Daniel D., HUTTO, Beyond Physicalism, (Advances in Consciousness Research), John Benjamins: Amsterdam/Philadelphia, 2000.Beyond Physicalism attempts to establish the nature of the limits to our understanding of basic experiences and to stake out the consequences these have for an appropriate metaphysics. The argument of the book advances in three stages which, for the purposes of this précis, I label Divisions I, II and III. I argue that two conceptual problems concerning consciousness must be distinguished. The first concerns the correct way in which to characterise the nature of consciousness itself - let us call this the 'phenomenology problem'. But, I also claim standard attempts to address this problem are constrained by their focus on a second problem, which I dub the 'metaphysical problem'. Solving this latter problem essentially requires us to provide an intelligible representation of the relation between the mental and the physical. Crucially, its very formulation links it to traditional physicalist conceptions of basic ontology. My claim is that it is not a problem to be solved - but one to be avoided. I propose that it dissolves if we reform our assumptions concerning what is natural and rethink our orthodox metaphysics. Nor is such a move ad hoc, as there are independent reasons for doing so. To this end I defend a form of absolute idealism that permits pluralism but which does not collapse into mere conceptual or linguistic idealism. I attempt to show that the version of absolute idealism I am interested in is not a 'dead' option by establishing that it is not at odds with modern science and that its central claims have not been already successfully refuted at the turn of the twentieth century.
Chapter by Chapter Synopsis
Division I
Chapter 1: Nonconceptual Experience
In this chapter I defend the view that basic experiences exist and that they are nonconceptual, against those who would dismiss them altogether or who attempt to equate them with linguistic or conceptual reports. I do not concern myself with higher-order forms of consciousness, such as self-consciousness for
. I believe that these require an awareness of awareness that in turn requires a conceptual capacities. Instead, drawing lessons from a broadly Davidsonian account of interpretation, I argue that the development and use of concepts presupposes the existence of basic forms of experience. To learn concepts it is necessary, at least in a range of simple cases, to be able to respond to a common world in ways that are similar to that of others and to be able to recognise the similarity of response in others. But, as these capacities are required for the development of a conceptual point of view, I argue that at root they must be nonconceptual in character. Hence, at bottom, I propose that joint attention should be understood as requiring simulative abilities to engage practically and emotionally with the situations of others and the objects of their concern and to recognise that others are similarly engaged with our situations and concerns. Yet such simulative capacities do not presuppose any conceptual abilities. On this basis I argue that mere informational sensitivity, no matter how complex, is not shareable in the appropriate way and hence it cannot play the role of experience in grounding concept learning and development... If it is accepted that our concepts of experience are not innate then it is possible to mount an argument, distinct from mere appeals to introspection, in order to establish that causally efficacious, but non-cognitive experiences exist. This has the further consequences that if conscious experiences are at root non-conceptual then, although they can form basis for linguistic expressions and 'reports' they cannot be equated with them.
Chapter 2: From a Sensational Point of View
In this chapter I consider some of the consequences of the thesis that experience is, at bottom, nonconceptual in character. It follows that if the qualitative character of experience is nonconceptual, then to the extent that Nagel, or Jackson, argue that knowing-what-it-is-like to undergo an experience imparts a special kind of factual phenomenal knowledge, they are wrong. A fortiori there can be no such knowledge. Talk of it is strictly nonsensical. Yet this technicality should not comfort physicalist overly much, for the real force of the Nagel-Jackson critique still has some punch. Their thought-experiments reveal that neither physicalists, nor anyone else, can give an account of experiences, which can capture or describe their nature. No such account is possible. While this does not establish that physicalism is flawed because it is factually incomplete, it does reveal its limits. To appreciate what-it-is-like for X to have an experience, we must actually undergo the same kind of experiences as X. This requires transformation, not knowledge per se. For this reason there is no grasping what-it-is-like in the case of those whose experiences we cannot simulate. Consequently, even though it is wrong to talk of experience as imparting a special kind of knowledge, Nagel's conclusion still stands. He is right to think that there is an inexplicable residue after the process of seeing is fully characterised, functionally and neurologically speaking. This residue is the qualitative character of experience. This is important, for if theorists of consciousness aim to provide an alternative means of understanding this aspect of experience they are doomed to fail.
But, is this their goal? After all, why would theorists of consciousness seek to tell us about the qualitative aspects of experiences? Theories generally tend to abstract away from qualities in their quest for underlying structures and causes. Consequently, it is still prima facie possible to develop a theory of consciousness, in the metaphysical and explanatory sense. This would be a theory designed to tell us about the essential physical and causal nature of experiences. This would be achieved if we could isolate a class, or sub-class, of phenomena, which would be identical to, or would serve as indices for, the class of conscious experiences. If this reducing class already fits with our established ontology, then we would have produced a theory of the metaphysics of consciousness. The claim that this is a sensible occupation is the pretext for advocating representational, functional, neurological and quantum mechanical theories of consciousness. Given this, it appears as if there is an important philosophical task involved in assessing the virtues and vices of the different metaphysical theories on offer. However, using the strong representational thesis as a foil, I highlight the key feature that frustrates any and all attempts to provide a theory of metaphysics of consciousness.
One internal problem with the representationalist approach, which proposes that experiences can be equated with representational contents, concerns the nature of representations themselves. For, only if we think of representations as functionally or dispositionally defined creature-relative responses, is the strong representational thesis plausible. Yet, for this very reason, we can see why it must be false. This is because representations cannot be understood in purely functional or dispositional terms. For if experiences are functionally defined by appeal to how a creature is disposed to respond, then they cannot to be understood in terms of how the creature is supposed to respond. If we want to avoid the problems of objectifying experience and instead maintain that experiences are creature-relative, or even subject-relative, responses which typically influence actions, then we are forced to distinguish intentionality and experience.
The more general problem is that strong representationalism fails to do justice to the subjective character of experience. It is a crowning virtue of the best accounts of representation that in order to avoid an infinite regress of interpreters they avoid postulating any internal interpreters in the first place. But while this is a good move when it comes to understanding intentionality, removing the subject is retrograde when it comes to explicating the nature of experience. For in order to understand experience it is necessary to bring the subject in on the act. Thus, representationalists - like all other theorists about consciousness - face the following dilemma; either accept Frege's thesis, according to which there cannot be experience without someone to experience it, or deny it. If they accept it, then they owe us an account of the nature of the experiencing subject. Furthermore, this account must be compatible with their favored theory, even if the subject doesn't play anything other than a supporting role in it. On the other hand, if they deny Frege's thesis, it is unclear how they can make sense of experiences at all. The point is that once we cleanly distinguish experience from intentionality, we are in good position to understand why the subjective aspect of the former foils attempts to intelligibly explain the place of consciousness in the material world. That is not to deny that experiences are real and causally efficacious and. Yet if they are, the need to provide a metaphysical and explanatory account of them becomes even more pressing. But I claim that physicalism is not up to this task.
Division II
Chapter 3: The Failure of Explanatory Physicalism
Although I am committed to the view that there are nonconceptual conscious experiences and that they make a difference I do not endorse functionalism. It is perfectly possible to accept that experiences have causal aspects or roles, as long as they are regarded as defining ones. In chapter three, in particular, I set out my reasons for rejecting any kind of functionalism that is wedded to some form of explanatory physicalism. The virtues of functionalism are reviewed on two fronts. One concerns the why-question, which asks 'Why is there conscious experience at all?'. The other concerns the how-question, which asks: 'How do the mental and physical inter-relate?'.
With respect to the first question, I argue that abstract versions of functionalism are inadequate, particularly when they are compared to biologically inspired homuncular teleofunctionalism. Indeed, because the latter might provide some insight into the possible origins of experience, I claim it provides the right kind of reply to the why-formulation of the hard question. Nevertheless no functionalists, who take the existence of experience seriously, be they of the abstract or homuncular persuasion, can adequately address the how-question by endorsing some form of explanatory materialism or physicalism. On the one hand, although abstract functionalists are ontologically non-committal in the first instance, their ontological openness leaves them without the internal resources to address the how-question. Thus it is standardly hoped that it will be possible to graft abstract functionalism onto some explanatory form of physicalism which would provide an intelligible account of the relation between the experiential and the physical. On the other hand, for those who treat experiences realistically there is no interesting distinction between homuncular teleofunctionalism and physicalism. As Lycan notes, his version of functionalism is in fact just a special instance of the psychophysical identity theory.
This leads us to ask; what, then, is the matter with physicalism? At base, physicalism is an ontological doctrine which holds all genuine existing things to be physical in nature, or in some way depend upon, or are generated by, or are composed of physical events, properties and their inter-relations. But it can also be characterised as a doctrine concerning the structure of knowledge. Although these two construals need not be at odds, it is crucial to distinguish those versions of physicalism that see it only as an ontological thesis from knowledge-based versions which accept that it carries an explanatory burden, however mild. In short, explanatory versions are concerned with issues at the level of sense, as well as those of ontology. With this distinction in hand, I use the remains of chapter three, to argue that explanatory physicalism, in both its reductive and non-reductive forms, fails to provide an adequate metaphysical basis for consciousness. I reserve comment on purely ontological versions of physicalism until chapter five.
The most straightforward way of making physicalism plausible would be to produce an adequate theory of consciousness that could then be incorporated unproblematically into a more basic physical theory. This style of approach best suits the methodological ideals of reductionists. But, if the arguments of the first two chapters are sound, then no such theory of consciousness is on the cards and so neither is a transparent account of psychophysical relations. The problem is that without such a theory we have no good reason to accept the psychophysical identity claim on explanatory grounds. Put bluntly, merely stipulating that Y and F are identical, even if true, provides no understanding of psychophysical relations since understanding must occur at the level of sense. Thus although standard replies to Leibniz Law objections are correct in claiming that intensional inequivalence does not rule out the possibility of extensional equivalence, they also highlight the very reason why standard identity theories are not plausible when advanced under an explanatory banner.
Nor do eliminative materialists fare better on this front. For experiences cannot be simply explained away. If one admits that experiences appear to exist, even as modes of presentation, then they must be accounted for. We are owed an explanation of what accounts for the fact that experiences appear to be as they are. Specifically, we must account for their apparent qualitative aspects. But this is tantamount to explaining experiences themselves. Thus, while eliminativism may be a successful strategy in other domains, to say that experiences are mere appearances hence not real does not reduce the materialist's explanatory burden.
In contrast, the problem with non-reductive physicalism is not that it fails to recognise that conscious experiences have unique features, but rather, in taking them seriously, its defenders systematically abuse the notion of the physical. Although increasingly popular, non-reductionism offers nothing by way of a serious resolution of the metaphysical problem. To do so it would have to explain how the non-reductive notions of composition, realisation or emergence are at peace with physicalism, as opposed to supplying a positive reason for its abandonment. But to the extent that non-reductive approaches refuse to compromise with respect to subjective character of experience, they necessarily fail to provide a coherent account of the physical. Standardly, defenders of non-reductionism press for a distinction between the epistemic and the metaphysical and on this basis advance a version of metaphysical physicalism. But on such accounts, the physical is robbed of any possible meaning as there are no principled boundaries, such as those established by the reductionists, by which to decide which phenomena are genuinely physical. In short, not only are all explanatory commitments shirked, one literally does not know what it means to be a physicalist.
Chapter 4: Intelligibility and Conceptual Limits
What becomes clear from reviewing reductive and non-reductive versions of explanatory physicalism is that the metaphysical problem only presents itself against certain background assumptions about what counts as 'the natural'. Indeed, the problem arises against the background of a certain entrenched view of metaphysics sponsored by the idea that the physical (in one form or another) describes the ultimately real. In chapter four, I offer a diagnosis of the assumptions that make the metaphysical problem intractable. My conclusion is that the tension arises because experience cannot be understood in terms of an object-based schema to which physicalism is implicitly committed. This is the real source of the intelligibility problem.
I begin by considering the dilemma, posed by Crane and Mellor, according to which physicalism is either a false or empty doctrine, depending on whether or not it defined in terms of a current or future physics. By way of reply, I argue that the doctrine can be given a non-vacuous boundary if we regard it as committed to the object-based schema. That is to say, if we define it as a science that has developed from and which applies to the class of objects which occupy space and time. This way of formulating it anchors its concern to a particular domain without foreclosing entirely on the possibility of developments in physical theory, within certain limits. Looking at things in this light also helps us to understand why inter-theoretical reduction is possible in certain cases. For example, the relation between classical physics and chemistry is intelligible because the entities of both sciences can be successfully modeled using an object-based schema. Talk of atoms, molecules, and the relations that hold between them are all part of a common schema that makes reference to spatio-temporal locations.
Conversely, the problem of intelligibility that arises in our attempts to understand quantum physics stems from the fact that it breaks faith with this kind of schema. In this respect its metaphysical problem has the same roots as that which concerns consciousness. In both cases, in order to overcome the tensions, it seems necessary to model the respective 'entities' in terms of an object-based schema. But doing so only generates philosophical confusion and nonsense. In the case of consciousness, such attempts only lead to ill-posed questions about the location of consciousness and phenomenal space and they mislead us into speaking, inappropriately, of conscious 'properties' and 'states'. To avoid such difficulties I argue that we must take seriously the idea that consciousness is not what is experienced, but the medium through which we experience things, making it clear that this requires presentation to a subject.
Chapter 5: Pluralistic Naturalism and Absolute Idealism
In light of the arguments of chapter three and the diagnosis of chapter four, it is claimed that the most stable and defensible version of physicalism is one which is only committed to an identity thesis and the kind of dependency relations required by some unsupplemented covariance thesis. However, in absence of an explanatory theory, why should we adopt such a minimalist ontological physicalism? The standard reply is an argument from causation. It states that if consciousness really does influence the physical, and vice versa, then it must be physical in nature. This is the fast road to psychophysical identity. But this does not help. Minimal physicalists, unlike their explanatorily ambitious cousins, accept that psychophysical connections have no prospect of being made intelligible. Given the incommensurable nature of the two domains in question, we cannot hope for a reduction. Nor will it do to appeal to the extensionality of causal relations in order to get us to look for this explanation under a different description - e.g., in terms of physical-physical causation. This would not explain what needs explaining. The point is that the extensional manoeuvre does nothing to clear the air of the mystery that hangs over psychophysical causation. This is hardly surprising since a suitably cautious minimal physicalist makes no attempt to resolve the intelligibility problem. To solve this problem would require an explanatory theory of consciousness. The point is that postulating identities makes no advance in this regard.
What then, if anything, is the advantage of endorsing minimal physicalism, as opposed to substance dualism? Minimal physicalism; is attractive, not because of explanatory superiority, but for its metaphysical economy. The fact that it is monistic enables us to postulate causal closure even if we cannot fully comprehend the nature of causal interaction in extension. But this raises the question: If monism is all we are after, what entitles us to endorse the physicalist claim that physics, or an ideal physics, provides a privileged, true picture of the universe in all its nakedness? The minimal physicalist is faced with the following dilemma. On the one hand, if they hold that the world in extension is nothing but the world described by an ideal physics, then how can consciousness be a genuine phenomenon? On the other hand, if we reject the claim that the world in extension can be correctly characterised by any one discourse, then what is 'physicalist' about the token identity thesis? Indeed, to drop the claim that ideal physical descriptions are privileged when it comes to describing reality is, in effect, to endorse a version of absolute idealism.
With this analysis in hand, I argue there are at least two good reasons to prefer absolute idealism over a minimal physicalism. First it does not commit us to the problematic view that reality in extension can be rightly characterised in physical terms. It shares this important feature with quietism, which asks us to reject any deeper metaphysical projects. Secondly, it has the aesthetic virtue of postulating a truly unified metaphysics. Ironically, it is this latter criterion that some materialists appeal to when they wish to make a case against dualism.
Division III
Chapter 6: Defending Absolute Idealism
In chapter five, I argue that we can best deal with the metaphysical problem by adopting a modified version of Bradley's form of absolute idealism, which respects the autonomy and diversity of differing conceptual schema while retaining an unbiased commitment to a unified metaphysics. Absolute idealists claim that there is no possibility of making intelligible the relations between various schema, theories or sciences, such that a single discourse could ever completely describe reality. They recognise the need for plurality at the level of explanation and sense; thus they make no attempt to explicate the metaphysical unity which they postulate. Accordingly, there is no possibility that an ideal physics could completely describe the world as it is.
However I anticipate that many philosophers will baulk at accepting this proposal because they either think that absolute idealism has already been refuted at the turn of the century or that it is incompatible with our best views about science. In chapter six I deal quickly with the first charge and devote my efforts to dealing with an aspect of the latter one. I argue that the idea that absolute idealism is anti-scientific is at best contentious and at worst false. Although it is clearly the case that absolute idealism is at odds with a science-based metaphysics, it doesn't follow that it is at odds with our best understanding of science. That is to say, it is at odds with the view that science and metaphysics are necessarily interwoven and hence it rejects the vision of science inspired by analytical philosophy, which sees it as providing a complete, true picture of reality. By drawing on examples from biology and quantum physics I make a case against the idea that good science necessarily strives for, or can attain, completeness. If this analysis is correct then we cannot expect science to provide a unified basis for metaphysics.
Chapter 7: Truth and the Whole Truth
Having argued that absolute idealism is not at odds with a proper characterisation of the scientific enterprise, with respect to issues of unity and completeness, I conclude by considering its position on the nature of truth and, relatedly, its prospects for making sense of scientific progress. I claim that the absolute idealist view of truth is largely consistent with our best understanding of truth and that it provides a sound basis for a plausible and sophisticated version of scientific realism. In making this case I begin by analysing the role truth is supposed to play in our understanding of the scientific enterprise.
It is hardly news that lying at the heart of traditional scientific realism, that is realism with respect to theories as opposed to entities, is a commitment to some form of correspondence theory of truth. But the idea of correspondence can be fleshed out in a variety of ways, some more modest than others. I argue that the strongest form of theoretical realism, which are committed to the discursive picture of reality, according to which a final truth is what our best science will one day reveal, are unworkable. However, along with many other contemporary philosophers of science, I agree that if truth is understood against an ingredient in the interpretative process, then a modest version of scientific realism can be successfully advanced. In this regard, I see Davidson as a contemporary champion of the cause of Bradleyian relative truths or, as I prefer to call them, contextual truths. Such truths include the truths of science. The crucial point of difference between these two thinkers is that Davidson shuns all attempts to analyse truth, whereas Bradley, although he too will say nothing substantive about it, identifies it with reality itself. This move gives truth the potential to play an important metaphysical role as an independent norm, which can both motivate inquiry and set a limit to it. Furthermore, in conjunction with this kind of modest theoretical realism, it is possible to endorse a version of entity realism that is needed in to help account for the successes of modern science, since these stem in part from our experimental interactions with the world.
Ultimately, the position defended is suitably complex. It sees us anchored to the world nonconceptually, by our senses and experimental techniques, but recognises that in order to have understanding, however imperfect, we need social standards, dialogue and debate. Initially, when learning our concepts - scientific or otherwise - we inherit a picture of reality but it develops and changes as we question and probe it. Part of what constrains theoretic change are our nonconceptual engagements with reality. But there are other constraints such as the requirement that, when and where possible, our theories ought to fit together, and at the very least they should not contradict one another. This is a goal, but it is not always or ultimately achievable.
We are the measure of all things but, practically speaking, that which we measure is surely independent of us. Yet given the nature of our measuring tools, be they perceptual or conceptual, we can only measure things imperfectly. Imperfectly does not mean uselessly. The world is real enough, with or without science - but prior to the development of a language and the capacity for conceptual abstraction - aspects of it cannot be said to be true or false. This does not mean that there is no theoretical progress to be had in creating fertile scientific metaphors or that we don't get better at successfully engaging with aspects of reality by means of technology. But there is no final, complete, objective truth to be told about reality. Not even one stateable in the longest imaginable proposition. Consequently, not even an ideal physics can discursively characterise the world as it is in extension. This being so, we must look beyond physicalism to a metaphysics that permits both tolerance and unity.
©2000 Daniel D. Hutto - Published in SWIF Philosophy of Mind - ©2000 Luca Malatesti