Having it both ways in materialism

(An improved and updated version of this material has appeared in my book Why Materialism Is Baloney. The version below is kept for legacy purposes.)

The online discussions that followed my previous article have brought my attention to a peculiar point about how the reigning idea that consciousness is (generated by) brain activity is supported today. Uniquely among the sciences, neuroscience seems to be able to use not only knowledge, but also lack of knowledge to defend its current materialist assumptions about the mind-body problem. Allow me to elaborate on this a bit more.

Science operates on the basis of models of nature. A model contains a set of abstract elements that represent entities of nature, these elements being related to one another on the basis of mathematical equations. The dynamics implied by these equations determine how the elements of the model vary with respect to one another. These variations should then correspond accurately to observations of nature, otherwise the model is incorrect. For instance, we model temperature variations in a material by the intensity of vibration of the molecules composing the material. So when our mathematical equations predict that the molecules should be vibrating with higher intensity, we should be able to measure the temperature of the material and take a reading that corresponds to that prediction. This way, the observed entity of nature (that is, the material's temperature as measured with a thermometer) should follow the predicted dynamics of the elements of the model (that is, the vibration intensity of molecules) very closely, otherwise our model does not correspond to reality.

This correspondence between the dynamics of a model and the dynamics of observed phenomena in nature is called a mapping between a model and nature. Determining an unambiguous mapping between a proposed model and nature is the sine qua non of science. The proposed mapping sets the rules of the game by which a theory can be verified through experiment and observation. The theory is validated when the mapping is confirmed through observation, or invalidated when such correspondence is not observed.

When it comes to the problem of consciousness, we have no intuitive way, even in principle, to deduce the properties of consciousness from the properties of matter, as David Chalmers argued. In other words, there is no intuitive way to deduce the intensity, structure, coherence, or continuity of a subjective experience from mass, momentum, spin, or charge. Indeed, nature abounds with structures wherein mass, momentum, spin, and charge combine in extraordinarily varied ways and yet seem to lead to no subjective experience at all. Even in the human brain that is still the case: Much of the neuronal processing in our heads, entailing the exact same kind of neurons that apparently lead to awareness, are completely unconscious. This seemingly insurmountable difficulty in logically deducing the qualities of experience from unconscious matter is called "the explanatory gap," or "the hard problem of consciousness." One would think that such gap would only make it more difficult for neuroscience to defend its current assumption that consciousness is (generated by) material brain activity. But in fact, and surprisingly, the opposite seems to be the case. Here is why.

Since there is no intuitive way to establish a mapping between the qualities of electrochemical brain processes and the qualities of conscious experience, neuroscientists are free to postulate any mapping. This is, per se, not necessarily a problem. The problem is that they seem to also feel free to replace the mapping (not only refine it) depending on circumstances. This is akin to changing the rules of the game after you've already started playing it. For example: Many studies relating the qualities of experience with brain activity indicate a direct proportionality between the intensity, richness, and/or complexity of the experience and the amount of corresponding brain activation – that is, the number of neurons firing and how fast they fire (note that the opposite is not necessarily true, since not all brain activity is accompanied by subjective experience). For instance, the brain of someone sitting quietly with his or her eyes closed will be less activated than that of someone listening to music, which in turn will be less activated than that of someone watching audio-visual erotic material and engaging in sexual intercourse at the same time. This motivates neuroscience to postulate the following mapping: Subjective experience is caused by neuronal firings, therefore the former is directly proportional to the latter (note, again, that this proportionality works in only one direction, since not all brain activations lead to subjective experience). However, in some studies such proportionality is not there at all. Then, neuroscientists may postulate one of a host of different mappings: that experience is correlated to specific neurons, not to the amount of neurons firing; or that experience is correlated to something else going on inside the neurons, but not necessarily their firings; or that experience correlates to some undefined interplay between inhibitory and excitatory brain processes, in a way that breaks the proportionality; etc.

Again, neuroscientists can only do that because, due to the "hard problem of consciousness," we have no intuitive basis to judge whether a certain proposed mapping is reasonable or not. And by changing the mapping depending on what is observed, they can always claim that there is a conceivable way to explain any conscious experience through the assumptions of materialism; there is always a way out through ignorance. This way, on the one hand, they can claim that the proportionality between the intensity/richness/complexity of experience and brain activity, observed in many situations, can be explained under materialism because conscious experience corresponds to the amount of neuronal firings. On the other hand, in other situations they can also claim that, if there is a single neuron alive in the brain, an experience can be attributed to that one neuron because of specificity (never mind proportionality) and, therefore, materialism still holds. Well, one can't have it both ways in one's attempt to defend materialism; one certainly can't have it all ways. These different mappings proposed in different situations contradict one another; they can't all be true.

Neuroscience cannot claim ignorance, leave every door open to defend its materialist assumptions, and then turn around and still claim that they are making enormous progress in understanding consciousness on a material basis. Either they are making progress and can propose one unambiguous mapping between the qualities of physiological processes and the qualities of conscious experience that can be falsified through experiment and observation; or they just don't know what the mapping is and, therefore, there is no theoretical support for materialism as far as the mind-body problem.

In science, one can't have it both ways.


  1. This is my second post to your blog -the first was at the 'new paradigm wanted' thread.

    I'm tremendously enjoying your incisive analysis of the materialist viewpoint.

    I have found your analysis provides excellent compatibility - to a point - to the work executed in 'My Big TOE'.

  2. Bernardo,

    You will know by now that I'm no great fan of the idea that everything has a materialistic explanation. So I raise the following merely as a discussion point.

    The heart beats faster as a result of an increase in physical activity. It also increases as a result of the secretion of adrenaline, not necessarily associated with a marked increase in physical activity.

    It may also beat faster than normal in certain types of clinical conditions (one old woman I saw in hospital while visiting a relative had a resting heart rate of 160 that doctors were trying to normalise).

    Who knows, there may be further ways for heart rate to increase. One can't insist that we can't have it more than the one way (viz. increased physical activity) for heart rate to increase. We can have several, possibly many, ways.

    Now, maybe my example isn't truly analogous, and if so, feel free to explain why not. But it accounts for why I don't feel completely confident that it’s only legitimate to have one kind of mapping to explain the intensity of the subjective experience. In general, this may indeed link to degree of neuronal excitation, measurable by fMRI.

    I do favour the idea that consciousness isn't merely an epiphenomenon of brain activity, and that the brain is quite possibly a filter of extra-brain consciousness. On the other hand, I also recognise that there will be neurophysiological correlates even if consciousness does not originate in the brain.

    fMRI correlates are one thing, but we can’t, *a priori*, eliminate the possibility that there could be other correlates which at present we can’t measure so well. I know full well that materialists want to have it all ways, but that doesn’t actually exclude the possibility that there may in fact be several ways of explaining the subjective intensity of experiences (and I’m not trying to play down the importance of this by use of the word “subjective” – in the end, I believe *all* experience is subjective).

    Any comments?

    Michael Larkin

    1. Hey Michael,
      I think it's enormously useful to try to defend the position contrary to our own; it's the only way to increase confidence in our own position. So I appreciate your comment!
      In your analogy, I would say neuronal firing is equivalent to hear rate. Yes, there are many ways neuronal firings could be made to increase, just like heart rate can increase due to many causes. But if experience correlates to neuronal firings, then, regardless of how it increases, it should still increase for more complex/rich/intense experiences. And, therefore, it should be measurable with an fMRI or EEG.
      Now, one could choose an alternative mapping and state that experience correlates to something else, other than neuronal firings. Hameroff does that with his microtubule quantum activity. That's OK, but then we still have to explain why so often experience does seem to correlate with firings.
      Either way, I think neuroscience cannot claim that materialist assumptions are being substantiated until they can pin down one unambiguous mapping that can be falsified through experiment.
      Cheers, Bernardo.

    2. Just to expand a bit more on my argument: Regardless of what causes heart rate to increase, when it does increase you can measure the increase. In the analogy, I'd claim the same for neuronal firings: regardless of what causes them to increase, the increase should be measurable with an fMRI or EEG. Now, if experience IS neuronal firings, then an increase in the complexity/richness/intensity of the experience should correlate with a measurable increase in neuronal firings, regardless of what caused it.

  3. Hi again Bernardo,

    Well, I was really trying to draw a parallel between heart rate and (subjective) intensity of experience rather than neuronal firing. Yes, we can measure heart rate, but we can also measure subjective intensity of experience (if we are inclined to believe experimental subjects, which I am).

    Now it may indeed be that generally, subjective intensity of experience correlates with degree of neuronal firing. But who is to say that it doesn't also correlate to something else other than that?

    Do you see the distinction I'm making? Your thinking that I was saying neuronal firing is equivalent to heart rate in my analogy wasn't what I actually meant, though maybe I didn't express myself clearly enough.

    Michael Larkin

    1. Yes, I understand now what you meant. In a way, it's an appeal to ignorance from the side of materialism: "It could be something else." Yes, that's true by definition. But from this it doesn't follow that materialism is well-supported by the evidence; on the contrary.

      I do see an issue in your analogy, though. Heart-rate increase may have many _causes_, but it _consists_ in the speed of contraction of heart muscle cells. Analogously, more complex subjective experiences may have many _causes_ (e.g. a fight with your wife, watching an IMAX 3D movie, going to an amusement park, etc.), but what does it _consist_ in? You see, it's the latter question that is in dispute: What _is_ subjective experience as far as material brain dynamics? Not necessarily what causes it.
      The question "What _is_ subjective experience?" should have a single, unambiguous answer, in my view, even if it's a composite answer (e.g. a function of more than one variables).

  4. This reminds me of Popper's comparative analysis between Einstein's Theory of Relativity and Freud's psychoanalysis - when the latter was shown to be able to change the models as he saw fit in order to match the empirical observation, Popper called it pseudoscience.

    Is Neuropsychology pseudoscience?