhaps because Armstrong does not want to appear too dogmatic, he goes on to acknowledge the possibility that empirical investigations might reveal “new evidence and new problems” that will force advocates of materialism to “reconsider the physico-chemical view of man.”5 Armstrong is nevertheless convinced (on the basis of a vague reference to advances in molecular biology) that empirically minded philosophers should join hands in formulating a theory which maintains that mind is “nothing but a physico-chemical mechanism” that functions to cause (and therefore explain) certain sorts of behavior, such as the behaviors involved in distinguishing, recognizing and perceiving objects, as well as speaking, and any other behavior animals (both human and nonhuman) are capable of performing.6 Compatible with his broader logical, metaphysical, and commonsense commitments, Armstrong maintains that mind plays a causal role in our cognitive and behavioral lives. In this strict sense, Armstrong is like Descartes in that he too attributes causal power to the mind itself; rightly so, they have no sympathies with theories that deny the reality of mental causation, such as behaviorism and epiphenomenalism. Despite their allegiance to the reality of mental causation, these two theorists differ quite radically about the ontology of mind. Unlike Descartes’ stance on a variety of substance dualism, Armstrong believes that every mental state can, and most likely will, be explained in terms of material states without remainder; after all, a mental state is, for Armstrong, nothing but a cause in the central nervous system that is capable of bringing about behavior.
Even though Armstrong argues primarily on conceptual grounds that his materialism is explanatorily better than both classical and logical species of behaviorism, he also acknowledges that any view that reduces mind to a material cause inevitably faces a serious problem, what has become known as the hard problem of consciousness—the problem of trying to accommodate first-person conscious mental phenomena within a purely third-person, materialist explanatory framework.7 Armstrong’s purported solution to this hard problem is to explain first-person conscious mental phenomena in terms of “a self-scanning mechanism in the central nervous system.”8 He claims that this mechanism is essentially a selective mechanism that enables an animal with perceptual capacities to distinguish between the features of objects of any given visual scene, such as the kind of selective behavior that would be required to distinguish between a green object and a red object. After highlighting some of the positive aspects of Armstrong’s position, I present an empirical case against Armstrong’s claim that consciousness is nothing but a self-scanning mechanism.
2. Classical and Logical Behaviorism
Armstrong finds classical and logical species of behaviorism implausible because both entail the denial that mental states are the actual causes of behavior. Before examining Armstrong’s case against behaviorism, it might be helpful to start by reflecting on a core theoretical claim made by B.F. Skinner, a pioneering advocate of the classical behaviorist view. Skinner claimed that, under the hypothesis of behaviorism, conscious mental states are ultimately reducible to behavior: “We may take feeling to be simply responding to stimuli.”9 This type of claim is compatible with reductive materialism in the philosophy of mind, which maintains that all mental states can be accounted for in terms of material states without remainder. Classical behaviorists supposed that once you have explained the whole range of our behavioral lives, you have explained the whole range of our mental lives. Armstrong provides a pithy summary of the philosophical upshot of classical behaviorism: “Thought is not an inner process that lies behind, and brings about, the words I speak and write: it is my speaking and writing. The mind is not an inner arena, it is an outward act.”10 Because behavior is a type of material state, Skinner’s claim is compatible with reductive materialism.
Though Armstrong takes a materialist stance about the nature of mind, he finds classical behaviorism unsatisfying because it ultimately denies the reality of our common experience of mental states independent of overt behavior: “One obvious difficulty is that it is our common experience that there can be mental processes going on although there is no behavior occurring that could possibly be treated as expressions of those processes.”11
We might label this objection the mind without behavior objection. I can have a conscious thought about eating an apple without actually eating an apple. We should not underestimate the intuitive power of the forgoing objection. Armstrong suggests that the mind without behavior objection is directly responsible for inspiring Ryle’s logical form of behaviorism.
Under Ryle’s hypothesis of logical behaviorism, a mind can possess a mental state, such as a disposition, without at the same having to manifest that disposition in the form of behavior. Thus a mind can have mental states independent of behavior. For example, Smith can possess a disposition of anger without acting angrily. But had you said one more insulting word to Smith, he would have burst out in anger. Whereas classical behaviorism sought to define the mental in terms of the behavioral, logical behaviorism provides room for talk about the mental in both dispositional and behavioral terms.
Armstrong points out, however, that the trouble for logical behaviorism is that it defines dispositions too weakly. A disposition, on the logical behaviorist’s account, is simply a liability or tendency that is manifested when a specific stimulus is present; a disposition, by Ryle’s lights, is not a state that causes behavior: “To possess a dispositional property is not to be in a particular state, or to undergo a particular change; it is to be bound or liable to be in a particular state, or to undergo a particular change, when a particular condition is realized.”12 For example, if an individual possesses the disposition to laugh, and the relevant external condition is brought about, then the individual will laugh. For example, if Cherry were to hear just one more joke about the Oompa Loompas’ song and dance, then she would erupt in laughter. One more joke was uttered and Cherry burst out in laughter. Any behavioral-hypothetical suggests that specific environmental conditions play the role of eliciting responses. Even though common sense might tempt one to think that Rylean advocates of logical behaviorism could assign causal power to mental dispositions, this would be a mistake because Ryle, much like Skinner, winds up denying the existence of mental states as the actual causes of behavior. Fodor’s concise observation is relevant here: “Logical behaviorism is just radical behaviorism in semantic form. . . . What does not really exist cannot cause anything, and the logical behaviorist, like the radical behaviorist, believes deep down that mental causes do not exist.”13
Having examined why Armstrong and others find both classical and logical species of behaviorism to be less than satisfying materialist accounts of mind, we are now positioned to consider Armstrong’s central state materialist approach to mind.
3. Armstrong’s Central State Materialism
In contrast to logical behaviorism’s weak notion of dispositions, Armstrong argues that dispositions are not mere liabilities understood in terms of a behavioral-hypothetical, but are instead conceived as the actual states that lie behind and cause behavior: “Perhaps mind can be defined not as behavior, but rather as the inner cause of certain ranges of behavior. Thought is not speech under suitable circumstances; rather it is something within the person that, in suitable circumstances, brings about speech.”14 Armstrong offers a theory of mental causation that agrees with our introspective intuitions about the mind in relation to behavior: minds (or mental states) are causally involved in producing behavior. For example, Stevo perceives the cobra and jumps back because he sees the cobra is striking. He jumps back because he believes that cobras are dangerous and furthermore desires to preserve his life. Stevo’s perceptions, beliefs, and desires—his mental states in general—are the actual causes of his jumping.
Armstrong recognizes that an analysis of dispositions as the inner causes of behavior does not exclusively warrant materialist conclusions. For if we suppose that our language is topic-neutral about the mental, any description of mind as the inner cause of behavior would be logically compatible with Cartesian and certain other non-reductive accounts of mind. In this way Armstrong acknowledges that dualism accords with common sense about mental causation as well. But Armstrong argues that if language is topic neutral about the nature of the mental, then dispositions, as the inner causes of behavior, are at least compatible with materialism; after all, central state materialism says that neural events occupy the appropriate causal role of mental states (dispositions), which are the sole causes of behavior. Armstrong then claims that since science favors materialism, mental states are probably identical to material states (i.e., internal causes of a material system) of the central nervous system: “the verdict of modern science seems to be that the sole cause of mind-betokening behavior in man and the higher animals is the physico-chemical workings of the central nervous system.”15 Armstrong reduces (or identifies) mental states to nothing but the cause of specific sorts of behavior. In order to clarify the concept of a disposition in a scientifically acceptable manner, Armstrong points out that, unlike the behaviorist who has denied that a disposition is a state internal to a system, modern science describes dispositions as actual states of systems (e.g., the weak molecular bonds of a vase make the vase brittle). In the case of brittle glass and elastic rubber, the modern scientist will say the following:
Faced with the phenomenon of breakage under relatively small impacts, or the phenomenon of stretching when a force is applied followed by contraction when the force is removed, he will assume that there is some current state of the glass or the rubber that is responsible for the characteristic behavior of samples of these two materials.16
In the case of brittle glass, brittleness is understood to be a specific molecular pattern possessed by the glass. It is a pattern of the glass itself; thus, when struck lightly, the molecular pattern is causally involved in the glass’s shattering. If dispositions are understood as actual states of material systems, Armstrong reasons that we can infer that dispositions are “actual causes, or causal factors . . . A certain molecular constitution of glass that constitutes its brittleness is actually responsible for the fact that, when the glass is struck, it breaks.”17 Analogously, mental dispositions function as the inner causes of behavior: “mind is nothing but that of an inner principle apt for bringing about certain sorts of behavior.”18 Because an explanation of mental dispositions can be shown to be consistent with science, Armstrong concludes that his central state materialist (CSM) theory of mind is scientifically plausible and superior to both logical behaviorism (LB) and Cartesian dualist theory (CDT). At this stage in the analysis, it might be useful to provide a clear summary of Armstrong’s argument:
1. A theory of mind is scientifically plausible only if mind is the cause of behavior. (The assumption is that scientific explanations are causal explanations.)
2. If LB is scientifically plausible, then mind is the cause of behavior.
3. LB contends that mind is not an internal state that causes behavior (but only a liability or tendency).
4. Therefore, LB is not scientifically plausible.
5. Mind is the cause of behavior on the CSM theory.
6. Mind is also the cause of behavior on the CDT. (The assumption is that our language about mind is topic-neutral and thus compatible with both CSM and CDT.)
7. The verdict of modern science favors CSM over CDT.19
8. Therefore, CSM is scientifically plausible (and superior to LB and CDT).
4. CSM and the Problem of Consciousness
Armstrong recognizes that the character of conscious experience is a serious problem for his own theory. Analyzing mind in terms of an internal material cause alone coheres with a third-person form of analysis but is nonetheless logically compatible with the absence of a first-person perspective:
Now can we say that to be conscious, to have experiences, is simply for something to go on within us apt for the causing of certain sorts of behavior? Such an account does not seem to do any justice to the phenomena. And so it seems that our account of the mind, like Behaviorism, will fail to do justice to the first-person case.20
A purely physico-chemical explanation can get along just fine without ever mentioning a first-person point of view. One rightly wonders how Armstrong can accommodate first- person phenomena within his purely third-person, reductive causal account of mind. Armstrong proposes that first-person consciousness is ultimately a higher order function of perception: consciousness refers to a “self-scanning mechanism in the central nervous system.”21 The idea is that some part (or process) of the central nervous system scans another part (or process).
Armstrong claims that the self-scanning mechanism is essentially a selective mechanism that enables an animal with perceptual capacities to distinguish between the features of objects of any given visual scene, such as the kind of selective behavior that would be required to distinguish between a green object and a red object: “We can think of the animal’s perception as a state within the animal apt . . . for selective behavior between the red-and green-lighted pathways.”22
5. Appraising Armstrong’s Hypothesis
Though Armstrong formulates a logical analysis of consciousness that is compatible with central state materialism, one could fill in the details with recent findings in neuroscience about the role that selective neuronal behavior plays in relation to distinguishing an object’s features when two or more objects comprise a visual scene. For example, if a visual scene were comprised of a red square (RS) and a blue triangle (BT), selective neuronal behavior would somehow have to assign features to the right objects. For if the neurons that fire in response to RS and BT were to fire at the same time, the brain might easily confuse the features of objects, such that the square would be seen as blue and the triangle as red. In order to explain how we consciously perceive the features of competing objects in a reliable way, there must be a higher order form of selective neuronal behavior that functions to assign features to the right objects. This notion of higher-order selective behavior is commonly thought to be carried out by the neuronal mechanisms of top-down selective attention and is discussed in a vast literature pool in the neurosciences.23
For our purposes, the critical question concerns whether consciousness is nothing but an activity of the neuronal mechanisms of top-down selective attention. Some of the experimental work of Anne Treisman and colleagues is useful here. In one experimental set-up, individuals were briefly shown two colored letters at the same time: a green T and a red O. The experimenters found that when top-down selective attention was prevented by means of a brief presentation of the letters, individuals would nonetheless experience illusory conjunctions: the individuals reported seeing “a red T when a green T and a red O” were presented at the same time.24 Consequently, red, rather than green, was bound to T and the result was an experience of illusory conjunctions. Several inferences can be drawn from this study. First, consciousness is not reducible to the mechanisms of top-down selective attention on grounds that conscious states can emerge without the deployment of the mechanisms of top-down selective attention. Second, binding features to form a unified object of consciousness (even if the feature-unified object in question does not reflect the correct features) can occur without the deployment of the neuronal mechanisms of top-down selective attention. Thus object-feature binding per se is not reducible to activities carried out by the mechanisms of top-down selective attention.
What functional role do the mechanisms of top-down selective attention perform? They perform the role of disambiguating the features of objects when competition arises in the cortical hierarchy.25 Thus what Armstrong proposes for an account of the metaphysics of consciousness is really an account of the epistemology of consciousness: the neuronal mechanisms of top-down selective attention explain how our cognitive systems function to assign features to the correct objects.
Fine, the materialist might say. Perhaps all we mean by consciousness is an ability to carry out some epistemic function, such as the ability to recognize an object. As David Lewis remarks, “knowing what it’s like is the possession of abilities: abilities to recognize, abilities to imagine, abilities to predict one’s behavior by means of imaginative experiments.”26 We might call this Lewis’ attempt to shore up the weaknesses in CSM. Lewis does after all characterize his own view regarding the nature of consciousness as consistent with, and an expression of, CSM.27
I think there is important empirical evidence to the contrary. For instance, recent neuropsychological evidence has shown that persons who undergo damage in the inferior temporal lobe (or IT) wind up with associative agnosia disorder, the inability to recognize objects. Even though associative agnostics can no longer recognize objects, they can nevertheless see them.28 The following is an example of an elderly man diagnosed with this type of agnosia:
A sixty-year old man . . . woke from a sleep unable to find his clothes, though they lay ready for him close by. As soon as his wife put the garments into his hands, he recognized them, dressed himself correctly, and went out. In the streets he found he could not recognize people—not even his own daughter. He could see things, but not tell what they were.29
Although he could not recognize objects, he could see them. This suggests that consciousness cannot be adequately explained in terms of certain abilities, such as the ability to recognize. Therefore, Lewis’s attempt to solve the problem of consciousness in terms of abilities is less than convincing. One might also say, in light of Chalmers’ distinction between hard versus easy problems, that the scope of Lewis’s purported solution to the problem of consciousness is, at best, consistent with a solution to one of the easy problems of consciousness (e.g., the problem of recognition); but no solution to an easy problem entails a solution to the hard problem. Hence, what Lewis (and by logical extension, Armstrong) sets out to explain and what he actually explains are not the same.
Finally, one could motivate a conceptually-based objection that would apply to any brand of reductive materialism through the conceivability of a philosophical zombie.30 But for those adopting an empirical approach to the problem of consciousness, a philosophical zombie might not be the most appealing concept. That does not mean that a zombie concept might not be useful to the empirical sciences in some sense. For example, what if we inverted the characteristics of a philosophical zombie? Let us call such a creature an inverse zombie. An inverse zombie would have none of the behavioral characteristics and responses of a philosophical zombie, but would nevertheless be conscious. Inverse zombies are not only conceivable; they actually exist: for example, individuals who experience “anesthesia awareness” fall into such a category. From an external observer perspective, these patients appear unconscious during general anesthesia. However, in 1 to 2 out of a 1000 cases, patients under general anesthesia may be aware of intraoperative events, and sometimes without any objective indices.31
The application of an inverse zombie to our current discussion could go as follows. Armstrong claims that “mind is nothing but that of an inner principle apt for bringing about certain sorts of behavior.”32 Now if Armstrong’s causal reductive hypothesis is feasible, then knocking out the mechanisms that are apt to cause certain sorts of behavior would involve knocking out the person’s conscious mind. However, in inverse zombie cases, the mechanisms that are apt to cause behavior are knocked out and yet the person is still consciousness.
One could provide further support to this empirically based counterexample by simply noting that persons who suffer from a severe form of locked-in syndrome (also known as total locked in syndrome) can no longer bring about certain sorts of behavior because the mechanisms that are apt for causing certain sorts of behavior are no longer apt for causing certain sorts of behavior (due to chronic motor deficits); yet persons who are locked-in their brains are still conscious of what is going on in the strongest and most literal sense.33