A Causal Argument for Physicalism

Good arguments have been given to show that mental states are not reducible to physical states of the brain (neural transmissions etc.); and intuitively they don’t seem like brain states. Thus, some degree of dualism obtains, of one sort or another. But no one ever argues that mental causation is, or must be, non-physical. It is commonly assumed that all the causal machinery of the mind resides in the brain—in the physical correlates of the mind. Three kinds of mental causation can be distinguished: input psychophysical, output psychophysical, and intra-psychological—for example, perception, action, and perception-cognition (seeing causing a belief, say). The picture we have is that the correlated brain states causally act in such a way that certain effects are triggered, mental or physical. We don’t need to invoke a special type of causation to explain mental causation. If Cartesian dualism were true, we would presumably need to, since the mind could have causal powers without being accompanied by a brain. But most theorists today accept that the brain is the causal powerhouse of the mind. Mental causation is identical to brain causation: we can have an identity theory of it. There is no such thing as irreducibly mental causation—though there may be irreducibly mental properties or states or events or stuff. The case is like certain other characteristics of the mind: location, duration. In so far as mental occurrences have a location, it surely coincides with the location of the brain; there is no other type of location that the mental might have (the big toe, somewhere ten miles from the brain). Evidently, mental occurrences have a date and a duration: the pain or thought occurred at a specific time on a certain day, or took up a certain amount of time (the pain went on for ten minutes, say). Surely, these temporal coordinates coincide with those of the correlated brain processes; they don’t operate autonomously. If you know the timing of the correlates, you know the timing of the mental occurrences. We entertain no notion of irreducibly mental time: mental time is physical time. Mental causation thus acts in the same way—it comes down to physical causation. The place, time, and causal power of a mental occurrence are nothing other than those of the correlated brain occurrence. That doesn’t mean that all the properties of the mental occurrence are so reducible, but those three are; to that extent at least physicalism holds sway. In particular, causal physicalism seems true and is generally accepted.[1]

But the resulting position is uncomfortable, because it is hard to see how a more far-reaching physicalism could fail to be true also. How could it be that mental causation is completely physical while the causative factors are not themselves completely physical? That would mean that the causes and effects have properties over and above their causal profile. How could a mental cause not be physical if its causal powers are completely physical? The only way out is to embrace epiphenomenalism, but we are assuming that mental causation is real. If mental causation is real and reduces to physical causation, how can the mental event not be physical? If biological causes in the body operate by virtue of physical causation, aren’t they thereby physical—constitutively so? And we know they are: they are composed of chemical and physical stuff. So, the same thing must be true of mental causes: they must be made of the stuff that constitutes their causal machinery. How can the mind not be the body if its causal machinery is completely bodily? It would be different if the mind could act causally without a brain, but we are ruling that out. Causal physicalism thus entails ontological physicalism. The correlation can’t be mere correlation. If it were, and mental causation is real, we would have to say that the causal machinery is not completely physical; but it is, so the relation has to be more than mere correlation. The nature of the mental state must reduce to its causal nature, which is physical. Mental causes can’t be non-physical if mental causation is physical.

We could conclude from this argument that classic physicalist reductionism must be true, despite the cogency of arguments against it, or we could revise our view of physical causation as it occurs in the brain. That is, we could propose that the cerebral causal machinery is not classically physical: it isn’t just electro-chemical transmissions, or they are not purely physical as we currently understand the physical. Here panpsychism, or some variant of it, beckons: hidden in the brain there are properties or processes that go beyond what we habitually attribute to the brain—and these figure in the causal machinery on which mental causation depends. The mind as we know will then be reducible to these unknown properties. This is not physicalism in any recognized sense (or even pansychism); it is rather a doctrine with no name or currently-intelligible content. It might be called “Whatever-ism”: the mind reduces to whatever is in the brain that goes beyond what we now know of the brain and functions as part of the machinery of mental causation. At any rate, we have a puzzle here with no easy solution—the puzzle of mental causation. How can the mind not be physical if its causal machinery is physical?[2]

[1] When a sensation of pain causes a withdrawal of the hand, say, the causal mechanism consists in the fact that the neural correlate of pain initiates a sequence of neural impulses that culminate in the contraction of certain muscles. It isn’t brute and immediate, or strangely immaterial; the causal process is straightforwardly material—the brain doing what brains do. The causal relation is a physical relation.

[2] I think this has always been a prime motivation for mind-brain identity theories, though I have not seen it stated explicitly (though ideas close to it have been). Once you discover that the brain does all the work of moving things around in the mind you wonder how the mind can be anything but the brain; the only question is the design specifications of the brain machinery. We certainly know a lot about these specifications, but do we know everything about them?

Errors of Philosophers

I asked myself if there are any errors philosophers regularly commit, through the centuries and still today. This would be useful information to have. I came up with three. The first is that they seldom if ever acknowledge that positions they oppose generally have things to be said in their favor; they behave as if only their own position is remotely credible. But this is almost never true: there are always reasons that support other positions and account for adherence to these positions. Philosophers need to cultivate an attitude of mind that combines endorsing one position with recognizing the force of alternative positions. You may reject materialism, say, but you should not be oblivious to the genuine reasons that speak in its favor. The same is true of views like emotivism, idealism, Platonism, instrumentalism, eliminativism, skepticism, etc. There are reasons to accept such positions, even if you find them ultimately implausible. Blind dogmatism is never helpful or rational (this is really a plea for intellectual honesty). Second, philosophers have tended to treat the human being as privileged—as if this species alone is the proper subject-matter of the discipline. That is anthropocentric and non-evolutionary. We exist contingently and other animals are proper subjects of investigation. If you are interested in knowledge, say, you should consider animal knowledge of various kinds—and the same for other mental categories. You should also consider hypothetical cases of alien species to test your theories. If there were actually existent species superior to us, we would do well to study the philosophy they produce as well as our own. Our human philosophy has been speciesist and the narrower for it. I hear the Vulcans have some pretty interesting things to say. Third, and connected, philosophers have not been mysterian enough: they have been unwilling to accept human intellectual limitation (though some of been aware of the point). This has led them to adopt implausible reductive and eliminative positions (I write about this in Problems in Philosophy). A healthier attitude recognizes that we might be allowing ourselves to embrace absurdity instead of accepting limitation. These three tendencies have shaped philosophy from its earliest days, and still shape it today. We do well to keep them in mind as we go about our business.

A New Metaphysics

I am about to propose a new metaphysics—a new metaphysical system. I say “new” with some trepidation, there being nothing new under the sun; but I am confident that nothing like this has ever been formulated before, or not in any tradition with which I am familiar—though there are faint echoes of it in Aristotle (it is quite anti-Platonic).[1] I will begin by stating the theory as baldly, indeed stridently, as possible, so that no one is in any doubt about the pretensions of the theory. Then we can move to more careful formulation and motivation. I call the theory “logical causalism”, though “causal logicism” would also do; this may be abbreviated to “causalism”, as “logical positivism” was abbreviated to “positivism”. There are two main claims. The first is that logical entailment is a type of causation: premises have the causal power to produce conclusions. Entailment is (causally) generative, productive, consequential—a type of force, active and effectual. Logical consequence is a type of causal consequence, not something altogether different from ordinary causal consequence (bridges collapsing etc.) The second claim is that causation is a logical relation: it has a logical structure, a logical form, a logical essence. It is logically constituted. It works by combining a major premise and a minor premise—the former being a law of nature, the latter a particular fact. For example: all matches cause fires when struck (in suitable conditions); this match was struck; therefore, this match caused a fire. The conclusion is deducible from the premises and may be inferred from them. The antecedent facts (general and particular) logically imply the consequent fact. It is as if the match reasons thus: “I am a match and I was struck; all struck matches cause fires; therefore, I will cause a fire”. It doesn’t really reason this way, of course, but the causal sequence consists of a pair of facts that entail the final state of that sequence. Causation is a kind of logical procedure: it leads from premises to a conclusion—from cause to effect. It is as if the world is engaging in logical reasoning. The process is logical because it embeds a logical necessity—given the antecedent facts a certain fact must result. There is a logic to causation; it isn’t just random or brute or “one damn thing after another”. It is intelligible, as much as modus ponens is. Putting these two claims together, we can say that logic is causal and causation is logical. Deduction (or induction) is based on a causal power—the power to produce conclusions; and causal reasoning is based on a logical structure—the structure of a general law and a particular fact combining to produce a certain result.

Now the metaphysical picture is as follows: the entire world (reality) consists of logic and causality. The two are essentially identical and they run through everything—the mental, physical, mathematical, and moral. The mental is subject to causation and hence the logic inherent in it; ditto the physical.[2] The mathematical concerns facts that have causal explanations: e.g., the number 4 is even because it is divisible by 2 and all numbers divisible by 2 are even. The latter two facts causally explain the former fact—they give rise to that fact; make it the fact it is. Mathematical facts have the power to produce other mathematical facts (a proof seeks to uncover these facts). The fact that a triangle has three sides is causally connected to the fact that it has three angles: the two facts lead to each other, generate each other, are the reason for each other. In the case of morality, we can say that according to utilitarianism an action is good if and only if it is caused to be so by maximizing utility: this is what explains or grounds the goodness of the act. The deontologist will give a different explanation of what makes an act good (it’s good because it falls under moral rules). Causation extends beyond what we have come to call efficient causation modelled on a limited class of cases, as when a sculptor shapes a statue. Causation tracks reasons-why; the reason why something is the case is properly described as the cause of it (see below). The world does not divide into a causal part and non-causal part (as Aristotle recognized); it is shot through with causality in every nook and cranny. Everything has a reason for being—and that includes logic itself. The cause of the validity of an argument is that it instantiates certain logical laws; these laws cause (explain, ground) the validity of the argument. The only uncaused things are things that exist without reason—the basic facts of the universe. It isn’t that logic only exists in the realm of language and thought, and causation only exists in the changing world of matter and mind; logic and causation jointly penetrate everywhere. Thus, logical causalism is a monistic metaphysics: all reality is unified around its two pillars. Existence is coterminous with logic and causation—existence is logico-causal. The only things that escape them are non-existent: the nonexistent has no causal power and it need not obey any logic (it can be nonsensical). Every real existent thing has logic and causality woven into it. We thus have mental causation, physical causation, mathematical causation, and moral causation—as well as logic of these types. Everything obeys the laws of logic, and everything is subject to causation. We may not understand causation that well (see Hume), but we know it is everywhere. And logic and causation work together, not being separable in their essence.

I mentioned Aristotle: he too had a very expansive conception of causation. His four types of cause reach into everything—the mental, the physical, the mathematical, the moral. But I am not aware that he ever thought of the syllogism as causally grounded: he didn’t add logical causation to his list of causes. Still, I think he was right to use the concept of cause in a highly inclusive manner. We need a concept of cause (causal explanation) commensurate with our concept of a reason, and that concept is everywhere. The natural (metaphysical) kind Cause is as broad as reality itself; it is as broad as the concept of an object or a property or the concept of a concept. It enables us to unify all the heterogeneity of the universe. We actually need it to unify Aristotle’s four causes: what do material causes, formal causes, final causes, and efficient cause all have in common? They are all causes, of course—productive, generative, reason-providing. And what ties them all together instead of leaving them in a jumble? Logic, naturally: it is only logical for nature to harmonize the four causes into a coherent package. You can’t have the final cause of a table being to hold things up and the material cause being composed of air or fire. Something has to explain the harmony of the four causes, and that will be a logical cause. Aristotle could have chosen to regard logical consequence as a species of causation, but evidently he didn’t (he seems rather Platonic about logic). In any case, such a view is not part of the tradition stemming from Aristotle. The obsession with so-called efficient causation (the label itself is hardly pellucid—aren’t all his causes equally efficient?) in post-Hume philosophy precluded recognition of a broader notion of causation. In its classical meaning it referred to causation by a separate agent on the body in question, as in a sculptor shaping a statue, which clearly covers only a subclass of cases—it says nothing about the intrinsic causal power of a tree to grow, for example. Nor does ordinary language provide any warrant for enforcing such a narrow conception of causation upon us. And once we widen the concept in the natural direction, we can see our way clear to accepting a causation-based metaphysics. Logic has traditionally been supposed to apply universally, but has not been supposed to be causal itself; I am adding that wrinkle. First, we must pull logic away from language and mind, placing it in the objective world; then, we must subsume it under a causal umbrella, thus producing a monistic logico-causal metaphysics. The world is all that is the case logically and causally.[3]

It is fair to say that causation used to be under a cloud in analytical philosophy. Hume was thought (wrongly) to have undermined the coherence of the concept, and logical positivism fought shy of it for lack of verificationist approval. But that changed sometime in the 1960s when causal theories became all the rage. All well and good, but efficient causation was the preferred causal concept, though not much scrutinized (is all causation really from other bodies?). I am recommending casting off the suspicion more radically (along with the empiricism that leads to it): causality is ubiquitous, protean, and indispensable. Logic, too, has cast off earlier inhibitions, allowing in the modal, epistemic, practical, deontic, etc. It has become a lot less anal. True, causation is somewhat mysterious, especially to the empiricist (anal) mind, but that shouldn’t stop us from using it to bring order to an otherwise chaotic universe. Mysterian logical causalism is not to be ruled out of court a priori. Indeed, there is nothing to stop us being naturalist mysterian logical causalists.[4]

[1] For some background, see my “Causal and Logical Relations”.

[2] Classical mechanics has a logical flavor to it, being largely a priori; and psychology (folk and scientific) is inextricably bound up with logic because the role of rationality in mental life.

[3] That would include God, if he were to exist—he is clearly the logical and causal being par excellence.

[4] Seriously, did anyone ever truly believe that an all-encompassing metaphysics would be free of mysterious elements? Do you think the ultimate truth about the whole universe would be transparently given to the human mind (brain) at this precise moment of evolutionary history? Is Berkely’s metaphysics mystery-free, or Leibniz’s, or Descartes’? Is a totally physics-based metaphysics devoid of mystery? Hardly.

Consciousness and the Fetus

The brain grows gradually in the fetus, going through various stages of development. This development is now well mapped out embryologically. But nothing is known about the inner consciousness that accompanies this brain growth. It is reasonable to believe that it goes from simple to complex, primitive to sophisticated—how could it not? It has often been supposed, on good evidence, that the fetus itself passes through stages resembling its prehistoric ancestors—the fish is often mentioned. Could it then be that the ontogeny of consciousness recapitulates the phylogeny of consciousness? Could the earliest stages of fetal brain development resemble the earliest stages of ancestral brain development? Might the consciousness of the fetus in its early life be like the consciousness of the first organism on earth to develop consciousness? We don’t know what this consciousness was like, or what the fetus’s early consciousness is like; but it is not unreasonable to surmise some homology (perhaps a feeling of comfort or discomfort). Then, as consciousness develops in the fetus, there is a mapping onto the evolutionary development of consciousness in ancestral species, leading up to the consciousness of species close to the human. The old forms of consciousness are preserved in the genes and have their moment in the sun as the fetus develops. Thus, the fetus lives through millions of years of phenomenological evolution, getting a taste of what things used to be like long ago. All this is forgotten as the adult human brain takes shape, but it was there nonetheless. The fetus in effect knows the consciousness of our earliest ancestors. Yet this knowledge is not preserved, memory being what it is. If only it could be recorded and then uploaded into the adult brain! We would learn a lot about the evolution of consciousness from the inside. The fetus knows more about the evolution of consciousness than we adults will ever know.[1]

[1] The human fetus doesn’t know what it is like to be a bat, since echolocation was not a feature of our ancestral line, but it might know what it is like to be a fish or lizard.