By Matthew Owen
In her 1985 hit song Material Girl, Madonna summed up the prevailing view in philosophy of mind: “We are living in a material world and I am a material girl.” Madonna comes close to echoing a description of ‘materialism’ and its cousin ‘physicalism’ in the Cambridge Dictionary of Philosophy: “Physicalism in the widest sense of the term, [is] materialism applied to the question of the nature of mind” (Shoemaker 1995: 617). Materialism claims that everything that exists is material, and physicalism applies this to the mind. So according to materialism/physicalism, Madonna is indeed a material girl with a material mind living in a material world.
With leading philosophers widely embracing it (not to mention a hit song in its favor) materialism could not fail to prevail. In the early nineties John Searle (1992: xiii) made the following assessment concerning contemporary philosophy of mind: “mainstream orthodoxy consists of various versions of ‘materialism.’” True, volumes like The Waning of Materialism and After Physicalism published in the last decade by Oxford Press and Notre Dame Press respectively do suggest that the times, they are a changin’, as the prophet Bob Dylan put it.
Nevertheless, materialism has been the dominant view in philosophy of mind since the middle of the twentieth century (see Göcke 2012: 1). Yet some philosophers, such as Laurence BonJour (2010: 3), are perplexed by this definitive embrace of materialism:
I have always found this situation extremely puzzling. As far as I can see, materialism is a view that has no very compelling argument in its favor and that is confronted with very powerful objections to which nothing even approaching an adequate response has been offered.
If one resonates with BonJour’s puzzlement, like a doubting Thomas in modern philosophy, it’s natural to ask: Why do so many brilliant contemporary philosophers embrace materialism?
Two and a half decades ago Searle (1992: 3-4) put his finger on a powerful motivation for the widespread embrace of materialism, which remains persuasive: Materialism is considered the only scientific option. But this ought not be due to methodological naturalism, which is a methodological presupposition, not a scientific conclusion. Nor can it be due to the idea that physics is causally closed; for although this is a position held by many scientists, it’s not a testable hypothesis proved by any observer of the whole physical universe (cf. Lowe 2003).
However, there’s a more potent premise that hinges on neuroscience. Simply put, the premise is that neuroscience says Madonna is indeed a material girl. The idea, which is quite popular, is that modern neuroscientific discoveries entail that Madonna’s mind and everyone’s mind is purely physical and thus physicalism/materialism is the only scientific option. Many philosophers agree with Nancy Murphy’s (1998: 13) claim that neuroscience provides “dramatic evidence for physicalism.” The principal data from neuroscience comes in the form of neural correlates of consciousness (for brevity NCC). It’s NCC that allow us to map parts of the brain that correspond to mental states and actions (see brain-map.org).
Nearly three decades ago, the contemporary search for NCC was instigated by Francis Crick and Christof Koch (1990). Nobel Laureate, Francis Crick co-discovered the double helix structure of DNA before focusing on the mind-body problem and NCC during the latter part of his influential career. His longtime collaborator and close friend, Christof Koch is now the President and Chief Scientific Officer at the influential Allen Institute for Brain Science in Seattle, Washington. Koch (2016: 307) defines an NCC as “The minimum neural mechanisms jointly sufficient for any one specific conscious experience” (cf. Chalmers 2000). The famous, but perhaps mistaken, example of an NCC is C-fiber activation in one’s brain that corresponds to one’s mental state of pain.
Fundamental to identifying neural correlates of consciousness is finding neural activity that consistently corresponds with certain conscious states. Imaging brain activity is central to this endeavor. Over a century ago Italian physiologist, Angelo Mosso (1846- 1910), laid the conceptual basis for brain imaging techniques vital to the contemporary search for NCC (see Sandrone et al.: 2014). A chief challenge to studying the brain is that it’s enclosed in a hard protective casing – i.e. the skull. Mosso worked with a patient named Bertone who suffered extensive damage to the top of his skull, consequently much of it was missing. Where Bertone’s skull was missing, Mosso placed a cap made out of a rubber-like substance, gutta percha (Glickstein 2014: 343). This flexible cap made it possible to record brain pulsations of blood pressure correlated with mental activity such as emotional arousal and doing arithmetic (Glickstein 2014: 343).
Mosso’s study confirmed a straightforward hypothesis. That is, if the brain works harder there will be increased blood flow to the brain, so if the brain works harder when the mind works harder, there will be increased blood flow to the brain when the mind works harder. Put differently, (a) increased mental activity means (b) increased brain activity, which means (c) increased blood flow in the brain. So (a) increased mental activity correlates with (c) increased blood flow in the brain. Mosso confirmed this by measuring the increased blood flow in Bertone’s brain that took place when Bertone’s mental activity increased.
However, the method Mosso used to measure pulsations in Bertone’s brain had a limitation. It was effective only if the patient had an abnormal skull breach (Sandrone et al. 2014: 622). Mosso’s ingenious ‘human circulation balance’ was invented to overcome this limitation (Sandrone et al.: 2014: 622). Mosso had his patients lay on a table that was essentially a balance intended to measure pulsations of blood flow that would tip the balance. Whether or not Mosso’s human circulation balance was reliable, his work laid the conceptual foundation for noninvasive functional brain imaging techniques (Sandrone et al. 2014: 621-622). Noninvasive functional brain imaging is done while the brain is active and without being invasive to the brain by penetrating it in any significant way. Such brain imaging is vital to the search for NCC. Since it allows us to discover brain activity that corresponds with a conscious subject’s mental activity.
Needless to say, noninvasive functional brain imaging technology has advanced significantly since Mosso’s day. In the 1920s a German psychiatrist, Hans Berger, discovered it’s possible to record electrical activity in the brain from the human scalp (Glickstein 2014: 338). This type of recording is called an electroencephalogram (EEG) (Glickstein 2014: 338). Positron emission tomography (PET) and functional magnetic resonance imaging (fMRI) have also since developed. A PET scan can reveal blood flow, glucose metabolism, oxygen metabolism, or concentrations of dopamine transporter indicative of brain activity (Johnson and Becker 1999). An fMRI reveals increased brain activity by revealing increased blood flow. The blood carries oxygen on molecules of hemoglobin carrying iron that changes the thermodynamic and magnetic properties of the brain area, which is detected by magnetic resonance imaging (see Bulte 2011: 4:15-5:00).
This technology is invaluable in the contemporary search for NCC. It allows us to identify the neural activity in the brain that correlates with our conscious mental activity. By observing brain activity via fMRI while subjects in a study report their conscious experience, we can identify the NCC of that experience (cf. Guta 2015). And the methodological “no-report paradigm” is now being used as well. This paradigm has arisen out of a concern that the standard method confuses NCC of the conscious experience of seeing a red image with the NCC of reporting seeing a red image.
What NCC Entail
While the evidence for neural correlates of consciousness may be clear, it’s not obvious what such correlations entail regarding the nature of the mind. At this point a common fallacy – post hoc, ergo propter hoc – is tempting to commit. This fallacy is committed when one infers that f caused j simply because f is correlated with j. If one inferred that Barack Obama winning the democratic nomination in 2008 caused the stock market crash of 2008 simply because there’s a correlation between the two events, they would commit this fallacy. Correlation doesn’t entail causation. One needs further rationale to infer that a correlation is best explained by (or suggests) a causal relation.
Likewise, a correlation by itself doesn’t entail dependency, identity, or that one correlate is reducible to the other. Given that f is correlated with j, we need more information to justifiably conclude that f depends on j, is identical to j, or is reducible to j. Suppose that all around the world whenever any philosopher heard a knock on their door they found a packaged philosophy book on their doorstep. On the basis of this correlation alone we couldn’t justifiably infer that the packaged books cause the knocks, lest we commit the fallacy mentioned above. But likewise, we couldn’t infer that the door knocks depend on the philosophy books. Nor could we infer that the knocks are identical to the books or in someway reducible to them. Further information would be required to justifiably make such inferences.
In some cases where we’ve identified an NCC we might have additional data that justifies further inferences. For example, suppose that whenever Fern was in a mental state of remembering her childhood in Kansas, a particular part of her brain lit up. And suppose further that the same area lights up in Kathryn’s brain whenever she remembers her childhood in Washington State. Moreover, presume the data related to Fern and Kathryn confirms numerous studies with many human subjects. Given this, we could know that Fern and Kathryn’s mental states of remembering their childhood correlate with neural activity in a particular part of the human brain.
From this correlation alone we couldn’t infer that Fern and Kathryn’s mental state caused the neural activity, depended on it, was identical to it, or reducible to it. However, suppose that Fern and Kathryn lost the part of their brain with the neural correlates and directly after this they could never again remember their childhoods. Given this, the data set would then include more than just the correlations. With this additional data it would be justifiable to conclude that their mental state of remembering their childhood was not only correlated with the neural activity, but also depended on it.
If a physicalist assumed physics is fundamental before mapping the correlation between Fern and Kathryn’s mental activity and the corresponding neural activity, she would likely conclude the correlation implies that the mental activity depends on the neural activity before gaining the additional data. And she would be justified in doing so to the degree that her assumption was well justified. But she would not be arriving at this conclusion merely on the basis of the correlation. Rather she would be justifiably arriving at the conclusion on the basis of the correlation coupled with her preexperimental assumption. Similarly, a dualist could justifiably conclude that Fern and Kathryn’s immaterial minds stood in some type of causal relation with the neural correlates, if she justifiably assumed that their mental states are not reducible to physical states. However the dualist, like the physicalist, would be arriving at her conclusion on the basis of the NCC and her preexperimental assumption.
Pessimists might roll their eyes at this point and remind us that everyone has pre- experimental assumptions. That’s true. Nevertheless, that doesn’t mean we can’t learn anything from NCC. It does mean, however, that it’s important to analyze our pre- experimental assumptions and to be aware of the justificatory role they play. And according to David Chalmers (1998: 227), “once we recognize the central role of preexperimental assumptions in the search for the NCC, we realize that there are limitations on just what we can expect this search to tell us.” Given this, it’s fitting that Thomas Metzinger (2000: 4) writes in his influential volume Neural Correlates of Consciousness:
However, mapping does not mean reduction. Correlation does not mean explanation. Once strict, fine-grained correlations between brain states and conscious states have been established, a number of theoretical options are still open…Assume that we find a strict and systematic correlation between a certain brain property or type of neural event N and the subjectively experienced phenomenal property of “sogginess” S. This is entirely compatible with Cartesian dualism: The underlying relation could indeed be a causal one, namely causal interaction between events in two ontologically distinct domains. If the ideas of Descartes or those of Popper and Eccles (see Popper and Eccles; Popper 1996) were correct, then we would certainly find neural correlates of consciousness.
Metzinger’s words are echoed by Christof Koch. In personal correspondence he reminded me of the following:
Note that the NCC themselves are neutral from the point of view of physicalism/materialism or one of the various shades of dualism. Under any reading, consciousness will have physical correlates. The question is what are those correlates, where are they and what can and will they tell us about how consciousness is generated in the first place. They also don’t speak to whether or not consciousness can be analyzed using reductionism or other mereological assumptions.
Both Metzinger and Koch point out that dualism is consistent with NCC. Interestingly, dualist critic, Jaegwon Kim has gone a step further. In Physicalism, Or Something Near Enough, Kim (2005: 124) acknowledges that some dualist views even entail NCC (cf. Oderberg 2005: 90; Moreland 2016: 116-119).
The data revealing that there are NCC is well established and will become more so as the contemporary search for neural correlates continues (cf. Shulman 2013). This is interesting data, but the fact that NCC exist doesn’t tell us much about their nature. And as Koch points out, the real question is: What are the correlates, i.e. what’s their nature? And as Metzinger (2000: 4) points out, we need to ask whether particular views are even possible? The answers to these questions will tell us much more than the fact that the correlates exist. But it’s important to recognize that these questions of nature and possibility are metaphysical questions pertaining to the metaphysics of the mind. Thus to discover more about the conscious correlates of neural correlates studied by neuroscientists, we’re brought back to metaphysics (cf. Tahko 2012: 41-42; Moreland 2016: 118). Perhaps neglect of the requisite metaphysical analysis explains why philosophers like John Searle (1992: 4) have found much contemporary mainstream philosophy of mind to be “obviously false” (cf. BonJour 2010:3-4; Plantinga 2007: 135-136).
Returning to the main question: Does neuroscience say Madonna is a material girl? Perhaps a better question would question this question. Neuroscience provides us with invaluable information about neural processes that are integral to our mental activity. But the idea that the latter is the former doesn’t follow from such information provided by neuroscience. What really matters is what metaphysics of mind will tell us about Madonna, and whether this material girl is merely material.
Author’ note: I am indebted to Christof Koch and Mihretu Guta for helpful conversation about the ideas presented here. That’s not to say, however, that they agree with me on every point.
Matthew Owen teaches philosophy at Heritage University. He’s also a doctoral researcher at the University of Birmingham. His research applies neo-Thomistic hylomorphism to mental causation and neural correlates of consciousness. He blogs at mkowen.org.