Panpsychism – is it testable?
By Paul So
In recent years (late 90’s to the 21st century), Panpsychism has been enjoying some resurgence in philosophy. David Chalmers, Galen Strawson, Thomas Nagel, and others are seriously considering Panpsychism as a viable solution for the Mind-Body problem. But philosophers aren’t alone. In the natural sciences, some prominent scientists, even the well-known Christof Koch, are seriously considering Panpsychism too. This recent resurgence might give outsiders the impression that Panpsychism is becoming popular. However, the truth is it remains controversial because the majority of philosophers and scientists still find Panpsychism highly implausible.
What is it about Panpsychism that many people find implausible? In order to understand people’s incredulity, we need to know what Panpsychism is. I’m not going to provide a strict definition of Panpsychism. Instead, in the context of contemporary Philosophy of Mind, I will spell out what virtually all contemporary Panpsychists have in common. Panpsychists tend to believe that the microentities (e.g. elementary particles) that inhabit the fundamental level of the universe are conscious. Consequently, everything possesses some degree of consciousness. This is what many people are incredulous about: how could fundamental particles have consciousness?
To be clear, proponents of panpsychism aren’t arguing that matter has the same kind of mind/conscious states that we do. Rather, proponents or sympathizers argue that at the very least matter instantiates a very primitive conscious experience. When bits of matter combine together in an appropriate arrangement, they will produce a more complex and enriched form of consciousness like ours. There is a rationale or strategy behind this view that can only be understood in the context of the Explanatory Gap.
We have a story for how we get from the most fundamental matter to the most complex organism. When we go from an atom to a molecule to a neuron to a neural network, there is a smooth or continuous transition that connects them together. Once we arrive at a bunch of firing neurons, we want to find out a continuous transition from neurons to consciousness. But there seems to be a gap between firing neurons and consciousness; we simply don’t know how we got from a bunch of neurons to consciousness. We do know how we get from atoms to molecules to cells, but how did we get from neurons to consciousness?
Supposedly, this is a problem for Materialism, the view that consciousness is exclusively a physical phenomenon. This view can be spelled out in many different ways (e.g. type identity theory, token-identity theory, functionalism, etc), but they all agree that consciousness is not a non-physical substance or property floating above an underlying physiological process. One way to demonstrate that consciousness is a physical phenomenon is to reduce it to an underlying and familiar physical process. Some people who endorse the identity theory would like to show that consciousness is identical to some range of neural states. Others who endorse functionalism would like to show that consciousness is identical to, or realized by, some causal state that can be occupied by anything (at least in principle) from carbon-based neurons to its silicon analog.
Unfortunately, the aforementioned attempts are far from successful. One argument against these views tries to show that if consciousness is physical, it is logically impossible to have a world in which everything is physically identical to our own except its inhabitants lack consciousness. But such a possible world is conceivable. If it is conceivable, it is logically possible. Therefore, Materialism has to be false. This argument is known as the Philosophical Zombie argument and it is one of the popular arguments against Materialism.
Dualists are convinced that arguments like the Philosophical Zombie show that consciousness is irreducible. Some dualists think consciousness is an emergent property that is anchored to the physical world, but stubbornly retains its subjective quality. Others are convinced that consciousness is a substance distinct from the body. Panpsychists agree with Dualists that the Materialist strategy isn’t working, but they do think that consciousness is reducible. Unlike Materialists, Panpsychists think consciousness is reducible to more primitive or fundamental conscious entities. For Panpsychists, fundamental particles are conscious insofar as they possess primitive conscious experience. When they combine in the right manner to constitute a complex structure, eventually they’ll produce or constitute an enriching and complex conscious experience.
So far the Panpsychist strategy makes sense. It is understandable why there is at least some appeal to Panpsychism, but there are some deep problems for it. For one, just as there is an explanatory gap for Materialism, there is also an explanatory gap for Panpsychism. How did we get from very primitive conscious experience to some very enriching and complex conscious experience? Combining primitive conscious experience together means more primitive conscious experience, but how does that amount to a quality of conscious experience that you don’t find in particles? This is known as the Combination Problem.
But I think there is another pressing issue for Panpsychism. Even if the Combination Problem is solved, this other question remains: Is Panpsychism testable in principle? I think that it is not. Allow me to illustrate with a thought experiment. Consider the following two scenarios. One scenario where our universe’s fundamental matter is non-conscious, but in another scenario fundamental matter is conscious. However, in both scenarios, everything else is the same. Universes from both scenarios have the same laws of nature, spatiotemporal structure, forces. causal history, entities, and other physical facts. Consequently, scientists in both scenarios have the same scientific models making the same predictions. Moreover, scientists from both scenarios believe their universe is physical. If Panpsychism is true, but everything else we know about the physical world remains intact, how can we know if it is true?
It is not enough that panpsychists postulate a primitive conscious experience in order to find a way to reduce consciousness. Such a postulation needs to imply some observable fact or prediction that would make the universe look different from a view of the universe where its fundamental matter is non-conscious. If fundamental particles are conscious, but they still behave as we already expect them to behave (e.g. particles still spin up or spin down) based on previously established experiments, we have no way of knowing whether it is conscious. Consider another thought experiment: suppose that in a possible world almost every elementary particle is non-conscious except a handful of them. In such a possible world, all elementary particles exhibit the same kind of behaviors such as spins. How can physicists differentiate them from each other if (1) they exhibit the same behaviors according to laws of nature (2) their conscious experience is private?
In order for Panpsychism to be testable, it needs to imply some observable consequence that would be noticeable for scientists. Moreover, it would have to be something that any materialist model of the universe wouldn’t anticipate. But it isn’t obvious what that observable consequence should look like. Furthermore, in principle, the total set of observable data wouldn’t suggest any reason to postulate a conscious experience at the fundamental level of the universe. While there might be good metaphysical reasons to consider Panpsychism, there isn’t a good scientific reason to postulate it to begin with.
A panpsychist could point out that Materialism is also not testable. So far, I argued that Panpsychism is untestable for three reasons: (1) it doesn’t imply any observable consequence unanticipated by a materialist. (2) a panpsychic universe would look indistinguishable from a physical universe in terms of their laws of nature, spatiotemporal structure, number of elementary particles, causal history, and other scientific facts. (3) scientists in both possible worlds wouldn’t be able to tell the difference. But (1), (2) and (3) also apply to Materialism. After all, (1) materialism needs to imply some observable consequence unanticipated by panpsychist, (2) a materialist universe would look indistinguishable from a panpsychic universe, and (3) scientists in a materialist universe aren’t in any better position than scientists in a panpsychic universe to determine whether or not their universe is either physical or conscious. Therefore, my argument for the untestability of Panpsychism also applies to Materialism.
There is at least one problem with the panpsychist response. Whereas Materialism is relatively conservative insofar as it sticks to known properties already described by physics (among other natural sciences), Panpsychism postulates an extra unknown property: a primitive conscious experience. If one is a materialist, one believes in every property already described or attributed by physics from spins to electric charges. But a panpsychist wants to postulate an extra property unknown to physics. There seems to be a special burden of proof placed on someone who wants to postulate an extra novel property not accepted by science.
Let me elaborate on the special burden of proof. Normally, we think that when we make any claim, we ought to support it. So if we postulate an extra property or entity that isn’t part of our scientific view of the world (at least not yet), we want evidence for it. Think of it like an exclusive membership club. Suppose a potential member wants to join your exclusive club and you tell that person that one has to meet certain conditions to be an exclusive member. In other words, a potential member has a burden to meet certain conditions before he or she can be a part of that club. It would be awkward and absurd if such a potential member insists that you have the burden to show why one shouldn’t be part of the club. Likewise, anyone who postulates an entity or property and desires that it be incorporated into the scientific view would have to satisfy certain conditions for incorporation.
Somebody might point out that this would rule out many metaphysical positions from mathematical platonism to realists about properties. However, this depends on what the conditions are for being incorporated into a scientific worldview. Empirical evidence is one of the conditions, but another condition could arguably be indispensability for doing science (check out indispensability arguments in Philosophy of Mathematics). Either way, there is no empirical evidence for panpsychism and it isn’t obvious that panpsychism is indispensable for doing science.
I think one way to state the problem with Panpsychism is that a (possible) panpsychic universe is observably indistinguishable from a materialist universe because conscious experience is private. Since they are observably indistinguishable due to the privacy of conscious experience, there’s no further empirical fact of the matter that would decisively determine which view is correct. If positing an extra property doesn’t make a new picture of the universe empirically distinguishable from the current picture, it is untestable in principle.
Is this really a problem for Panpsychism? Someone might point out that Panpsychism is a metaphysical claim rather than a scientific claim. Any attempt to show that Panpsychism is untestable in principle doesn’t undermine the original motive behind Panpsychism as a metaphysical claim. After all, Idealism and Platonism are metaphysical worldviews that aren’t subject to scientific scrutiny. Likewise, perhaps Panpsychism is a metaphysical worldview that cannot be tested by science.
There are several problems with this view. First, it isn’t obvious that a claim must either be a scientific claim or a metaphysical claim. For example, the claim that there are natural kinds is a metaphysical claim, but it bears so much relevance to the natural sciences such that it isn’t exclusively a metaphysical claim. Moreover, the claim that species are essences with strict membership conditions is a metaphysical claim, but it is falsifiable by discoveries in Evolutionary Biology. Presentism, which states that only the present exists (whereas the past no longer exists and the future does not yet exist), is also a metaphysical claim. However, one of the main reasons why many philosophers of time do not accept it is that Presentism seems contrary to Special Relativity.
Panpsychism isn’t immune to any scientific or empirical criticism by virtue of being an abstract metaphysical claim. Moreover, some sympathizers of Panpsychism such as David Chalmers present it with the explicit intention or hope to change the scientific view of the universe. If Panpsychists (or their sympathizers like Chalmers) want Panpsychism to be a new paradigm for physics, they might want to take the objection seriously.
Overall, Panpsychism is an interesting metaphysical view, but I don’t think it could ever become a viable scientific worldview because it is untestable. A panpsychist can’t merely postulate conscious experience at the fundamental level. Its postulation needs to imply some observable consequence that would never be anticipated by a physicist who is a materialist. While a panpsychist could argue that materialism is untestable, this would be unusual because materialism is relatively conservative insofar as it sticks to properties already described and discovered by the scientific investigation. Panpsychists are postulating an extra property, so they have the burden of proof to show that it exists.
Paul So is a graduate student in Texas Tech University’s Master’s Program. His main areas of interests are Philosophy of Mind, Cognitive Science, Metaphysics, and Philosophy of Action (or Free Will).