A Cybernetic Defense of Type Physicalism

Abstract
In this paper, I examine the tenability of type physicalism within the context of a second-order cybernetic analysis of phenomenality. I begin by describing the philosophical problem type physicalism attempts to resolve and follow up with an examination of arguments against type physicalism. I then describe how arguments against type physicalism tend to rely on the ontological distinction between system and observer. Next, I show that this distinction is purely conceptual and dissolves when phenomenality is analyzed from a second-order cybernetic perspective. Within this context, type physicalism remains a tenable solution to the mind-body problem so long as an isomorphic mapping between physical and psychological processes is possible.

Introduction
The motivation for type physicalism stems from empirical evidence of pervasive and systematic psychoneural correlations, that is, correlations between mental phenomena and brain processes. These correlations are systematic enough to allow scientists to successfully sense, transmit, analyze, and apply the language of neurons using an assortment of sophisticated imaging techniques and brain-computer interfaces. For instance, in 2003, Dr. Miguel Nicolelis, associate professor of neurobiology at the Duke University Medical Center, used a brain-computer interface system to successfully filter and utilize motor command impulses from the electrical activity of a primate brain to operate a robotic arm in real time. Foxborough-based Cyberkinetics Neurotechnology Systems, Inc., founded by John Donoghue, followed up this achievement by developing similar implantable brain-computer interface that proved capable of translating human thought into controlled cursor and even robotic arm movements. The Wadsworth Center in Albany, NY, run by Dr. Jonathan Wolpaw, has developed a less-invasive EEG cap which transmits and decodes electrical waves detected on the surface of the skull to accomplish similar feats. Clearly, we are beginning to better understand psychoneural correlations and the above examples support the notion that these correlations are in fact law-like (insofar as they occur in predictable patterns).
It is possible that scientists will one day discover that each type of mental event has a neural correlate that is necessary and sufficient for its occurrence. For example they may discover that the type of mental event we classify as seeing red requires the existence of a specific type of physical event, and that the existence of this specific type of physical event entails the type of mental event we classify as seeing red. Of course, the fact that we might one day discover that this correlation were true does not imply that the relationship between mental phenomena and physical processes has been explained. Even if true, the relationship between the two is still in question because this thesis is compatible with various dualist and monist solutions to the mind-body problem, including but not limited to Descartes’ causal interactionism, Leibniz’s pre-established harmony, Malbranche’s occasionalism, Spinoza’s double-aspect theory, epiphenomenalism, and emergentism. Undeniably, empirical evidence supporting the correlation thesis offers no solution to the ontological mystery.

A Case for Type Physicalism
Our scientific world-view makes it difficult to avoid looking for a causal explanation for psychoneural correlations. At the level of human experience, causal explanations describe the mechanism by which one event leads to a consequent event. In the case of psychoneural correlations, we either assume that there is a mechanism by which physical process P gives rise to mental event M or admit that the correlation is a brute, unexplainable fact. In either case, the sense of mystery remains. If we assume that a mechanism explaining psychoneural correlations exists, knowledge of this mechanism would do little to remove the sense of mystery. For if we discovered that P gives rise to M through mechanism T, we would then wonder how T gives rise to M. If on the other hand, we admit that there is no mechanism at play, and that the psychoneural laws are fundamental laws of nature, our curiosity is left unsatisfied.
Type physicalism attempts to dismiss the issue altogether by declaring that types of states of phenomenal consciousness are identical with types of brain states, for example, that the conscious state, seeing red, is identical to a specific neural process or neuronal activity within the brain (describe actual physical process). A more intuitive way of stating this is that we are capable of conceptualizing a single event type in at least two different ways: as a physical event and as a mental event. By replacing mechanism with identity, type physicalists avoid having to explain, as Huxley put it, how anything as remarkable as a state of consciousness comes about as a result of irritating nervous tissue.
This claim remains contentious however because there is no a priori justification for accepting that the correlation between mental and physical states is one of identity. As noted above, the correlation thesis is compatible with various forms of ontological dualism and idealism. Proponents of type physicalism hold that while the concept of mental events may be distinct from any physical concept, we can empirically establish that these concepts refer to the same natural entities. Type physicalism is therefore an empirical claim that assumes a physical ontology. As an empirical claim, an a priori justification seems out of the question. So, the type physicalist can do little more than show that type physicalism is tenable and provide convincing justification for the acceptance of the identity theory over other plausible solutions to the mind-body problem.
In Is Consciousness a Brain Process, Ullin Place was the first to characterize as a reasonable working hypothesis the idea that mental phenomena are really physical processes. In other words, he contends that the identity theory is not logically doomed; it is not necessarily false. Place suggests the misconception that the identity thesis can be disproved by logical arguments alone derives from the ambiguity of the term is in its many different uses. He recognizes that the term is serves three primary functions—to predicate, to define, and to indicate a relationship of compositional identity. For Place, the presumption that the logical independence of expression entails the ontological independence of the entities described by those expressions originates from the failure to distinguish between the is of definition and the is of composition. In other words, if one cannot differentiate between the function served by the term is in the statement, red is a color, and the statement, his table is an old packing-case, one will not be able to see the logical fallacy in assuming that statements or expressions with unrelated meaning are not capable of providing sufficient characterization of the same object or state of affairs. Such a person will be convinced that it is logically invalid to claim that consciousness is a brain process since, for example, it is not self-contradictory to suppose that someone feels pain when nothing is going on inside that person’s head. What this person fails to realize, according to Place, is that we are dealing with the case of compositional identity in this example, not one of definitional identity. In other words, the statement consciousness is a brain process is not true by definition, but must instead be verified by observation. In this sense, we are dealing with a case in which it is possible that two logically independent expressions, consciousness and brain processes, describe the same object or state of affairs in the same way that his table and an old packing-case do for the statement, his table is an old packing-case.
Once Place is able to show that the working hypothesis is not logically doomed, he is able to tackle the real issue; if it is logically possible, what evidential support is there to convince us that it is probable? Place points out that we must still set criteria for determining when two sets of observations are observations of the same object or event in order to recognize adequate evidential support for the identity thesis. In his words,
“we treat the two sets of observations as observations of the same event in those cases where the technical scientific observations set in the context of the appropriate body of scientific theory provide an immediate explanation of the observations made by the man in the street” (Place, 1956, p.48).
For Place, it follows “that in order to establish the identity of consciousness, and certain processes in the brain, it would be necessary to show that the introspective observations reported by the subject can be accounted for in terms of processes which are known to have occurred in his brain” (Place, 1956, p.48)
We see here that Place is very confident that we can account for subjective introspective phenomenal reports in scientific terms. However, such an account seems impossible to construct. While in some sense it is possible to account for the apparent property of water in terms of the underlying chemical nature of H2O, and perhaps similarly possible to account for lightning in terms of the underlying motion of electrical charges, it has yet to be seen how one would account for the phenomenal in terms of any underlying brain process.
Place accounts for the assumed difficulty in providing a physiological explanation of introspective observations as the product of what he coins the phenomenological fallacy—the mistaken conjecture that descriptions of introspective observations are primarily descriptions of phenomenal properties within a phenomenal field and only “secondarily, indirectly, and inferentially descriptions of the objects and events in our environment” as proposed by Kant.
We are perfectly justified in maintaining that only what is within ourselves can be immediately and directly perceived, and that only my own existence can be the object of a mere perception. Thus the existence of a real object outside me can never be given immediately and directly in perception, but can only be added in thought to the perception, which is a modification of the internal sense, and thus inferred as its external cause ... . In the true sense of the word, therefore, I can never perceive external things, but I can only infer their existence from my own internal perception, regarding the perception as an effect of something external that must be the proximate cause ... . It must not be supposed, therefore, that an idealist is someone who denies the existence of external objects of the senses; all he does is to deny that they are known by immediate and direct perception ... . (Kant, A367 f).
In this passage, Kant asserts that the existence of external things is inferred by our internal perception of them. Place argues that the reverse is actually the case, that we infer phenomenal properties from real properties. I will return to discussing the phenomenological fallacy later, because it plays a critical part in the cybernetic analysis of type physicalism. Before I return, I will introduce some of the traditional objections to the identity theory in general, and to type physicalism specifically.

Arguments against Type Physicalism
The main objections to the identity theory in general involve epistemic arguments: the conceivability, explanatory, and knowledge arguments. These arguments all reject physicalism on the premise that there is an epistemic gap between physical and phenomenal truths. While some proponents of these arguments contend that an epistemicgap implies an ontological gap, others admit otherwise. This oversimplification, of course, doesn’t do the individual arguments justice.
The conceivability argument basically states that since it is possible to imagine a world which is physically identical to ours yet lacks phenomenality, it is then metaphysically possible and therefore false that phenomena can be identical to physical processes. The premise that identity cannot be a relation which can hold contingently between objects (Kripke, 1972, p.x) is critical to this argument. As Saul Kripke points out, necessity can be established empirically for rigid designators, which appear to be only contingently related, by explaining away the apparent contingency. However, he goes on to argue that this is not possible in the case of psychoneural identities and concludes that the relationship between phenomenal and physical states cannot be one of identity. To arrive at this conclusion, Kripke relies on the premise that in the case of psychoneural identity statements, there is no confusing the phenomenal property with some different property which is only contingently connected with the correlated physical property, as is possible with identity statements of the theoretical variety. Rather, it is the phenomenal property itself that guides us in recognizing instances of the phenomenal type. Kripke therefore discredits the analogy between scientific identity statements, like heat is molecular kinetic energy, and psychoneural identity statements, such as pain is C- fiber stimulation, by showing that only in the former can we remove the illusion of contingency to reveal a necessary a posteriori relationship. For Kripke, the metaphysical possibility of the ontological distinction between the phenomenal and physical implies the possibility of an ontological gap that cannot be explained away.
In reaction to the intuitive leap (felt contingency) involved in the metaphysical force of Kripke’s conceivability argument, Levine proposed that even if we deny the apparent contingency of psychoneural statements, an explanatory gap still exists, shedding doubt on the physicalist position. Basically, Levine points out that an explanatory gap exists in any physicalist account of phenomena since (1) physical explanations of phenomenal concepts are limited to causal or functional descriptions and (2) there is more to our phenomenal concepts than their causal or functional roles, namely, their qualitative character; how it feels. According to Levine, this explanatory gap is exactly what makes psychoneural identity statements appear contingent and therefore vulnerable to Kripke’s modal arguments.
Related to the explanatory argument is the knowledge argument, which basically takes the same form as the explanatory argument in concluding that physicalism is epistemically false. The classic example of the knowledge argument was provided by Frank Jackson. To summarize his argument, a color blind person could know all the physical truths about the world, and in particular, all the physical information about the brain and the neurophysiology of vision, and still not know what it is like to see red (Jackson 1982, 1986). According to a similar example, someone could know all the physical truths about bats and still not know what it is like to be a bat (Nagel 1974).
All of the epistemic arguments against the identity theory summarized above depend on the notion of the qualitative character of subjective experience, or qualia. The inability to adequately account for qualia is an argument against physicalism in general and therefore also applies to the identity theory. These arguments would be easily dismissed by Place by appealing to the phenomenological fallacy. However, even if identity theorists are able to account for qualia and subsequently close the epistemic gap, they will still be left defending themselves against what many philosophers consider the most damaging objections to the identity theory: the multiple realizability argument.
In The Nature of Mental States, (1967), Hilary Putnam contends that functionalism provides a better solution to the mind-body problem than the identity theory by suggesting that it is more likely that functional states can be invariantly correlated with psychological states than physical-chemical states. Putnam’s contention is based on his belief that neuroscientists will have no problem finding at least one psychological state that can be applied to two separate organisms with distinctly different physical-chemical correlates. He mentions hunger as such an example. Clearly hunger is experienced by humans and octopuses alike, yet a common physical-chemical correlate is unlikely to be found between us and octopuses. For type-physicalism to be true, all organisms capable of experiencing hunger must share an identical, nomologically- correlated physical-chemical state. Since this is likely not the case, but rather that psychological states are multiply realizable, Putnam takes type-physicalism to be false.
Jerry Fodor strengthened the multiply realizability argument by arguing that the multiple realization of a psychological state applies not only “across physical structure- types” but to individual systems over time as well. In other words, he recognized that it was possible for the same organism to realize the same mental state type at different times in different physical structure-types.
Convergent evolution, corticalization, and the plasticity of the brain appear to provide enough empirical evidence to justify the belief that psychological states are multiply realizable. In the case of convergent evolution, the fact that unrelated lineages are capable of acquiring the same biological traits shows that it is possible for two lineages with dissimilar physical-chemical make-up to experience the same mental states. In the case of corticalization, the fact that brain functions can migrate from the subcortical centers of the brain to the cortex suggests that different physical-chemical structures are capable of realizing the same type of mental state. The possibility of brain plasticity strengthens this hypothesis since it suggests the instantiation of the same mental state type within different neuronal structures.

Title this section
Jaegwon Kim attempted to save type-physicalism from the multiple realizability argument by positing that the thesis fails to refute psychophysical reductionism or type- physicalism. He calls attention to the fact that Putnum’s conclusion that type-physicalism is false is based on the presupposition that reductive theories of mind imply the Correlation Thesis:
For each psychological kind M there is a unique physical (presumably, neurobiological) kind P that is nomologically coextensive with it (i.e., as a matter of law, any system instantiates M at t iff that system instantiates P at t).
To account for the possibility that many biological species, and perhaps non-biological systems, can realize the same mental property type, and for the possibility that the same organism may realize the same mental property type at different times in different physical structure-types, Kim adopts a less naïve formulation of the thesis, which he calls “the Structure-Restricted Correlation Thesis”:
If anything has mental property M at time t, there is some physical structure type T and physical property P such that it is a system of type T at t and has P at t, and it holds as a matter of law that all systems of type T have M at a time just in case they have P at the time.
Under this thesis, rather than identifying a particular mental state M with a particular physical state P, mental state M would be considered a disjunction of all its realizers: We would identify pain, for example, as neural correlate Nh at time t in humans, Nr at time t in reptiles, Nm at time t in Martians, etc.
If Kim’s Structure-Restricted Correlation Thesis amounts to an acceptable form of type-physicalism that successfully addresses the multiple realizability argument, and Place’s phenomenological fallacy argument is a sound objection to any of the empirical arguments, then type-physicalism appears to be a reasonably defensible thesis. Although Kim’s thesis possibly allows for an infinite number of disjuncts, there is no inherent problem with his particular take on type physicalism. Place’s phenomenological fallacy argument, on the other hand, seems to contradict the thesis which it is intended to defend.
If it is true that (1) we actually infer phenomenological properties from real properties, and (2) real properties are the “actual physical properties of the concrete physical objects”, then the thesis that consciousness is a process in the brain, which type-physicalism is intended to defend, appears to be false. These premises combined imply that the experienced properties, which are supposed to be type-identical to brain processes, do not exist within the brain. If brain processes exist within the brain and experienced properties do not, it is difficult to see how they can be type-identical. In other words, the experienced property red cannot be type-identical to the neurological processes nomologically correlated with seeing red because the experienced property is an input into the neurological process as opposed to being the process itself.
The above argument depends on the premise that properties of objects exist within the objects themselves and not within the brains of observers. Place seems to accept this premise when he states that
“We describe our conscious experience … by reference to the actual physical properties of the concrete physical objects, events, and processes which normally… give rise to the sort of conscious experience which we are trying to describe.

In this passage, Place explicitly admits that real objects posses actual physical properties, which cause conscious experiences. Although Place admits that objects themselves have properties independent of the observer, he argues that such properties cannot be similarly ascribed to experiences. It would be a mistake for example to ascribe the property red to the mental event seeing red. Place would likely counter the above argument by stating that the mental event seeing red is what is supposed to be type-identical to the neurological process, not the experienced property, which resides in the concrete physical object.

Like the phenomenological fallacy, second-order cybernetics challenges the traditional theories of perception which are based on the assumption that sense data are captured by the sense organs and converted into a synthesized internal representation of the external world. Whereas the phenomenological fallacy denies the existence of phenomenological properties, a second-order examination of perception instead denies the possibility of experiencing real properties and demands the existence of phenomenological properties, equating them to “subjective dynamic constructions that, by complex feedback paths within the observer and environment, move the system towards its emergent goals” (Bishop and Nasuto, 2005).

Generally speaking, second-order cybernetics is the cybernetics of cybernetics. Cybernetics is the study of the communication and control processes of regulated, closed- loop systems as observed by humans. Its primary aim is to elucidate how such systems are able to maintain their apparent goals in the face of various environmental disturbances. Like traditional theories of perception, cybernetics traditionally demands a distinction between the system studied and its environment. Second-order cybernetics, on the other hand, blurs this distinction by recognizing that the act of observation, which makes this conceptual distinction possible, plays a critical part in constructing the environment. When taking this into account, any system observed by humans, whether mental or physical, is understood as being a model of reality as opposed to reality itself.
From a second-order cybernetic perspective, properties are both phenomenological and real in the sense that they are both cause and effect of the dynamic constructive process of observing; properties are both object and representation of object. This causal-loop relationship can be represented symbolically as yt+1 = f(yt) where f is the act of observing, yt is the subjectively constructed representation of environment y at time t, and yt+1 is the subjectively constructed representation of environment y immediately following the act of observing y at time t.
Since cybernetics is concerned with those properties of systems that are independent of their concrete material or components, it can be used to describe both psychological and physical systems and to look for isomorphisms between them. If scientists can discover an isomorphic relationship between psychological and physical processes, there is good reason to suggest that they share an identical external structure and are simplified representations of the same external properties.
Sensations are not reactions to stimuli but rather a reaction to the changes in stimuli. Bibliography
Hill, C.S. (1991). Sensations: A Defense of Type Materialism. Cambridge: Cambridge University Press.
Kripke, S. (1980). Meaning and Necessity. Cambridge, MA: Harvard University Press.
Loar, B. (1990). ‘Phenomenal States’. Philosophical Perspectives 4, 81-108.
McLaughlin, B. (2003). ‘Color, Consciousness, and Color Consciousness.’ In. Smith, Q. and Jokic, A. (eds.), Consciousness. Oxford: Clarendon Press, 97-154.
Papineau, D. (2002). Thinking about Consciousness. Oxford: Clarendon Press.
Polger, T. (2004). Natural Minds. Cambridge, MA: MIT Press.
Putnam, H. (1973). ‘Psychological Predicates’. In Capitan, W. H. and Merrill, D. D. (eds.), Art, Mind, and Religion. Pittsburgh: University of Pittsburgh Press, 37-48.
Smart, J.J.C. (1959). ‘Sensations and Brain Processes’. The Philosophical Review, 68, 141-56.
Crooks, M. (2002). Intertheoretic Identificaiton and Mind-Brain Reductionism. Journal of Mind and Behavior, 23, 193-222.
Brown, R. (2006). What is a Brain State? Philosophical Psychology, 19(6), 729-742.
Kim, J. (2006). Philosophy of Mind. Boulder, CO: Westview Press.