Is Materialism Better Suited to the Scientific Outlook? (Column 593)

With God’s help

Disclaimer: This post was translated from Hebrew using AI (ChatGPT 5 Thinking), so there may be inaccuracies or nuances lost. If something seems unclear, please refer to the Hebrew original or contact us for clarification.

I have already discussed here in the past (see for example in column 397 and elsewhere) the notion of “emergence” as it arises in the context of the mind–body question. In lesson 28 of the series Doubt and Statistics, another point regarding this view came up, which I wished to present here. This discussion also connects to the recent series of columns on artificial intelligence, since, as I noted, many AI researchers tend to think that from a certain stage on, mental phenomena will also appear in it. They link these phenomena to material complexity, in effect implicitly assuming emergence. I wish to devote this column to a more systematic clarification of emergence and its scientific standing, and thereby also to add a few comments on the question of artificial intelligence.

Materialism and Dualism: A First Look

The debate between materialism and dualism is tied to other issues, like free will, etc. But usually the connection runs the other way: because they adopt a materialist view, people deny free will (for if the whole world, ourselves included, is physics—then apparently we have no free will). So in this column I will set aside the question of free will and focus on materialism.

Classical views tended to see the human being as a dual creature, composed of body and soul. I think this is also our initial intuitive sense today, since within us are mental phenomena that on the face of it do not seem to be something that can be ascribed to the body or even to our material brain. It seems one needs good reasons to give up dualism. The reasons to adopt a materialist stance nonetheless are based mainly on two advantages materialism has over dualism: simplicity and scientificity. 1) Occam’s razor tells us to adopt a view that contains fewer assumptions and fewer kinds of entities (only matter without spirit). Materialism is a simpler view and is therefore preferable by the razor, unless there are good reasons against it. 2) Science ostensibly teaches us that we are creatures made of matter, and there is no indication that we contain anything else. Put differently, the claim is that dualism is an unscientific (mystical) position, since it offers a thesis that is not falsifiable. Think how one could subject the thesis of the existence of a soul or spirit to an empirical test of refutation. Is there a scientific experiment that could examine this? At present it seems that all the processes familiar to us in the psyche can be explained from a physicalist starting point, and even if there is something we do not yet understand, it appears to be only a question of time and further research. In light of the advances in these fields in recent generations, it does not seem that there is a gap or fundamental problem here (except perhaps for free will, if one assumes we have it).

The Naive Materialist Attitude to Mental Phenomena

Many materialists identify mental phenomena with neuronal processes. When we think, want, feel, remember, and so on, one can consistently point to certain regions in the brain that are active and generate these phenomena. Therefore brain researchers link the mental phenomena with the corresponding neuronal activities that generate them. Free will is an exception in this regard, and thus most brain researchers tend to deny its existence. For this reason I will not enter here into the question of free will, but will focus precisely on the other mental phenomena, for which neuronal counterparts can be found.

I believe I once recounted here a conversation I had with Prof. Yosef Ne’eman of blessed memory, who was a materialist and a prominent, active atheist. After my book God Plays Dice came out, in which, among other things, I critiqued some of his writings, he called me and we had a long and very interesting conversation. He told me that many of his materialist colleagues identify mental phenomena with neuronal processes in the brain, and he was frustrated by his inability to explain to them the folly of this view. One may claim that the brain generates the mental phenomena, but one cannot identify a mental phenomenon with an electrical current. That is simply a category mistake. The electrical current produces in me (in my soul?) a desire for chocolate or raises a memory of Grandma Yokheved, but the electrical current itself is not the memory or the desire. It is merely a categorical muddle. By way of analogy I will again bring two examples I have often presented in the past.

Bertrand Russell, in his book The Problems of Philosophy, asks: what is the color yellow? It is commonly answered that it is an electromagnetic wave of a certain wavelength (corresponding to yellow). But that is of course a mistake. An electromagnetic wave is a physical phenomenon, whereas the experience of the color yellow is a mental phenomenon. The color yellow appears in my consciousness after the electromagnetic wave hits my retina and undergoes processing in my particular brain. The brain of another creature built differently might connect the eyes to the brain’s auditory center. In such a case that creature would respond differently to the striking of yellow light on its retina. It might hear a certain kind of sound or some melody (the Ninth Symphony) for each “color” (i.e., each electromagnetic wave of a given wavelength) that hits its retina. Does such a creature apprehend reality incorrectly? Certainly not. It apprehends it differently from me, but there is no right or wrong here. Each according to its sensory system. Color and sound do not exist in the world itself but only in our consciousness, and they are a product of its specific structure.

Another example is the folk parable about a tree falling in a forest when there is not a single living creature’s ear there listening. The question is whether the fall of the tree produces a sound. This may surprise many of you, but the answer is negative. Of course a pressure wave in the air is produced, but so long as there is no eardrum for that wave to strike, no sound will be created. Sound is a phenomenon of consciousness, not a physical phenomenon, and it arises as a result of the physical phenomenon (the pressure wave) striking the eardrum, and from the structure of our brain (how it processes auditory information). So too color arises from an electromagnetic wave, which is a physical phenomenon, and from the structure of our consciousness.

Identifying electrical currents in the neuronal system with mental phenomena is similar to the mistaken identification of an electromagnetic wave with color or of a pressure wave in the air with sound. At most there is a cause-and-effect relation here, but not identity. Incidentally, it is not always clear what is cause and what is effect. Does neuronal activity necessarily generate the mental phenomena, or are there cases where they are the ones that generate it?[1] This is the naive materialist approach, and clearly it is mere confusion.

Emergence

Returning to my conversation with Ne’eman: I told him that I was about to write another book precisely on this subject (mind and brain; in time my book The Science of Freedom was published), and he replied that I was preaching to the choir. But the question that may arise here is what Ne’eman himself thought on this matter. After all, he was a materialist who did not believe in the existence of non-material entities. So if he did not identify mental phenomena with physical phenomena in the brain, how does that align with his materialism? This brings us to the thesis of emergence.

Ne’eman’s being a sober, non-confused materialist does not make him a dualist. He believed wholeheartedly in materialism and denied the existence of an additional human substance (spirit or soul). For him, mental phenomena are features or functions that appear at the collective level of our neuronal whole. The claim is that when many neurons are connected into a network, as in the brain, mental phenomena suddenly appear (emerge). According to the emergentist view (emergence), there is no need to posit the existence of a spirit in order to speak about mental phenomena, even if one does not identify those phenomena with neuronal phenomena. The claim is that at the ontic level (the level of entities) there is only matter in us, and the mental phenomena are properties of the material whole that is us.

Thus, for example, the color yellow or sound are not entities in the world. They are properties of objects in the world. The yellow color of a car or the sound of a piano are not entities but properties of entities. Does that mean they are subjective? Certainly not. They exist in the world itself but not as entities. The sober materialists argue that mental phenomena are properties of the brain, i.e., they exist and are not fictions (and certainly not electrical currents), but there is no entity of a different kind here (a soul or spirit).

It seems to me that emergence is the dominant view today among brain researchers and philosophers who deal with the subject, since it appears to them to be a perfect solution to the tension between our intuitions and science (and Occam’s razor). You need not posit another kind of entity (spiritual), and on the other hand you need not give up your intuitions regarding mental phenomena. As noted above, this is also apparently what brain researchers and AI people think when they assume that in a sufficiently complex AI program, mental phenomena (such as consciousness, emotions, desires, etc.) will also appear. They take these phenomena to be properties of physical complexity as such, and that is precisely emergence. There is no need to posit a “ghost in the machine.”

[Parenthetically, in fact they assume something much stronger: that these are properties of a complex brain-like structure that performs thinking (or more precisely, “computation”). In their view the dependence is only on the function of thinking, regardless of desires, etc. The view is that thinking generates emotions, desires, consciousness, and more. Hence in their opinion a digital analogue that performs only “thinking” (or computation) necessarily possesses those mental properties or capacities. I argue this is more novel than the previous claim, since there was room to say that even if the mental is indeed a property of human bodies (emergence), it does not follow that a digital imitation that mimics only our thinking would also produce all the other human mental phenomena (like will, emotion, consciousness, etc.). Why not suppose that only one who has thinking, emotions, and desires will also have consciousness? Or conversely, that one who has consciousness will also have emotions and thinking? Their assumption is that the existence of thinking—or, in fact, computation—alone is a sufficient condition for the emergence of the mental dimension. To me this is highly speculative and far-reaching.]

The Connection to the Question of Vitalism in Biology

We must remove another needless ghost from the table. Until not long ago there was a fierce debate among biologists about vitalism. Some argued that the phenomena of life cannot be explained without positing a vital principle beyond chemistry and physics. Today “vitalism” is a dirty word, for biologists believe we possess a complete conceptual framework, i.e., a framework that can explain all life phenomena without resorting to a vital substance. The gaps regarding what we still do not know can also be filled in this way, without needing the vitalist assumption.

It is important to understand that I, as a dualist, can also accept this scientific thesis. It is quite possible that biology can be explained without resorting to the assumption of the existence of a vital component in matter. This means that biology emerges out of physics and chemistry. We do not yet have a full account of all the steps from physics to biology, but it seems we are certainly on the way. My question concerns the explanation of mental phenomena, not biology. That is a completely different question, and there, as we shall see shortly, the situation is apparently different.

Put differently: the methodological assumption of today’s biologists is that there is no need to posit the existence of a vital substance. This is an efficient and fruitful methodological assumption. But it is not necessarily correct to infer from this philosophical or ontological conclusions about what truly exists. It is entirely possible that there is a vital component in us even if it is not needed to explain biology. Consider the assumption about God’s existence. Science does not need this hypothesis, as Laplace said. Does that mean there is no God? No. It only means that as a methodological assumption, atheism is very fruitful and efficient. The question of whether there is in fact a God is another question.

The question, of course, is why posit a vital component, or the existence of God, if we manage perfectly well without it? Occam’s razor strikes again. If there is no scientific necessity for it, the simple conclusion is that our methodological assumption is also the ontic truth. But there is still a leap here, which is not necessarily correct. For example, if we have good arguments for the existence of God (say, we believe He revealed Himself to us, or that He is the only foundation for the validity of morality), then even if there is no scientific necessity for His existence we will believe in Him. After all, for me there is also no necessity for your existence, but I met you and therefore I know you exist. And what about vitalism? Here we must return to mental phenomena and see whether they might compel us toward it. If there is no good explanation for mental phenomena, perhaps that will lead us back to vitalism, even if in terms of biology it is unnecessary.

As we shall see, emergence—the most reasonable materialist–scientific explanation proposed for mental phenomena—also requires problematic assumptions at the scientific level.

Emergence

The philosopher John Searle (of “the Chinese Room” fame), in his book Mind, Brain and Science, was one of the first to raise the possibility of emergence as an explanation for mental phenomena, though in his work it was not yet called that. To clarify his intent he used the example of liquidity. A single water molecule is not liquid (nor solid or gas). States of matter are properties of a large collection of molecules, not of individual molecules. Thus a collection of water molecules can be liquid, while at the same time none of the individual molecules themselves has that property. Here we have properties that arise at the collective level, even though they are not relevant to, and do not exist in, any of the parts that compose that collection.

Searle argues that this example leads to a new possibility for understanding the relation between mental phenomena (our mind) and the body. It may be that these phenomena are properties of the material collective (the body, or the brain), although they do not characterize any single part (cell or molecule) that composes it—just as liquidity does not characterize an individual water molecule.

Weak and Strong Emergence

The problem with Searle’s example is that a property like liquidity, or a state of matter in general, can be grounded in the properties of the individual parts that compose it. Physics knows well how to derive the state of matter from the properties of the molecules that compose it. The field around a molecule is responsible for the bonds it forms with other molecules, and those bonds determine the entire phase diagram of the substance (a map describing its state of matter at every given pressure and temperature).

By contrast, the situation with respect to mental phenomena is entirely different. There is currently no way to explain how they are brought about by the neuronal currents and electrical fields in our brain. We saw above that the identification made by naive materialists between the brain’s chemical–electrical phenomena and mental phenomena is mistaken. This is also true of any identification one might make between the properties of a single water molecule and the liquid state of a mass of water. But now I add another claim: unlike the case of water, the leap from electrical currents to mental phenomena is something for which we have not a shred of explanation. Not only do we lack an explanation; we lack any conceptual framework that would allow us to formulate and seek such an explanation. This is a black hole in our understanding of the reduction of neuroscience to the laws of physics. As I explained above, all we find in brain research are correlations. Such-and-such a physical phenomenon in such-and-such a region of the brain produces such-and-such a mental phenomenon. But how does it do this? Can we perform a calculation whose inputs are physical phenomena in the brain and whose output is a mental phenomenon? Absolutely not. We focus on and stop at determining correlations, and at most cause-and-effect relations (as I noted above, that is not the same thing) between the physical and the mental. That is all.

Well, perhaps this is something that will be deciphered later on. Maybe it is a gap that further scientific research will cover. Perhaps yes, perhaps no—but the fact is that for now there is not even a direction. Therefore assuming that an explanation will emerge is a rather wild speculation in light of current knowledge. It is no wonder, then, that even atheists seek other explanations, i.e., they insist on materialist–emergentism without positing that future science will find an explanation for this gap. The standard proposal in this area is based on a distinction between weak emergence and strong emergence.

The claim is that the liquidity of water is the product of weak emergence; that is, an emergence that does have an explanation in terms of the properties of the parts. One can reduce the collective phenomenon to the properties of the individuals that compose it. Moreover, all emergent phenomena currently known in science are of the weak kind. There is no other known scientific example of strong emergence. But in principle strong emergence could also be possible, namely the emergence of a property of the collective that cannot be reduced to the properties of the parts that compose it. In the future science may discover additional examples of such phenomena, and in particular discover that mental phenomena are indeed of this kind. Therefore even today one can assume that mental phenomena are products of strong emergence, and that suffices to reject dualism (i.e., the assumption of a spirit’s existence).

Can Strong Emergence Be a Scientific Claim?

Is that a sufficient explanation? It is of course possible, since none of us knows how to predict the future. But this proposal suffers from a flaw that places it under a big question mark. My claim is that the fact there is currently no known example of strong emergence is not a mere coincidence. There cannot be such an example, and therefore even in the future one will never be able to demonstrate scientifically the existence of strong emergence. Strong emergence is, by definition, a claim that cannot be refuted or confirmed scientifically, and so it is doomed to remain so in the future as well. I claim that this is a fundamental gap that further scientific research will not close. I will now try to explain this claim a bit more.

Try to think of a situation in which science discovers an example of strong emergence. This would mean that science discovers a collection of items X that has a property A that does not exist in any of the items (i.e., an example of emergence). At the same time science would also determine that there is no way to reduce the collective’s property to the properties of the composing items (otherwise it would be weak emergence). What would you say in such a case? Would we all accept the conclusion that this is in fact strong emergence? Is that a scientific conclusion? Surely not. Because two further interpretive alternatives always remain in such a situation: (a) such a reduction exists but has not yet been discovered. If so, this is actually weak emergence. (b) there is an additional component in this collection beyond the set of X items that we did not factor in, and it is responsible for the emergence of A. Therefore there is no emergence here at all. Note that neither of these two interpretations can be ruled out by scientific means. Whenever we are in the described situation, three interpretive possibilities remain: strong emergence, weak (as yet hidden) emergence, or not emergence at all. There is no scientific way to rule any of the three out, and therefore we cannot scientifically conclude that we are dealing with strong emergence.

Note that this conclusion follows from the very logical structure of the problem/claim. It follows that there cannot be a future situation in which we discover a phenomenon about which we can scientifically determine that it involves strong emergence. This means that the claim that mental phenomena are cases of strong emergence is not a scientific claim. It can neither be confirmed nor refuted. It is a philosophical assumption (to my mind a very odd one), and nothing more.

Is Materialism Suited to the Scientific Outlook?

As I mentioned, materialism tends to present itself as a scientific thesis, seeing itself as a logical and reasonable alternative that stands firm on a rational and scientific basis against dualism, which is presented as a kind of mysticism. We saw that it attacks dualism on the grounds that its claims cannot be scientifically confirmed or refuted. You must admit that against this backdrop the picture I have presented here is rather strange. We have reached a situation in which the claim made in the name of science against the mysticism of the dualists runs into a scientific dead end, and it turns out that it, too, is not scientifically falsifiable. What then is the difference between it and the dualist claim? Moreover, even if it is not a scientific claim, it is unclear why materialists consider it preferable to the highly intuitive dualism.

One can of course resort to Occam’s razor (which we saw above also serves as a basis for the argument against dualism). If we have two explanations of equal value (as we saw, both unscientific and unfalsifiable), then ostensibly we should choose the simpler one. The materialist can claim that he chooses his position and rejects dualism because of the consideration of simplicity.

But the consideration of simplicity never stands alone (see on this in column 426). If one of our two explanations has some advantage, then it is reasonable to take that into account and not choose the other solely because it is simpler. Simplicity is a criterion for choosing between two equivalent explanations. But in our case we have seen that although both explanations are unscientific and unfalsifiable, there are still philosophical considerations that lead to the preference for dualism and to the rejection of the possibility of strong emergence. In such a situation, choosing materialism because of Occam’s razor is a puzzling step. In short, it is neither reasonable nor logical that there should be in the collective something that is not a product of the parts that compose it (unless there is an additional component—dualism). And I have not even mentioned our free will (which, in my opinion, exists; and certainly we must admit that the feeling of free will exists—what is its source?), which, unlike all our other mental functions, has not even a neuronal analogue responsible for it, and it is certainly unlikely that it will have a reductive explanation in the future.

The conclusion is that strong emergence is a kind of mystical claim, and on its face it contradicts all the foundations of scientific thinking (which explains complex phenomena by decomposing them into their components). It is much more reasonable to suppose that if there is something here that cannot be reduced to its components, there is probably an additional component we do not see that is responsible for it. That is how science always works: if I see an unexplained phenomenon I infer that in the background there exists some theoretical entity that accounts for it. Thus a scientific theory is formed.

If so, the materialism that speaks in the name of science not only advances unscientific claims, but these claims also stand in stark opposition to the very mode of scientific thinking. Choosing an illogical explanation merely because it is simpler seems absurd to me.

True, all this concerns only the claim of strong emergence. But we are apparently left with a third alternative: perhaps in the future an explanation will be found that succeeds in grounding mental phenomena in the properties of neurons, and then we would have weak emergence. But one does not go to the grocery store with hopes, particularly hopes like these which, at least at present, seem to have no reasonable basis. One could also claim that man is immortal because perhaps in the future a technique will be found to prevent death. In any case, this is certainly not an option more suited to science and scientific thinking than dualism, and to my judgment it is also, on its own terms, far less plausible than dualism.

I will conclude by addressing Roy’s comment on the previous columns.

How Can We Know?

Roy wondered there how one can know whether there is in us—or in thinking machines (now or in the future)—something beyond matter and computations. If this has no functional manifestation (after all, the assumption is that the machine is a perfect imitation), how can one determine or even estimate whether mental dimensions are present or not? This is whether we see them as emergent properties of the material whole or as functions appearing in the soul.

This wondering has its roots in a scientific outlook that examines positions or claims through experiment and measurable functional characteristics. Here we have seen that this very outlook suffers from a lack of scientificity. According to my dualist position, which holds that there is in us something beyond matter, it is quite likely that machines do not have such a thing. We know what we put into them, and there is no reason to suppose they contain anything beyond that—unless there will be functional indications of it. Precisely as I rejected strong emergence as a hypothesis without real basis. Dualism, with regard to me, is founded on an immediate sense of self. I experience that there is a spirit within me. Even if one accepts the emergentist view with respect to ourselves, this is only because it is clear to us that there is within us a mental dimension. Without that, there would be no reason to assume it. Therefore in a machine, so long as we do not observe mental phenomena, there is no reason to assume they exist. The fact that the machine “thinks” (more accurately, “computes”) says nothing about mental dimensions.

One may, of course, wonder about the problem of the other, what is called “other selves.” How can we know that others have a mental dimension or even a spiritual component? After all, in their case I do not have such an immediate apprehension. I can, of course, rely on their report—but perhaps they are lying?! The similarity between them and me leads me to the conclusion they are not lying and are probably built like me. The similarity between me and a machine, however, is decidedly not of that kind, even if it reaches a marvelous level of “thinking.” It is merely an imitation (even if a very successful one) of our thinking processes, and nothing more.

[1] At least regarding free will, the relation is probably the reverse. See chapter 17 of my book The Science of Freedom and in particular here for experimental confirmation of the claim made there.