In this post I recap my preferred framework for tackling the hard problem of consciousness, and then try to extend the analysis a bit deeper. I have favored a “divide and conquer” strategy that recognizes two dimensions to the problem. Our conscious experience has both a qualitative character and a subjective character, and both can seem difficult to reconcile with scientific descriptions of the world. However, I have thought that if one takes a Russellian monist stance with regard to question of qualitative character, then this can clear the way toward seeing subjectivity as a question for science—specifically as an essentially biological phenomenon.
Below I consider an objection alleging that this strategy still leaves a lingering philosophical problem when it comes to subjectivity (an “explanatory gap”). I will try to offer the beginnings of a response by doing something I have previously criticized: that is, adding a dose of fairly speculative metaphysics! This involves plumbing my preferred theory of causation for “proto-subjective” ontological features.
Russellian Monism and its Variants
I have argued that the best philosophical perspective on the mind-body problem is that of Russellian monism (named for Bertrand Russell and the view he developed in his 1927 work, The Analysis of Matter). The approach begins by noting that physical theories feature mathematical models that seek to represent the structure and dynamics of natural systems. However, there is more to concrete reality than these abstract representations. Natural phenomena also have an intrinsic character, and we have some insight into this character by virtue of our own conscious experience. There is no reason to think that there is a sharp divide between the nature of the physical world and the nature of our conscious experience. It is just that physical descriptions leave out an aspect of the world that is important for understanding consciousness.
In developing the view, Russell described this intrinsic character in terms of qualities. We know from perception that the world contains qualities (while abstract models do not). We can infer that unobserved physical systems generally have a qualitative aspect as well. Some theorists take the Russellian insight and make what sounds like a much bolder claim. The view called panpsychism says that all natural systems are intrinsically conscious (at least in some sense). The reason we are conscious is that we are part of a world suffused with consciousness.1
To see the difference between Russell’s view and the more dramatic panpsychist stance, note that the distinctive character of conscious experience has two dimensions: its qualitative character, and its subjective character. The latter dimension refers to the fact that experience is private, not public: it is “for-me”. Both dimensions of experience seem to contribute to what philosopher David Chalmers labels the “hard problem” of reconciling consciousness with the physical world.2
The panpsychist seeks to solve the problem by locating consciousness, including both of its dimensions, in the intrinsic character of all natural systems, including the most fundamental building blocks. In contrast, Russell’s own framework (which can be given the somewhat awkward label of “panqualityism”) asserts that qualities are ubiquitous, but theorizes that the subjective character of experience can be separately accounted through an understanding of the causal structure of conscious entities. One of the attractions of panqualityism is that it can be considered just an enlightened sort of physicalism: one that understands the strengths and limitations of our abstract representations of “physical” systems.
There is a third variation as well, that falls roughly in the middle: panprotopsychism. This view agrees with panqualityism that we should not assert that consciousness per se is ubiquitous a la panpsychism. It sees the intrinsic nature of natural building blocks as somehow providing the ingredients for human-style consciousness but doesn’t stipulate that this only concerns the qualitative dimension, as in the case of panqualityism. The subjective dimension may also have a foundation in the world’s fundamental nature (this is the notion I will look to flesh out in the conclusion of this post).
These approaches all face theoretical hurdles. The primary challenge for panpsychists is to understand how familiar, human-style, conscious experience is composed from or otherwise derived from intrinsically conscious parts: presumably the conscious experience of fundamental particles, for instance, is quite unlike our own. How does a concatenation of these small consciousnesses result in our own?3 This is known as the combination problem. There are actually several different challenges embodied in this problem, but the most acute is what can be termed the “subject-summing” problem: why should adding multiple subjects of experience together result in a new one?4
If panqualityism is correct, the subject-summing problem is avoided. But one still needs to ask: how exactly are conscious subjects (like us) causally constructed? The quick answer just involves pointing to progress in biology, neuroscience, cognitive science, etc.—the sciences will figure it out! Indeed, there is a burgeoning effort to analyze the functions and structures involved in basic subjective awareness, or sentience, in humans and animals. Science will explain how subjects of experience are built: the contribution of “armchair philosophy” is to point out that the distinctive qualitative character of subjective experience is due to the fact that the physical world itself has a qualitative nature.
Does Panqualityism Leave an Explanatory Gap?
While I think panqualityism is a more promising approach than panpsychism, a skeptic may press the case that it still runs into certain difficult conceptual hurdles. For instance, Chalmers (2017) provides a longer list of “combination problems” beyond “subject summing”, and notes that some of these are concerns for panqualityism. They include the quality combination problem (how do the qualities inherent in microphysical entities combine to make up the qualities of familiar conscious experience?) and the structure combination problem (how is it our conscious experience comes to have its characteristic structure?). While these are worthy questions to contemplate, I tend to discount them as reasons for skepticism: our armchair intuitions are simply ill-equipped to grapple with how microphysical systems and their properties create particular macro-sized phenomena (it can seem mysterious that a very limited set of elementary particles can give rise to tables, tigers and tsunamis, but they do).
Panpsychists on the other hand, while they struggle with the subject-summing problem, may also complain that the panqualityist assumption that causal structure can account for subjectivity doesn’t do justice to the distinctive nature of what it means to be a subject: the raw experience of the world from a first-person perspective. How can a causal structure represented by science, no matter how complex, account for this special phenomenon? Here, I think the allegation of an explanatory gap may be better motivated.
Below, I will sketch how my preferred approach to causation may help to address this gap. First, however, I want to also briefly explain why I believe the intuition behind the concerns about the gap can be partly disarmed through study of how simpler forms of biological subjectivity arise in nature.
Subjectivity as a Biological Phenomenon
Here I am influenced by Peter Godfrey-Smith’s work on tracing evolutionary developments that appear important to a discussion of subjectivity.5 Even the simplest organisms are localized, self-maintaining systems that sustain themselves via interactions with their environment across a continuously re-created boundary. One could argue that all life exhibits subjectivity in virtue of the maintenance of this self-other distinction.6 Godfrey-Smith, however, highlights slightly more advanced developments. One of these is the ability of an animal to distinguish those sensations that result from its own movement from other environmental influences. The internal causal feedback loops that make this possible establish a primitive kind of self/other modeling within the organism. The phenomenon of perceptual constancy (why we perceive a penny as round when viewing from an oblique angle) is a sophisticated development of this phenomenon that marks the strongly perspectival dimension of perception in humans and many animals.
In addition to the sensory/perceptual story, Godfrey-Smith highlights another by-product of the demarcation of organisms from environment: the evaluative dimension attached to interactions. The distinction between helpful and harmful interactions is, of course, crucial to survival, and we infer in the case of more sophisticated creatures that these align with the experience of pleasure and pain. An important step for subjectivity may be when animals develop internal systems to register these evaluations and link them to future action through associative learning.
In creatures like us, the brain/nervous system has the capacity to build especially detailed models of the world: models that build upon these simpler kinds of perspectival systems. The most sophisticated type of subjectivity, of course, arrives with our ability to think about the first-order perceptions and feelings that arise in the course of our continual causal interactions. But it seems restrictive to posit that the threshold for subjectivity requires this kind of reflective self-consciousness.
The panqualityist stance does not need to take a position on exactly how far subjectivity extends into the realm of living things. One could maintain it is a feature of life itself, or that it develops in more advanced multi-cellular creatures, or perhaps that it is restricted to humans and animals with fairly complex nervous systems. I am sympathetic towards the more expansive views. But note that all these stances would agree, contra panpsychism, that inanimate natural systems are not subjects.
I think exploring the biological origins of subjectivity helps to demystify the phenomenon. But the story sketched here still associates subjectivity with functions and causal structures. As a result, a stubborn panpsychist still may demand: why should these functions be accompanied by experience?
Deeper Roots in the Nature of Causation
Russell proposed that subjects, like familiar macroscopic objects, are causally constructed. He did not, however, offer a detailed theory of causation. My own account, which was independently motivated by my work on scientific explanation, has some features that may help address this remaining intuitive sense of an explanatory gap.
Briefly, I advocate a modified version of Wesley Salmon’s causal process account (Salmon, 1984). Here the basic entity or object is labeled a causal process, and there are two dimensions of causation: propagation and production. Propagation refers to the evolution of a causal process in the absence of interaction, while production refers to the discrete change that causal processes undergo when an interaction occurs.
A propagating process bears a particular causal influence between locations in spacetime. This influence I characterize (here departing from Salmon) as a cluster of dispositions toward various possible interactions—aka a dispositional profile. The interactions produce a change in this profile.7
Together, causal processes and their interactions form a causal network or web. The entities and properties described by science correspond to features of this web. For example, an electron corresponds to a causal process, and its properties (mass, charge, spin) represent its dispositions to produce change in interactions with other systems.
Importantly, I supplement this basic causal story with an account of composition. A composite causal process is formed when two or more sub-processes repeatedly interact to form a pattern. This pattern of interacting sub-systems gives a composite process its own dispositional profile, accounting for how it will itself interact with other systems of like scale. This account of causal composition enables the theory to describe the causal terrain that underlies phenomena investigated by chemistry and biology, as well as physics.8
So how does this help in thinking about subjectivity?
First, note that every entity in this account is a proto-subject in the following sense: it has an internal character (its dispositional profile) that is altered in every interaction with its environment. This means that while we may correctly limit the scope of true subjectivity to certain complex composites (organisms) whose constituting pattern includes a model of the self/world distinction, all causal processes undergo internal change with every interaction.
The second, more speculative, idea is that subjective experience is rooted in the fact that these internal changes involve a passage from a possibility to its actualization (in the form of a disposition becoming manifest in the change of profile). The idea that change involves this kind of passage or becoming is a very old (though contentious) philosophical notion.9
Finally, I note that experience is also characterized by a kind of autonomy or spontaneity. A possible connection to low-level ontology here lies in noting that for elementary causal processes, as reflected in quantum mechanics, the specific result of a manifestation is undetermined (although there is a constrained set of options): each outcome is spontaneously selected.
To summarize: every causal interaction entails that a process undergoes an internal change; this change involves a passage of a possibility into actuality; and the details of the change are spontaneously realized. These features that support proto-subjectivity are then combined with panqualityism. To accomplish this, one stipulates that (at least some of) the ontological elements also have a qualitative character. This could be the dispositions themselves or perhaps whatever accounts for their hanging together to characterize a process. The resulting (panprotopsychist) metaphysics provides support for the causal/functional picture of conscious experience as a biological phenomenon.
While this venture into metaphysics is, as always, a risky exercise, I do think that the particular route I take has some merit for the following reason. I believe that accounting for the epistemic success of science requires theorizing about the nature of causation.10 If the resulting theory seems to have resources to help with the problem of consciousness, then so much the better.
[UPDATE: March 5, 2022] I thought of another item I might have added to this list of proto-subjective elements inherent in my causal ontology. Dispositions/powers have a kind of directness or aboutness (they are directed toward manifestations). Some philosophers have pointed out the similarity between this feature and the notion of intentionality, which is often seen as a core characterisitic of mental states. For instance, I discussed this in a 2007 post on my old blog about the work of metaphysician George Molnar, who discussed this similarity.
References
Chalmers, D. (1995). “Facing Up to the Problem of Consciousness.” Journal of Consciousness Studies, 2(3):200-19.
Chalmers, D. (2017). “The Combination Problem for Panpsychism.” In (G. Bruntrup and L. Jaskolla, eds.) Panpsychism. Oxford University Press, 2017
Godfrey-Smith, P. (2019). “Evolving Across the Explanatory Gap,” Philosophy, Theory and Practice in Biology, 11:1.
Goff, P. Consciousness and Fundamental Reality (2017). Oxford: Oxford University Press.
James. W. (1890). The Principles of Psychology. New York: Holt.
Kriegel, U. (2005). “Naturalizing Subjective Character.” Philosophy and Phenomenological Research, 71:1.
Martin, C. (2008). The Mind in Nature. Oxford: Oxford University Press.
Russell, B. (1927). The Analysis of Matter. Nottingham: Spokesman.
Salmon, W. C. (1984). Scientific Explanation and the Causal Structure of the World. Princeton: Princeton University Press.
Thompson E. (2007). Mind in Life. Cambridge: Harvard University Press.
1 A prominent and prolific recent defender of panpsychism is philosopher Philip Goff.
2 I believe I first encountered the idea that these two dimensions of consciousness should be addressed separately in the work of philosopher Uriah Kriegel (see here). For Chalmer’s seminal paper on the hard problem, see here.
3 There is a version of panpsychism called cosmopsychism that reverses the direction of scale and asserts that our own consciousness is derived from a universe-level consciousness (Philip Goff defends this view in his 2017 book, Consciousness and Fundamental Reality. I won’t discuss this idea here. I note that Goff’s view, while still featuring cosmopsychism, has changed somewhat: in a recent manuscript he presents his own version of a “divide and conquer” approach, utilizing a hybrid approach to dealing with the qualitative (or phenomenal, to use his term) and subjective dimensions of consciousness.
4 This problem was famously discussed by William James in his Principles of Psychology (1890).
5 For a recent Godfrey-Smith paper on this topic see “Evolving Across the Explanatory Gap,” (a related paper is “Varieties of Subjectivity”).
6 I’m reminded here of Evan Thompson’s book Mind in Life (2007), which argues for a “deep continuity” between the phenomena of life and mind. [UPDATE: March 13, 2002] A new paper from Thompson explores the issues surrounding the question of whether all life is sentient.
7 This change is the mutual manifestation of the respective dispositions of the interacting processes. The idea of mutual manifestation (among “reciprocal disposition partners”) was proposed by philosopher C.B. Martin (see his 2008, especially Ch. 5).
8 I have a series of blog posts that discuss aspects of this in greater detail. A take connecting the causal theory to quantum physics is here; a overview of the theory of causal composition is here; and a deeper dive into molecular composition is here. Importantly, a causal approach like this avoids problems with the concept of emergence that plague non-causal (synchronic) approaches to composition.
9 One particular challenge is that it is tricky to connect this idea to more familiar temporal concepts we use when describing experience. While I have ideas about this, I’ll defer discussion to another time.
10 This is the topic of two recent blog posts here and here.
1 comment