Shall we tango. No, but thanks for asking

Citation preview

Daniel C. Dennett

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Shall We Tango? No, but Thanks for Asking I have learned a lot from Evan Thompson’s book — his scholarship is formidable, and his taste for relatively overlooked thinkers is admirable — but I keep stumbling over the strain induced by his selfassigned task of demonstrating that his heroes — Varela and Maturana, Merleau-Ponty and (now) Husserl, Oyama and Moss and others — have shattered the comfortable assumptions of orthodoxy, and outlined radical new approaches to the puzzles of life and mind. The irony is that Thompson is such a clear and conscientious expositor that he makes it much easier for me to see that the ideas he expounds, while often truly excellent, are not really all that revolutionary, but, at best, valuable correctives to the sorts of oversimplifications that tend to get turned into mantras by sheer repetition in the textbooks and popular accounts of these topics in the media. Philosophers have a delicate task: squeezing the tacit assumptions and unnoticed implications out of every ill-considered dogma without lapsing into nitpicking or caricature. Thompson does better than most; he is not a gotcha!-monger or sea lawyer, but he does set up a few strawmen (strawpersons?) which I will duly expose as such, while showing that his revolutionaries are not really so revolutionary after all. Reformers are the bane of would-be rebels, of course, taking the wind out of their sails just as they get started, and in the cases I will discuss, reform-minded critics — myself among them — have already pointed out the caveats that pre-empt these assaults on orthodoxy. Might these caveats and concessions be mere lip service? Have the reformers underestimated the seriousness of the challenges, papering over the cracks that will in due course bring down their edifice? Correspondence: Email: [email protected]

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 23–34

24

D.C. DENNETT

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Perhaps, but that case remains to be made, and it must begin by acknowledging that the problems have not gone unnoticed, just — at worst — underestimated. I will concentrate on four themes in the book, since they all involve my own work in one way or another: 1.

autopoiesis as a radical new foundation for evolutionary theory (in contrast to the ‘Ultra-Darwinism’ he attributes to Dawkins, Maynard Smith and me),

2.

developmental systems as an alternative to ‘genocentrism’, adaptationism and my design stance,

3.

the ‘autonomous’ view of semantic information (as contrasted with my intentional stance, among others),

4.

the first-person methodology he opposes to my heterophenomenology.

(Thompson also discusses my role in the mental imagery debate between Kosslyn, Pylyshyn and others, but in the interests of brevity, I will postpone comments on that for another occasion.)

1. Autopoiesis Is there a ‘theory of autopoiesis’ (p. 97),1 or is it just a good way to think about the requirements of life? I’ve been struggling with the idea of autopoiesis off and on for many years, since first meeting Umberto Maturana and Francisco Varela. Thompson succeeds in making the idea of autopoiesis as clear as anybody could want. The upshot, however, is that autopoiesis turns out to be, well, the cell theory, made a little deeper and more general by being divorced from the particularities of biochemistry and reduced to its simplest specifications. A similar — indeed virtually synonymous — approach is the chemoton proposed by Tibor Gánti (1971, but available in English only since 2003), the minimal functional/structural requirements for a living, metabolizing, reproducing organism (autopoiesis treats reproduction as an option, not a necessary condition, of life). I don’t see anything in autopoiesis as limned by Thompson that would upset — or particularly inform — a mainstream cell biologist, aside from the fancy language in which the specs are couched. As Thompson says, ‘The paradigm [of an autopoietic system] is a living cell’ (p. 44). The cell theory is, so far as I can tell, good biology indeed, so I have no [1]

Unless otherwise indicated, all page numbers in the text refer to Thompson (2007a).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

SHALL WE TANGO? NO, BUT THANKS FOR ASKING

25

quarrel with it at all. In fact I rather like some of the formulations due to Maturana and Varela, and some of Thompson’s as well, such as ‘a cell stands out of a molecular soup by creating the boundaries that set it apart from what it is not’ (p. 99), which nicely draws attention to the fact that the concentration of large molecules outside a cell is not all that different from that inside. The cell’s boundary is its most salient and robust feature but there are exceptions to Thompson’s claim: the boundary is, as everybody notes, semi-permeable, and which things count as inside and which as outside is not always clear. There are transition zones, and besides, is something inside it food, waste, or a proper part? If it too is an autopoietic system, is it an invader or a symbiotic ally? So far as I can see, autopoiesis is an excellent summary of what it takes for a collection of molecules to be alive, but it doesn’t predict anything in biology that hadn’t already been well understood by earlier theorists, or dissolve any puzzles that had been bedevilling those theorists. Thompson clearly disagrees with this bland verdict, and articulates his disagreement with a series of what I will call ratherings. A rathering is a rhetorical move much beloved by some of the other ideologues of biology, such as Steven Rose and the late Stephen Jay Gould. The general form of a rathering is ‘it is not the case that blahblahblah, as orthodoxy would have you believe; it is rather that suchandsuchandsuch — which is radically different’. Some — not all — ratherings are little more than sleight-of-hand, since the word ‘rather’ implies — without argument — that there is an important incompatibility between the claims flanking it. I counted over a dozen ratherings, and six of them occur on two adjacent pages, 206 and 207. Sometimes Thompson eschews the word ‘rather’ and settles for the slightly terser ‘not… but…’ idiom. A case in point: Such systems need to be seen as sources of their own activity, specifying their own domains of interaction, not as transducers or functions for converting input instructions into output products. (p. 46)

Why can’t they be seen as both? And don’t we need a small loophole to allow that autopoietic systems specify their own domains of interaction when all is going well? (A similar concern affects Varela’s ‘Closure Thesis’ discussed on p. 48.) I like the recognition by the autopoiesis view that replication (e.g. DNA replication) could be an optional add-on. After all, mules are certainly alive and cannot replicate. Interestingly, mules can be seen as a curious sort of ecto-parasite, dependent on us (on mule-fanciers, more particularly) to maintain their numbers. Other kinds of

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

26

D.C. DENNETT

intermediate cases could also be listed. In short, we see many variations on (or exceptions to) autopoiesis in nature. Chemotons are autopoietic systems with replication. Autopoietic systems are viruses with a metabolism (cf. p. 104). Just as a hermit crab can acquire its shell instead of ‘self-producing’ it, a proto-living form can acquire its membrane in a similar way (this is a currently well-regarded conjecture about the origin of life, discussed by Thompson on pp. 113ff). Is Gaia an autopoietic system? As Thompson notes (pp. 121ff), this isn’t clear. Autopoietic systems, as defined by Thompson (following Maturana and Varela), are perhaps owed pride of place in the cornucopia of life because of their ‘autonomy’, but even this feature is, as Thompson notes, ‘a “heuristic notion” — so whether a system is autonomous is context-dependent and interest-relative’ (p. 50). And it is worth noting that non-autonomous systems often display key features of life and cognition. For instance, my intentional systems are predictable and explicable from the intentional stance whether or not they are autonomous systems, or proper parts of autonomous systems, or fully heteronomous. Even if one granted that autopoiesis is the most apt definition of the property of life, this would not establish that only autopoietic systems can be the proper (literal, underived, etc.) bearers of various biological predicates. After all, viruses are not alive, are not autopoietic systems, and yet the theory of evolution by natural selection applies as truly and unmetaphorically to them as to any living thing. Recently severed limbs, and transplantable kidneys and hearts, are living, but not themselves autopoietic systems, and so forth. Then there is the major question of whether autopoiesis provides a genuine alternative to standard neo-Darwinism (or ‘Ultra-Darwinism’, the term invented by Stephen Jay Gould — along with ‘Darwinian Fundamentalism’ — to try to reposition his own misgivings in the centre of the field; a term adopted, I’m sorry to say, by Thompson). Here the ratherings come in droves: ‘natural selection is not an external force but the differential propagation of developmental systems’ (p. 202). An ‘external’ force? Thompson repeats this charge at least three times, but just what mistake is it, and what evidence is there that anybody ever makes it? He quotes a rathering from Levins and Lewontin: adaptationism ‘implies that the organism is simply a passive object of selection rather than an active agent or subject of the evolutionary process’ (Levins and Lewontin, 1985). How does this implication run, and does anybody believe it? Thompson dismisses Dawkins’ concept of arms races, one of the most predictively fruitful insights in evolutionary biology, as ‘merely a questionable metaphor’

SHALL WE TANGO? NO, BUT THANKS FOR ASKING

27

(p. 205). This dismissal would be more persuasive if Thompson could show us a single instance in which autopoiesis (or developmental systems theory) has predicted or explained biological effects on a similar scale. Finally, Thompson says, regarding my claims about biology as reverse engineering:

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

To move from the claim that organisms can be interpreted from a reverse engineering stance to the claim that they are artifacts of design is to confuse a particular heuristic or interpretive framework with the phenomena themselves. (p. 211)

Since I have gone to considerable lengths over the years to show how ‘design-without-a-designer’ is no more a contradiction in terms than ‘splittable atom’, and since I have demonstrated over and over how the design stance works exactly as well for organisms and their parts and behaviours as it does for artefacts, I find this charge of elementary confusion on my part, presented with no support at all, simply incredible. Consider that Gánti and Maturana arrive at similar models for similar reasons: Gánti began his career as a chemical engineer and is explicit about his reverse-engineering perspective in chemoton theory; Maturana does not highlight the reverse-engineering in his approach, but Thompson’s own excellent clarification of the idea of autopoiesis nicely reveals its functionalistic rationales.

2. Developmental Systems The claim here is that ‘Ultra-Darwinism’ with its ‘genocentrism’ must be ousted and replaced by an ‘enactive’ vision of biology — and evolution — that is profoundly unlike orthodox neo-Darwinism. Thompson is not the only thinker tempted to declare that ‘evo-devo’ overthrows — rather than enlarging and repairing — the neo-Darwinian synthesis, but like the others he relies on something of a caricature of the genocentrism he seeks to banish. In particular I am unimpressed by the claims he makes about the ‘dualism of hardware and software’ which he imputes to genocentrism (e.g. pp. 174, 185), and ‘the myth of the gene as a unit of pure information’ (p. 179). First of all, I take it that he is not claiming that the hardware/software dualism of computer science is dualism of any objectionable (e.g. Cartesian!) kind. It is about as secure and useful as any ‘dualism’ in science (e.g. matter and anti-matter, or — closer to home — living and non-living). So it must be something about the way this useful and well-understood dualism of computer science is (mis-?)applied to genes that is the problem, but Thompson never spells this claim out. Some sciencephobes have used the epithet

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

28

D.C. DENNETT

‘dualism’ somewhat ironically (one supposes) to attack any science that uses the concept of information, as if the criticisms of Descartes’ brand of substance dualism could be somehow harnessed to their campaign against the use of computer science concepts in the humanities! I hope Thompson is not following their lead, but then I don’t see what he is saying. There seems to me to be a very clear and unobjectionable way in which we can draw the software/hardware distinction when discussing genes: the DNA molecule has an architecture that is ideally suited for enabling the replication of sequences of codons. These sequences, like software, are composed from a finite alphabet of salient and discriminable units, with the information carried by the particular sequence of codons so that scrambling them destroys the information. (The same is of course true of texts: shredders destroy the usable information in them by making the sequences unrecoverable.) Moreover, to a first approximation, thanks to the brilliant work of the molecular biologists since Crick and Watson set this investigation in motion, we can now show how these sequences, when fed into the machinery — the ribosomes, mainly — within the cell, yield proteins (and other effects). Like so many products spewing out of a CAD-CAM system in response to the specification — the information — provided as input. No whiff of objectionable dualism can be discerned here, I think. As for the ‘myth’ of the gene as a ‘unit of pure information’, I wonder if anybody has ever subscribed to that myth. Here is another rathering: Information is not intrinsic to the linear array of the DNA sequence. Rather, it is constituted in and by the cell as an autopoietically organized, three-dimensional entity — by the cell as a body. (p. 182)

Presumably Thompson would agree with this as well: Information is not intrinsic to the linear array of the letter sequence in a book. Rather, it is constituted in and by the reader as a cognitively organized, three-dimensional entity — by the reader as a body.

I take it everybody recognizes that the sentence ‘Snow is white’ only carries information about the colour of frozen precipitation on the assumption of the whole world of English speakers and readers. It carries no information at all intrinsically. Similarly, genes carry information only within the larger system of gene-readers, gene-recipeexecutors, etc., etc. I doubt that any ‘genocentrist’ has ever thought otherwise. No myth here. Thompson cites a passage of mine making roughly this point, but goes on to claim that I cannot draw a ‘principled line… between explicit (coded) and implicit (uncoded) information’ (p.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

SHALL WE TANGO? NO, BUT THANKS FOR ASKING

29

184). Quite frankly I don’t see the problem, and have gone on at considerable length about the difference in various places. I may be wrong, but Thompson doesn’t address my claims, relying instead on a passage quoted from Oyama, itself a series of non-sequiturs. Thompson says ‘It is simply not true that genes are prime-movers and cells their vehicles’, citing Moss (2003) as his authority. Moss’s argument isn’t given, and although I am sure there are flat-footed senses in which this is ‘simply not true’, there are others where it is not at all obvious that this isn’t true. We know that tiny changes in genes can yield huge downstream effects in population characteristics, ecological dynamics, large morphological changes (compare two branching lineages from a common mammalian ancestor: ungulates with hooves and whales), while similar changes in somatic line cells seldom result in amplified differences. Yes, ‘molecular biology can lose sight of the organism as a whole’ (p. 93), as Rose and Goodwin have argued, but while nobody can doubt that this is often true, few would agree with Rose or Goodwin that this is a disqualifying flaw in neoDarwinism. In his campaign against orthodoxy, Thompson cites a veritable Hall of Fame of the would-be revolutionaries of biology, from such warhorses as Levins and Lewontin, Margulis and Sagan, and Lovelock, to Kauffman, Keller, Oyama and Moss. This is a rhetorical strategy that can backfire. It caused me to conclude that if, throwing the kitchen sink at orthodox neo-Darwinism, this is the best Thompson can come up with, neo-Darwinism is in pretty good shape. Not all critics of neo-Darwinism are equal, and a more persuasive argument would concentrate on one or two main points and develop them in detail. Maybe one of Thompson’s radical biologists will prove, in the long run, to be right, but they haven’t individually made much of a dent on orthodoxy so far, and pooling them into chorus doesn’t persuade me, and won’t persuade many others, even if it ought to.

3. Autonomous Meaning-Construction Thompson insists on ‘the distinction between autonomous meaning-construction and heteronomous information processing’ (p. 54), but once again, I think we have a case of rathering here: From an autonomy perspective, individual neurons do not detect objectively defined features. Rather, assemblies of neurons make sense of stimulations by constructing meaning, and this meaning arises as a function of how the brain’s endogenous and nonlinear activity compensates for sensory perturbations. (p. 53)

Why not both? Thompson seems to think that the autonomy perspective permits the theorist to allow the neurons to do the sense-making,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

30

D.C. DENNETT

while in all versions of the alternative perspective ‘an observer or a designer stands outside the system and states what is to count as information’ (p. 52). This is reminiscent of Gerald Edelman’s (1989) insistence that his neural Darwinism does not count as a computational or AI model because the meaning of the (semantic) information in his simulations is not given in advance by the world, or by the modeller. But this is a red herring. Indeed, most AI systems devised so far have been built-to-order to represent certain states of affairs in the world defined by the system’s users, but this is obviously an optional matter. If an AI is designed (as some are, such as Edelman’s, but not only his) to muck about in the world and devise their own categories, the charge is vacated. As I and others have argued, all meaning in organisms is constructed by self-regarding processes that gerrymander the ‘given’ categories of physics to suit their purposes. This idea has been around for a long time, in Sellars’ distinction between the scientific image and the manifest image, for instance, and in Quine’s discussion of biases in the property-spaces of living things. Still, as an observer, one can in principle identify the categories adopted/constructed by an organism objectively — such as the categories of the sweet and the cute, the sexy and the ugly — once one understands in detail the evolutionary/ ecological predicament of the organism in question. In fact, Thompson quotes at length from my own account of this (p. 160) and then adds: ‘This passage could easily stand as a gloss on what I have been calling sense-making.’ The difference, he notes, is that his version is based on autopoiesis and mine on selfish-gene theory. That counts against my version only if selfish-gene theory is in error, which Thompson fails to show, but in any case it belies the rathering quoted above: the ‘autonomy perspective’ is not required for making sense of sense-making. I say that agency is born ‘in the first macromolecules that have enough complexity to “do things”’. Thompson points out, truly enough, that such macro-molecules cannot exist without being parts of autopoietic systems, but still they are agents, even if not fully autonomous. Think of motor proteins — little porters trudging along on their actin or tubulin highways carrying freight to where it is needed. Think of proof-reading enzymes. One of my favourite quotations from Maturana in the book is this: ‘Living systems are cognitive systems, and living as a process is a process of cognition. This statement is valid for all organisms, with and without a nervous system’ (1970, quoted on p. 124). Indeed, and the cognitive processes (in this somewhat extended sense of ‘cognitive’) that even bacteria and their proper parts engage in can best be appreciated from a reverse engineering perspective: there must be

SHALL WE TANGO? NO, BUT THANKS FOR ASKING

31

transducers that detect the life-relevant conditions and effectors that are appropriately linked to those transducers so that the ‘self-producing’ prowess of the autopoietic system can be maintained and enhanced. (See the excellent book by Dennis Bray, Wetware, 2009, for a lucid discussion of the computational networks of molecules in bacteria and other unicellular life forms.)

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

4. First-Person Phenomenology vs. Heterophenomenology Thompson lists five criticisms of heterophenomenology. First, it is wrong, he claims, for me to look for confirmation or disconfirmation of subjects’ heterophenomenological reports by discovering processes occurring in their brains. ‘One is describing one’s subjectivity at the personal level in a way that is completely noncommittal about the subpersonal workings of one’s brain’ (p. 305). Right he is, and I have not forgotten or abandoned the personal/subpersonal level distinction, which I introduced (1969), so I am quite alert to this point. I go out of my way to explain that subjects’ convictions about, or commitments about, what is going on in their brain are simply not part of the heterophenomenological data. People have all sorts of bizarre theories about what they are talking about and I am looking at what might be called unwitting reference. That is why we need to adopt the peculiar agnosticism recommended by heterophenomenology (at least a close kin to Husserl’s epoché); but once we have gathered our heterophenomenological story, it is time to start theorizing and experimenting, and that is where we find the leverage to discover that, for instance, since there are no images being processed in the brain when subjects say they are rotating mental images, their heterophenomenological reports must be interpreted as unwitting fictions of a sort. (For more on this see my response to Schwitzgebel, 2007, in Dennett, 2007.) Second is a point about intentional acts that I don’t understand: ‘From a phenomenological standpoint, however, there is no such thing as an intentional or notional object without a corresponding intentional act, and intentional acts are not to be identified with beliefs’ (p. 305). This must be a special, phenomenological meaning of ‘intentional acts’ since manifestly there are intentional objects without intentional acts in the everyday sense. Your misidentification of the tree stump on the side of the road as a pedestrian was not an intentional act, but it creates an intentional object. And although the process (jumping unconsciously to a conclusion, or whatever we call it) that creates this intentional object is not itself a belief, it generates a

32

D.C. DENNETT

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

belief — perhaps fleeting — to the effect that there is a pedestrian out there. I wonder what I am missing. Third, Thompson describes the ‘unavoidable need to make use of first-personal modes of access to mental phenomena’. I don’t deny that. This objection is a red herring. Heterophenomenology is advertised by me as the way — the best way — of taking first-person data seriously within the third-person domain of cognitive science; that is why I went into considerable detail about the ways in which we can transform first-person modes of access into (usable) third-person sources of data by proper scientific method. Still, Thompson avers, ‘Nothing in heterophenomenology would lead it to envision — let alone take the step — of working with experience in this direct phenomenological way’. On the contrary, the whole point of heterophenomenology is to give the subject the best possible opportunity to let it all hang out, in effect, to tell the world about what it is like to be that subject! It just requires, as all science does, that there be interpersonal standards of data-gathering.2 To the extent that Lutz et al. (2002), the flagship attempt at a first-person methodology for studying consciousness, does not meet these standards, it has been quite appropriately criticized (see, e.g. Piccinini, 2010). Fourth, ‘to take statements about experience as expressions of beliefs about experience seems strained’ (p. 307). How so? It ‘overintellectualizes’ the experience and ‘runs the risk of overinterpreting subjects, thereby distorting their experience’. This objection actually supports what it sets out to criticize. Suppose a Turkish subject says (what seems to English observers to be the sentence): ‘I’m rotating the left image clockwise.’ We will have no way of using this raw datum at all if we can’t interpret this as an expression in English of his current belief (or his expression, in Turkish, of some other belief that just happens to sound like the English sentence!) We can’t let subjects’ expressions count as evidence for their conscious experience — aside from utterances like ‘Ouch!’ and ‘Ahhhhhh!’ — without interpreting them as propositions asserted in some language, and sincerely asserted at that; in other words as utterances expressing their current beliefs. Thompson goes on to say that my proposal ‘collapses the crucial distinction between conscious experience and belief about conscious experience’ (p. 307). He suggests that ‘pervasive moods’ and ‘rapid and transient emotional experiences’ are experiences we don’t believe we have but ‘First-person and second-person methods work [2]

Thompson’s claim here was first presented in Thompson (2007b) and responded to by me — in somewhat more detail than here — in Dennett (2007).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

SHALL WE TANGO? NO, BUT THANKS FOR ASKING

33

directly with these sorts of experience and thus do indeed garner more usable data than heterophenomenology does’ (p. 308). Perhaps, but since Thompson doesn’t say how these methods accomplish this, we will have to wait for a verdict. An example of such a method in operation would be most helpful. Fifth, he says that ‘heterophenomenology is no mere extension’ of ordinary science because it involves using the intentional stance. Thus it stands in ‘an interpretive, intersubjective, and interpersonal relation to its subject matter’. Indeed. That makes it very different from chemistry and biology — we don’t have to whisper when experimenting with proteins for fear of their learning what we hope to discover. That is why I went to all the trouble to show that in spite of its being interpretive, we could achieve usable levels of intersubjective agreement on the part of experimenters — availing ourselves of the standard techniques of double-blind experimentation with independent scoring, etc. Instead of having to be hidebound behaviourists, eschewing all verbal interactions with our subjects, we can use language rigorously and cautiously to get at what it is like to be them, by asking them (second-personally) and letting them tell us (from the first-person point of view). Thompson says in the end that heterophenomenology, by relying on these ‘second-person’ methods, is not strictly speaking a third-person method. Perhaps, then, what he should next acknowledge is that his own proposed method is really heterophenomenology after all rather than a radical alternative. If he will adopt my method, I will adopt his name for it: second-person heterophenomenology. Let me close with a limited defence of setting up strawmen! Think of a dance instructor, who, seeing her pupil unwittingly doing something slightly awkward, says ‘You’re doing this!’ — and produces an hilarious parody of the poor student’s ungainly move. Ouch! Unfair! But probably a very good way, and maybe the only way, of getting the student to notice an unrecognized weakness in her dancing. Now consider what Thompson has done: he sees — or thinks he sees — that contemporary evolutionary biologists and cognitive scientists (and some philosophers thereof, e.g. me) are all too comfortable with the oversimplifications they have learned to love, and tend to overlook or underestimate important complications that would actually enrich their understanding of the phenomena they are dealing with. How to get them to notice? Poke them in the eye with a parody! The tactic can backfire, of course, and leave everybody grumpy, but a more constructive response is to take the criticism to heart, note, for the record, the misrepresentations in the challenge, and say ‘Thanks, I needed

34

D.C. DENNETT

that!’ Such a response co-opts the revolutionary, turning him and his heroes into mere reformers, but such is life. Thompson would love to get us all to dance the tango, but — speaking for myself — it ain’t gonna happen!3

References

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Bray, D. (2009) Wetware: A Computer in Every Living Cell, New Haven, CT: Yale University Press. Dennett, D.C. (1969) Content and Consciousness, London: Routledge & Keegan Paul. Dennett, D.C. (2007) Heterophenomenology reconsidered, Phenomenology and the Cognitive Sciences, 6 (1–2), pp. 247–270. Edelman, G. (1989) The Remembered Present, New York: Basic Books. Gánti, T. (2003) The Principles of Life, Oxford: Oxford University Press. Levins, R. & Lewontin, R. (1985) The Dialectical Biologist, Cambridge, MA: Harvard University Press. Lutz, A., Lachaux, J.-P., Martinerie, J. & Varela, F.J. (2002) Guiding the study of brain dynamics by using first-person data: Synchrony patterns correlate with ongoing conscious states during a simple visual task, Proceedings of the National Academy of Sciences USA, 99, pp. 1586–1591. Maturana, H. (1970) Biology of cognition, in Maturana, H. & Varela, F.J. (eds.) Autopoiesis and Cognition: The Realization of the Living, pp. 2–58, Boston Studies in the Philosophy of Science, Vol. 43, Dordrecht: D. Reidel. Moss, L. (2003) What Genes Can’t Do, Cambridge, MA: MIT Press. Piccinini, G. (2010) How to improve on heterophenomenology: The self-measurement methodology of first-person data, Journal of Consciousness Studies, 17 (3–4), pp. 84–106. Schwitzgebel, E. (2007) No unchallengeable epistemic authority, of any sort, regarding our own conscious experience — contra Dennett?, Phenomenology and the Cognitive Sciences, 6 (1–2), pp. 107–113. Thompson, E. (2007a) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: MIT Press. Thompson, E. (2007b) Look again: Heterophenomenology and mental imagery, Phenomenology and the Cognitive Sciences, 6 (1–2), pp. 137–170.

[3]

Yes, I know, the tango is from Argentina, not Chile, but since neither I nor the rest of the world knows of a national dance of Chile that has the passion and drama of the tango, I have to settle for this readily understood metaphor, with apologies to all my Chilean friends.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Lucia Foglia & Rick Grush

The Limitations of a Purely Enactive (Non-Representational) Account of Imagery 1. Introduction In Chapter Ten of his book, Thompson’s goal is to sketch a phenomenological analysis of imagery experience as part of a confrontation of the two leading imagery theories, descriptivism and pictorialism. While these two theories are, strictly speaking, theories of the subpersonal mechanisms that support mental imagery, they both appeal to phenomenology, and Thompson argues that both rival theories get the phenomenology wrong because both assume imagery to be subjectively similar to the experience of seeing pictures. Rather, he argues that imaging or visualizing is not experiencing a mental picture of an object or scene, but rather ‘mentally enacting or entertaining a possible perceptual experience of that thing or scene’ (p. 269). And much of the chapter is devoted to describing what the phenomenology of such overt perceptual episodes is like. We don’t intend to comment on Thompson’s analysis of the phenomenology of perception, other than that his analyses seem reasonable. Rather, our focus will be Thompson’s endorsement of the enactive theory of perception and imagery, and its concomitant antirepresentationalist leanings. Enaction, as put forward by Varela and defended by other thinkers (notably Alva Noë, 2004; Susan Hurley, 2006; and Kevin O’Regan, 1992), departs from traditional accounts that treat mental processes (like perception, reasoning, and action) as Correspondence: Lucia Foglia Email: [email protected] Rick Grush Email: [email protected]

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 35–43

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

36

L. FOGLIA & R. GRUSH

discrete, independent processes that are causally related in a sequential fashion. According to the main claim of the enactive approach, which Thompson seems to fully endorse, perceptual awareness is taken to be a skill-based activity. Our perceptual contact with the world, according to the enactionists, is not mediated by representations but is enacted, and the notion of representation, belonging to the classic computational paradigm, has no place in this alternative approach. Though Thompson does not pronounce directly on the issue of representationalism, he is most definitely keeping the company of anti-representationalists, and in that context it is not unreasonable to take his silence for consent. In this paper, we will argue that the enactive approach to imagery is unworkable unless it makes appeal to representations, understood in a particular way. Not understood as pictures, to be sure. Or sentences for that matter. But those aren’t the only options.

2. A ‘representationalist-enactive’ account of mental imagery Here is the main argument to the effect that the enactive approach needs representations in order to account for some key features of imagery. Consider a situation in which a subject solves a few Cooper-Shepard rotation tasks, and consider four cases: a.

1-match: The subject is presented with a 1 and a 8 and determines that they are congruent by mentally rotating the 1 until it conforms with the 8.

b.

1-non-match: The subject is presented with a 1 and a 3 and determines that they are not congruent by mentally rotating the 1 until it is apparent that it is incongruent with the 3.

c.

6-match: The subject is presented with a 6 and a 3 and determines that they are congruent by mentally rotating the 6 until it conforms with the 3.

d.

6-non-match: The subject is presented with a 6 and a 8 and determines that they are not congruent by mentally rotating the 6 until it is apparent that it is incongruent with the 8.

Now, we will suppose that the subject is solving these problems by using mental imagery, and in particular by rotating a mental image. We take it that Thompson and his enactive allies agree with us that

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

AN ENACTIVE ACCOUNT OF IMAGERY: LIMITATIONS 37

imagery is at least sometimes used to solve this sort of problem, and that imagery is a simulation of a corresponding overt perceptual situation. There are details one might worry about, but this first approximation is shared ground, and the first approximation will be enough for us to make our point. What might such an overt perceptual situation be? One possibility would be an actual overt manual rotation of a sheet of paper with a large 6 printed on it, or perhaps manually rotating a block of wood shaped like a 6. And in such cases the overt behaviour would help the subject solve the problem because after the rotation it is perceptually obvious whether the shapes are congruent or not. Sticking with the overt situation for a moment, we can see that it has two distinct components. First, there is the subject’s behaviour strictly understood — in this case a manual grasp and a counterclockwise hand rotation. And second there is the object that the behaviour is directed towards, in this case a sheet of paper with a 6 printed on it or a block of wood. These are separable in the sense that each contributes a component to an activity which can be compositionally substituted-for. A different situation might involve the same object, but a different action — the subject might twist the sheet of paper with the 6 clockwise rather than counterclockwise, for instance. Another situation might involve the same action (a counterclockwise twist), but a different object — a sheet of paper with a 1 on it, rather than a 6. Crucially, it should be pointed out that both components are required to solve the problem overtly: a piece of paper that the subject doesn’t rotate won’t make the solution perceptually apparent, nor will an empty hand twisting clockwise or counterclockwise. Both components are present and required in the overt situation. Now what about the covert situation where the subject solves the problem using imagery? Well, on one sort of theory, which we will call the ‘simulation theory’, the subject engages, in an off-line way, the same set of active behaviours that would be involved in the overt case. What this means exactly will be addressed shortly, but to keep things simple for now, we can gloss it as something like the covert operation of some of the same motor areas, producing activity that is relevantly similar to the activity they would produce if acting overtly. More on this shortly. On the second approach we will discuss, which we will call the ‘emulation theory’ (Grush, 2004), the subject engages, in an off-line way, the same set of active behaviours that would be involved in the overt case, and in addition targets those behaviours on an internal model of whatever it is that, in the overt case, the overt behaviours

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

38

L. FOGLIA & R. GRUSH

would act upon. The difference, as we are defining them, between the simulation theory and the emulation theory is the material in the italicized phrase in the previous sentence; namely, the existence and use of internal models. That the brain entertains models to produce imagery does not imply that such models are always accurate and complete: indeed there are reasons to think that such internal models are often inaccurate and sketchy. Although the degree of accuracy and completeness are not an issue here, this clarification will hopefully remove one possible irrelevant objection to the notion of internal models. If one thinks that pictures are always fully specified, then this will be one difference between models, as we understand them, and pictures. On the simulation theory, the subject is producing the same actions, the same behaviours, covertly. And this means the motor commands corresponding to the twisting hand. According to the proponent of the emulation theory, this is insufficient, just as an overt twist of an empty hand is insufficient to make the solution apparent in the overt case. What is needed in the covert case is something corresponding to the piece of paper or block of wood, call it an internal model, on which the covert behaviours act, and whose post-manipulation state corresponds to the post-manipulation state of its overt counterpart. The covert situation enables the subject to solve the problem exactly because the manipulated model makes the solution apparent in the same way that the rotated drawing on paper would make the solution apparent in the overt case. Another way to make the point is this: suppose that when we make the overt situation covert, only elements corresponding to the twisting hand (the subject’s actions strictly understood) are made covert, and nothing corresponding to the piece of paper or block of wood is made internal or covert. Then there would not be a difference between imagining rotating a 6, and imagining rotating a 1, for the simple reason that the actions, strictly understood, are the same in both cases — a clockwise hand rotation, say. In the overt case the difference between the two is supplied by different objects being acted on (by the same action). In the covert case, if there is nothing internal corresponding to the rotated sheet or block of wood, only a covert hand rotation, then the two putatively qualitatively different imagery episodes would not in fact be qualitatively different. But they are qualitatively different. Therefore, there must be more to what is going on covertly than just the strict action component. The upshot of this point is that enactive accounts of imagery are well and good, but if they are presented in such a way that makes it sound as

AN ENACTIVE ACCOUNT OF IMAGERY: LIMITATIONS 39

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

though they are adequate to the phenomena and are non-representational, then they are trying to sell a bill of goods. Before discussing the notion of representation involved, we need to address a possible objection. Objection: there are different ways to specify a motor behaviour. There are specifications in terms of muscle contractions, or equilibrium point settings, and on such a scheme, a given behaviour might be specified in low-level terms. But there are also specifications in terms of goals, such as ‘grasp the cup’, in which lowlevel terms are beneath the surface, and what is relevant is a higher-level specification. Your argument depends on specifying the covert behaviour in low-level terms. If the behaviour is specified in suitably high-level terms, like ‘rotate the 1’, then the problem is solved. It is the higher-level behavioural goal specification that is being covertly simulated, and this does not require anything corresponding to an internal model of the sheet of paper. This objection fails. It is vague, and could be clarified in either of two very different ways. On one way of clarifying it, it is just an obfuscating way of restating the emulation theory, and on the other way, it is simply and demonstrably unfit to explain the empirical facts. First, one way to spell out what the ‘rotate the 1’ (as stated in the text of the objection above) means just is the emulation theory. What is being covertly executed is both a covert motor action, corresponding to the ‘rotate’ part of the gloss, and an internal model, corresponding to the ‘the 1’ part of the gloss. So this interpretation doesn’t leave the objection standing as an objection. The other way to spell it out is as something akin to a goal specification. Let’s move to a different example for a moment. The behavioural specification ‘grasp the cup’ has the goal state, the end state of a cup being grasped, of the behaviour in its specification. Higher-level motor areas seem to work with specifications at such a level. And so perhaps by analogy, the higher-level specification of the covert behaviour involved in imagery is also in terms of end-states, like ‘make the 1 into a 8’. If motor behaviours are given in these terms, then perhaps (the objector believes) we can make sense of the idea of an internally simulated covert behaviour that doesn’t seem to need anything corresponding to an internal model. Here’s why this won’t work. The production of imagery is supposed to explain how subjects solve the problem, but this proposal requires that the problem already be solved before this solution can begin. The crucial question is: how does the motor system know that the behaviour it is supposed to simulate is (i) ‘make the 1 into a 8’,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

40

L. FOGLIA & R. GRUSH

and not (ii) ‘make the 1 into a 3’? Presumably the reason that imagery is required in the first place is that the subject does not know, before simulating the behaviour, what the figure will look like after being rotated. But a behavioural specification like (i) presumes that the answer is known before the simulation is executed, and is worked into the goal-level specification of the behaviour. A common sort of data appealed to be proponents of the simulation theory involves finding subtle motor behaviours during imagery and cognition. For example, studies showing that spontaneous eye-movements, in the absence of sensory signals, occur in the interpretation of a linguistic description of a visual scene (Spivey and Geng, 2001). But this sort of result is predicted by both the simulation and emulation theory, and so does not decide between them. To do that, we need another kind of data. In a quite interesting study, Presson and Montello (1994) blindfolded subjects who were asked to point to specific objects after real and imagined walks. As people move through the environment they change their location with respect to surrounding objects and accordingly they update their knowledge of the relative locations of objects in a fairly quick and accurate manner. Similar updating occurs when they imagine moving through the environment, but it is less quick and accurate. When blindfolded participants actually rotated their body 90 degrees they were just as fast and accurate to point to the specific location of objects in the room as they were before rotating. When they were asked to imagine such rotation and to point to specific objects again, they were slow and inaccurate. There is no obvious explanation for this difference if the only component in imagery were the covert action and no internal models would be involved. First, an enactivist might claim that covert actions are simply not very good at generating imagery, they only imperfectly correspond to overt actions. But since people can in fact generate very detailed imagery — in some cases so detailed that it is not easy or even possible to distinguish the perceived from the merely imagined, as with hallucinations — this explanation seems flawed. A more typical explanation in these sorts of situations is to appeal to interference. For example, in a study conducted by Marc Wexler and colleagues (1998) subjects were asked to perform the Cooper-Shepard mental rotation task while executing a motor rotation of a joystick handle in a given direction at a previously learnt speed. Consistently with their hypothesis, if visual image transformations and execution/control of physical movements share the same neural substrates, then ‘simultaneous mental and motor rotation should interfere with each other’ (Wexler et al.,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

AN ENACTIVE ACCOUNT OF IMAGERY: LIMITATIONS 41

1998, p. 81). The experiment showed specific and detailed interactions between mental rotation of abstract objects and motor rotation of a joystick handle: specifically, mental rotations were faster when they were in the same direction of the motor rotation, rather than in the opposite, and were highly influenced by the speed of the motor rotation of the joystick handle. It should be noted that the interaction between the two rotations appeared without any visual feedback from motor rotation: subjects could neither see the joystick handle nor watch themselves using their hands during the trials. But in the Presson and Montello study (and many others like it), the subjects are not engaging any interfering overt behaviour, and so there seems to be no resources to explain the impaired imagery. But the proponent of the emulation theory does have resources, namely, the internal model. Even though subjects are asked to engage in an imagery task, they still maintain a non-manipulated model of the relative location of objects in their environment (if you now close your eyes and imagine walking into another room, you don’t lose track of where your body really is, and where objects are around your actual body; this is because you are maintaining two models, one you take to represent reality, and another that represents an imagined alternate situation). Interestingly, the most vivid kinds of imagery, hallucinations, are precisely those where the distinction between imagination and reality is not maintained, where, that is, there are not two mutually interfering models, but only one. While motor behaviour is covertly involved in imagery as described above, clearly by itself it cannot determine the nature of the imagery (which objects will be recalled or described). So to summarize this section, the simulation approach to imagery, defined as one that tries to explain imagery as covert behaviour without anything corresponding to an internal model on which that behaviour is directed, is a non-starter. On a low-level specification of behaviour, the best it can do is covertly reproduce the empty twisting hand, which doesn’t help a subject solve the problem any more than an overt twisting empty hand. On a high-level specification of the covert behaviour, it is either a restatement of the emulation theory (if it allows for a model), or it presupposes the solution to the problem it is intended to solve (if it does not allow for a model, and tries to get by with a high-level specification of the behaviour in goal terms). The emulation theory, on the other hand, does not face these problems. It posits imagery as covert sensorimotor behaviour, which includes not only covert stuff corresponding to the agent’s actions, but also covert

42

L. FOGLIA & R. GRUSH

(internal) stuff corresponding to what those actions act on, like the sheet of paper or the block of wood.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

3. The nature of models and why they should not be feared by the enactivists Debates about representationalism typically revolve around two metaphors: pictures and sentences. The debate, indeed, seems to be trapped into these two dominant metaphors of representation and, precisely because neither of the two really captures the character of mental imagery, some authors have preferred to abandon the notion of representation to explore different options. The anti-representationalist arguments to be found among the enactive camp are often of this character. The crucial question is whether there is any way of understanding what a representation is that does not rely on the metaphors of pictures or language? One possibility is that of a model. The crucial idea might be loosely described by saying that something, M, is a model of X (for some agent A) if A can use M as a stand-in for X. It might also be glossed as: A can interact with M in a manner analogous to how A interacts with X. On this account, arguably pictures are a subcategory of models. The feature of pictures that most compellingly suggests that they are representations is that we can engage with them visually in a manner analogous to how we engage with the pictured object or scene. Our eyes can move around a picture of a face in a manner analogous to how they would move around a physical face, we can engage the ‘face recognition’ areas of the brain on the picture in a manner analogous to how they would be engaged on a real face, and so on (a list of digits specifying the grey levels at every pixel of the photograph might be caused by the same face, but would not be something on which the face recognition areas of our brains might engage). There are limits to the kind of interactions that are supported by a picture: one can’t engage with a photograph pharmacologically as one might with a real person (to test cancer medication, for example), which is arguably why nobody thinks that pictures represent physiology. There are many details, of course, swept under the term ‘analogous’, but time doesn’t permit exploring them. The examples above should make the idea clear at a first pass. Consider a flight simulator, which is used as a model of an aircraft. What allows it to represent the aircraft is exactly that a person can interact with the simulator in a manner analogous to how she might interact with a real aircraft. But the simulator is neither a picture, nor a

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

AN ENACTIVE ACCOUNT OF IMAGERY: LIMITATIONS 43

sentence. Crucially, it might include pictures and sentences: the display representing the visual scene through the windshield is a video picture, representing the environment as it would be seen from a cockpit (the pilot’s sensorimotor interaction with the display will be analogous to her sensorimotor interaction with the visual scene as viewed from a cockpit window); and there might be sentences produced by some instruments. But in all cases, the crucial factor is that something is being used as a surrogate, a model, for something else, and hence represents that something else. Note a few crucial things about this understanding of representation. First, on this understanding representation is skull-transparent. Whether the model is inside or outside the skull (or spans both) is not a relevant factor on our account. What is theoretically crucial is whether the entity can be interacted with in a manner analogous to the represented entity. Whether the model is internal and implemented in neural tissue, or external and implemented in blocks of wood or flight simulation software, is not relevant. Second, action is still of central importance. What makes the model a representation is precisely that it can be interacted with in a certain way. So this theory shares a central commitment with the enactivist camp. Since Thompson does not discuss representationalism directly, we don’t know if what we say about it here is something that he sees as a challenge to his position on imagery, or as a friendly clarification. That’s for him to say.

References Grush, R.S. (2004) The emulation theory of representation: Motor control, imagery, and perception, Behavioural Brain Science, 27, pp. 377–442. Hurley, S. (2006) Active perception and perceiving action: The shared circuits hypothesis, in Gendler, T. & Hawthorne, J. (eds.) Perceptual Experience, Oxford: Oxford University Press. Noë, A. (2004) Action in Perception, Cambridge, MA: MIT Press. O’Regan, J.K. (1992) Solving the ‘real’ mysteries of visual perception: The world as an outside memory, Canadian Journal of Psychology, 46, pp. 461–488. Presson, C. & Montello, D. (1994) Updating after rotational and translational body movements: Coordinate structure of perspective space, Perception, 23, pp. 1447–1455. Spivey, M. & Geng, J. (2001) Oculomotor mechanisms activated by imagery and memory: Eye movements to absent objects, Psychological Research, 65, pp. 235–241. Wexler, M., Kosslyn, S. & Berthoz, A. (1998) Motor processes in mental rotation, Cognition, 68, pp. 77–94.

Daniel D. Hutto

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Philosophy of Mind’s New Lease on Life Autopoietic Enactivism meets Teleosemiotics What he saw was a shocking surprise. Every Who down in Who-ville, the tall and the small, Was singing! Without any presents at all! He HADN’T stopped Christmas from coming! IT CAME! Somehow or other, it came just the same! And the Grinch, with his grinch-feet ice-cold in the snow Stood puzzling and puzzling: ‘How could it be so?’ ‘It came without ribbons! It came without tags! It came without packages, boxes or bags!’ And he puzzled three hours, till his puzzler was sore. Then the Grinch thought of something he hadn’t before! ‘Maybe Christmas,’ he thought, ‘doesn’t come from a store. ‘Maybe Christmas… perhaps… means a little bit more!’ Dr. Seuss, How the Grinch Stole Christmas (1957)

1. Introduction Mind in Life is a big book — in more ways than one, and in more good ways than one. In promoting the idea that there is a deep continuity between life and mind, Thompson defends a boldly anti-representationalist version of enactivism: it challenges, root and branch, the ‘passive-cognitivist’ view of the mind-brain. It aims for nothing short of the eradication of misleading Cartesian and Kantian (of the Critique of Pure Reason) dualisms that, despite challenges, continue to Correspondence: Daniel D. Hutto, Professor of Philosophical Psychology, School of Humanities, University of Hertfordshire, de Havilland Campus, Hatfield, Hertfordshire, AL10 9AB Email: [email protected]

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 44–64

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

45

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

dominate analytic philosophy of mind and some branches of cognitive science. Thompson sees adoption of the enactive approach as a way to put aside the mind–body problem, once and for all, and to refocus our investigations on the more fertile — phenomenologically-inspired — body–body problem. We are meant to give up on the traditional input-output processing model of the mind, one that continues to pay homage, if only tacitly, to the idea of what is sensorially or informationally ‘given’; of content that is received or informs, on the one hand, and the idea that such gifts are intellectually categorized, conceptualized, and schematized — in a downstream and serial manner — by higher forms of cognitive spontaneity, on the other. By reconceptualizing mentality in essentially active, dynamic and loopy terms, Thompson undermines the traditional boundaries and dichotomies. The boundaries thought to hold between mind and body and between mind and world are revealed to be, ultimately, of only heuristic value, having no genuine metaphysical import. We are asked to trade in the modern, Cartesian, picture of the mind as a sort of mechanism for a more Aristotelian vision of mentality, which emphasizes its biological character and the special features it shares with all living systems.1 All of this is in the service of putting us in a better position to understand aright the place of consciousness in nature, and thus to deal with the infamous explanatory gap. In what follows, for reasons of space, I say nothing more about the book’s success in meeting its principal aim of enlarging and enriching ‘the philosophical and scientific resources we have for addressing the gap’ (p. x). I think it surely achieves this and I am wholly sympathetic to Thompson’s ambitions on this front and, by and large, the general sort of enactivism he promotes. This is not to say that this book closes the gap (it doesn’t seek to) or that it is the last word on consciousness (it doesn’t promise to be). Still, it is a genuine tour de force, that adds creatively and convincingly to our ways of understanding basic forms of mentality. I will take this much for granted here and work instead to ensure that Thompson’s work gets the kind of reception it should have. For the truth is that many working in analytic philosophy of mind and cognitive science continue to be utterly mystified about what enactive approaches have to offer. To remedy this, this commentary will seek to clarify certain core features of Thompson’s proposal about the enactive nature of basic mentality, as best it can, and to bring his ideas into direct conversation with accounts of basic cognition of the sort favoured by analytical [1]

For a discussion of this contrast see Thompson (2007, p. 80).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

46

D.D. HUTTO

philosophers of mind and more traditional cognitive scientists — i.e. those who tend to be either suspicious or critical of enactive/embodied approaches (to the extent that they confess to understanding them at all). My proposed way of opening up this sort of dialogue is to concentrate on the close similarities between Thompson’s biologically-based proposal about non-representational forms of basic cognition and what I take to be a reasonable modification to the ambitions of teleosemantic theories of content. In so far as today’s theories of mental representation are less concerned to understand content in properly semantic terms, they are moving ever closer to the sorts of account proposed by enactivists of the Thompsonian stripe — close enough to have meaningful debates about the nature of basic mentality. It is against this backdrop that I put a spotlight on the true promise and value of enactivism, providing some compelling reasons for wanting to go Thompson’s way. This will be achieved, in large part, by showing that there is no wholly agreed way in which the family of fundamental theoretical notions — content, representation and, especially, information — are understood by opponents of enactivism. Despite their centrality and critical importance to traditional cognitivism there are no agreed and well-defined accounts of the exact nature of contentful, representational or informational properties. As such, it can be difficult to determine the precise boundaries between representationalist and nonrepresentationalist approaches. Moreover, when we get down to brass tacks, it looks as if the most scientifically respectable attempts to make sense of these ideas leads straight into the arms of the sort of enactivism that Thompson proposes (or something near enough). In short, once we make necessary clarifying adjustments to the best proposals for understanding the kind of informationally-sensitive responding that constitutes basic mentality, Thompson-style enactivism may well turn out to be the most promising naturalist game in town, if not the only one.

2. The Information-Processing Challenge The last claim of the previous section is far from obviously true. Originally advanced by Varela, Thompson and Rosch (1991), the big idea behind enactivism is to treat consciousness and cognition as emergent phenomena constituted by, and thus to be understood in terms of, specifiable patterns of organismic activity. Enactivism of this stripe denies that the most basic forms of genuinely mental activity necessarily involve, or are to be explained by, the manipulation of contentful representations. Instead, enactivists hold that mentality emerges

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

47

from — is ‘brought forth by’ — the self-creating (autopoietic) activities of organisms, and that the latter are constituted by essentially embodied, diachronic environmental interactions that are the basis for new possibilities for self-creating activity in a dynamic way. Drawing on insights from phenomenology and dynamical systems theory, enactivists invert the familiar explanatory strategies of orthodox cognitive science by supposing that ‘Abilities are prior to theories… Competence is prior to content… [and that] knowing how is the paradigm cognitive state and it is prior to knowing that’ (Fodor, 2008, p. 10). The framework has proved attractive to many. A great variety of enactivist proposals have now been advanced about many topics, including: consciousness, perception, intentionality, attention, memory, emotion, intersubjective social cognition and self-consciousness. Nevertheless this take on basic mentality is viewed by many as at best something that might supplement existing theories of cognition, and at worst is nothing more than a confused and obscuring gloss which adds nothing positive to the already well-established cognitivist accounts. Varela-inspired versions of enactivism face a standard objection, given that their ‘main explanatory tool… is the theory of self-organizing and autonomous dynamic systems’ (Thompson, 2007, p. 26). For even among those who are prepared to accept that the enactive approach holds promise for understanding how organisms develop and interact over time, a standard verdict is that it lacks the independent explanatory resources to provide a genuinely alternative understanding of the basis of mentality (Ramsey, 2007; Clark, 2008). Clark (2008) describes the emphasis placed by theories like Thompson’s on the dynamics of the total state of systems as both a boon and burden. On the positive side, it allows the theorist to ‘accurately capture the way two or more systems engage in a continuous real-time, and effectively instantaneous dance of mutual codetermining interaction’ (ibid., p. 25). On the downside, it is problematic ‘insofar as it threatens to obscure the specifically intelligence-based route to evolutionary success’ (ibid.) And it does this to the extent that it fails to recognize ‘the brain as the principal (though not the only) seat of information-processing activity’ (ibid., emphasis original). Accordingly, we need not deny the importance of timing, action and coupled unfolding to cognition so long as we do not forget that these play support roles in intelligent responses ‘grounded in processes of information extraction, transformation and use’ (ibid., p. 19). In sum, the complaint is that any version of enactivism that relies entirely on dynamical systems theory under-appreciates the fundamental role played by information-processing mechanisms in making mental

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

48

D.D. HUTTO

activity so much as possible. Call this the Information-Processing Challenge. The Information-Processing Challenge would present a formidable problem for enactivists if it could be safely taken for granted that the standard computational/information-processing explanatory strategies of traditional cognitivism are in perfectly good order under standard renderings. But enactivists question just this. This is precisely what Thompson (2007) has in his sights when he doubts the truth of the received view in cognitive science. That view, he maintains, is committed to the idea that ‘in order to explain cognitive abilities we need to appeal to information-bearing states inside the system. Such states, by virtue of the semantic information they carry about the world, qualify as representations’ (p. 52, emphasis added). This is where the trouble in assessing this debate starts. For there is, in fact, less consensus about exactly what orthodox cognitive science is committed to than this statement suggests — in particular it is not clear that the kind of information or content that matters must be semantic. Still the conservative wing of cognitive science certainly does insist that intelligence depends on information-processing. Textbooks in the field tell us that the core assumption of traditional cognitive science is ‘that there are sub-personal contents and sub-personal operations that are truly cognitive in the sense that these operations can be properly explained only in terms of these contents’ (Seager, 2000, p. 27). As the Clark quotation reminds us, information is thought to be the basic currency of cognition; it is received, stored, manipulated, and transformed by intelligent systems. It is the fuel for cognitive engines. So conceived ‘information is a prime commodity, and when it is used in biological theorizing it is granted a kind of atomistic autonomy as it moves from place to place, is gathered, stored, imprinted, and translated’ (Oyama, 2000, p. 1). This standard metaphor suggests that cognitive operations really involve the manipulation or processing of information or content of some kind or other — but despite the foundational importance of this claim it is incredibly difficult to pin down, with any firm grip, the theoretical commitments of those who propound such stories. In particular, it is hard to get a clear sense if they are truly committed to this sort of picture and if so exactly what it is that is supposed to be processed by these intelligent systems and how this is done. I suggest that the more we work through the possible readings and home in on a credible account that is naturalistically acceptable, the closer we come to accepting the kind of enactivist proposal Thompson advocates. What then might possibly fuel cognition?

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

49

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

3. What Might Content Be? The sentences of natural language, as expressed in linguistically mediated beliefs and utterances, are clearly the paradigms of contentful representations, if anything is. It is also quite clear that when philosophers of mind first developed naturalized theories of content their aim was to explain how mental representations could have semantic properties of just the same sort possessed by linguistic representations. A major motivation for this project is that success in this endeavour would make it possible to explain how language could gain its semantic properties from underived mental contents, those which could be explained by appeal to non-semantic properties — such as causation or biological function. Assuming that an exact parallel exists between the content of thought and language ensures that whatever can be thought or judged can be said, in principle. Saying something really only requires finding a public means of expressing oneself. With this in mind psychosemantic theories of content, including teleosemantics, seek to explain the semantic properties of mental states where these are understood as having the very same sorts of semantic properties possessed by natural language expressions. Proponents of Fodor’s ‘language of thought’ hypothesis wear this commitment on their sleeves. They are interested in mental representations with semantic content of essentially the same kind as that which natural language sentences possess (this follows given that whatever content the latter have is wholly derived from the former). Fodor (2008) tells us that ‘the content of a mental representation is its referent’ (p. 216). If such content is to express truths then the referents in question would need to be states of affairs (or the equivalent) and the referents would have to be picked out intensionally, i.e. under some description or mode of presentation, in order to satisfy the platitudinous disquotational rule (‘Snow is white’ in L iff Snow is white). It is quite clear that others working in this area are also primarily interested in mental representations with content of this sort. Millikan (2005), for example, thinks that representational content is essentially truth conditional. Hence, ‘intentionality has to do with truth conditions’ (ibid., p. 93). For her this requirement goes all the way down, thus ‘the intentionality of language is exactly parallel to the intentionality of bee dances’ (ibid., p. 98). The same goes for Papineau who offers a ‘naturalistically acceptable explanation of representation: namely that the biological purpose of beliefs was to occur in the presence of certain states of affairs, which states of affairs counted as their truth conditions’ (Papineau, 1987, p. xvi). And, again, McGinn tells

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

50

D.D. HUTTO

us that the aim of teleosemantic theories of content was to show how ‘teleology turns into truth conditions’ (McGinn, 1989, p. 148). All of this connects with what Fodor claims is a truism — that ‘the mind’s main concern is not acting but thinking, and that paradigmatic thinking is directed to ascertaining truths’ (Fodor, 2008, p. 8). It follows that if the mind’s main business is to ascertain truths, and if mental representations are the tools for conducting such business, then they must aim at truth. Thus contents must be at least truth-apt. Ascertaining truth is typically a risky affair. A formal requirement on genuinely representational contents of this sort is that they might be false. Representation always admits of the possibility of error or misrepresentation. Some hold that certain types of mental states, such as perception and memory, are factive (Hopkins, 2010). Such states necessarily reflect the facts. They come with a cast-iron epistemic guarantee: if a token mental state of that kind represents it as the case that p, then it is the case that p. Here’s the rule: If it should turn out that the content of the state in question does not represent the facts then we are not dealing with a mental state of that kind. Note that factive here qualifies the mental states in question, not their contents. To be a representational content that expresses p always allows for the possibility that not-p. Now if this were the received view in orthodox cognitive science about the kind of content possessed by mental representations, which play a role in the Information-Processing Challenge, then it would be directly at odds with the kind of enactivist approach that Thompson defends. But things are not so simple. The first thing to note is that linguistically-mediated beliefs and utterances are conceptual representations. They represent intensionally (with an ‘s’), under guises; they represent things as this-or-that. As Fodor (2008) insists ‘if a symbol represents a such and such, it must represent it as something or other’ (p. 178). However, many theorists believe in the existence of non-conceptual content. This is traditionally understood as a kind of representational content which presents the world as being a certain way, despite the fact that the creature or system doing the representing lacks the concepts that would canonically express the content in question. Now it’s a nice question whether we can really make sense of the idea of a kind of truth-conditional content that is non-conceptual. We might worry that we can’t so long as we think that intensionality (with an ‘s’) and concepts necessarily go together. However, this worry can be avoided if there is a kind of content that is subject to norms other than those to do with truth and falsity. The

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

51

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

idea that this might be so is gaining in popularity. Thus, Crane (2009) rejects what he calls the ‘propositional attitude thesis’ about perception without surrendering the idea that perceptual states possess representational content. Instead, he claims such states have accuracy and correctness conditions that are not any kind of truth conditions. A main motivation for this is the observation that accuracy and correctness come in degrees whereas truth or falsity do not. Crane compares experiences to pictures in this respect. Like pictures, he holds, experiences can be more or less accurate, but they are not intrinsically true or false. Nor can pictures stand in logical relations. Thus although Crane thinks they have a kind of representational content he denies it is of the truth-conditional sort.2 They have a kind of content that is more primitive, more basic than that had by propositional attitudes such as beliefs. Fodor (2008) has advanced a similar line. He observes: ‘pictures don’t have truth conditions. In the root case, for a symbol to be true it has to pick out an individual and property and predicate the latter of the former; but iconic representations have no way to do either. So, the camera doesn’t lie, but nor does it tell the truth’ (ibid., pp. 175–6). This is his way of making room for the ‘possibility that some mental representation is nonconceptual’ (ibid., p. 179, emphasis added). He promotes the idea that there can be representing that isn’t representing as. Thus ‘X represents Y insofar as X carries information about Y, where “carries information about…” is read as transparent… [this allows] for representation that’s not “under a description”’ (ibid., p. 179). The idea that being an information-carrying state suffices for being a mental representation with nonconceptual content is a radical departure from standard thinking about what is minimally required to qualify as a representational state. As noted, many now question the idea that content is necessarily truth-conditional. We are told that ‘it isn’t apparent that an intentional state, event, or object about something other than a state of affairs should be evaluated in terms of truth/falsity’ (Gunther, 2003, p. 5). But it is generally held that ‘what is true of any state (event, experience, and so forth) with content, is that it is governed by semantic normativity. For whether its content is [2]

I am reporting Crane’s view, not suggesting that it is unproblematic. One immediate worry about this proposal is that it is hard to understand what it would be for an experience or picture to be accurate simpliciter. It looks as if to be accurate is to be accurate in this or that respect. And that would seem to imply that a condition on being an experience or picture is that the ways in which the picture or experience might be accurate or not would have to be independently specifiable. If so, that makes it look as for an experience or picture to be accurate or not (to whatever degree) depends on its being used for a particular purpose that determines in what respect it might be so.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

52

D.D. HUTTO

conceptual or nonconceptual, propositional or not, an intentional state presents the world as being a certain way; and intrinsic to this presentation, to its content, is a set of (semantic) conditions under which it does this correctly, truthfully, satisfactorily, appropriately, skillfully, and so on’ (ibid., pp. 5–6). On this view, ‘semantic normativity is the mark of intentionality’ (ibid., p. 6). We can assume, along with Millikan (2005) that the normativity at play here is of the nonevaluative (e.g. possibly of the merely biological) sort. But even if symbols and concepts are not involved or crunched, and even if this isn’t any kind of intensional (with an ‘s’), truth-conditional representing, there is more going on than being in states that merely carry or contain information in the sense Fodor describes. What appears to matter to those who invoke the normative requirement is the character of the organism’s response, understood as involving some kind of norm other than truth. I will pick up this thread in a more positive vein in the next section. Fodor’s proposal, however, is apparently much more radical than this. It allows that there can be mental representation with content but without ‘semantic normativity’ in the reduced sense just described. Simply being in a state that registers information suffices for representing (but not representing as). But there’s a tension in his calling nonconceptual states with this sort of content mental representations. For Fodor (2009) tells us that ‘The mark of the mental is its intensionality (with an ‘s’) that’s to say that mental states have content; they are typically about things… only what’s literally and unmetaphorically mental has content’. If this is taken to mean that intensionality (with an ‘s’) is a minimal, necessary requirement for genuine mentality then it turns out that states bearing only informational content not only lack semantic properties, they are not truly mental. This fits with Fodor’s earlier verdicts about the sorts of creatures that belong to the class of cognizers or mentalizers; ‘wherever precisely the line is to be drawn, and however thick it may be, it is vastly plausible that we fall on one side and the paramecium fall on the other’ (Fodor, 1986, p. 12). I shall pick up this thread more positively in the next section too. Frankly, all of this makes my puzzler sore. Content, it seems, is a bit like Christmas: it can come without truth conditions, without concepts, without intensionality (with an ‘s’), without semantics, without mentality. Content, I guess, means something more. It would be nice — very nice, indeed — if there were some agreed, unequivocal, nonslippery and fully stable understanding of just what ‘content’ and ‘representation’ are; given that mainstream cognitive science apparently

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

53

depends on these notions so heavily. It is perhaps too much to say that the history of the use of these terms is shrouded in equivocation, equivocation, equivocation. But it is no exaggeration that despite their utterly foundational importance to orthodox cognitive science they are extremely elastic, far from unambiguous and not yet very wellunderstood. One thing is certain. There is no point in looking to our pre-scientific folk intuitions to decide the matter. As Jackson and Pettit (1993) point out, ‘“Content” is a recently prominent term of art and may well mean different things to different practitioners of the art’ (p. 269). Nonetheless, Crane (2009) assures us that his usage of the term, which deviates from the propositional attitude rendering, corresponds to the way that many professional philosophers use it. To back that up he cites the pedigree of this usage, reminding us that the notion of content belongs with the theory of intentionality that Brentano offered us. But, be this as it may, it does not by itself provide us with a robust and perspicuously clear understanding of content or its properties, nor even any ready way to demarcate it from other phenomena. Indeed, as Crane (2008) has recently admitted in response to a challenge by Nes (2008), making sense of intentionality itself seems to require calling on the notion of representation to do foundational work. He says: ‘It is the notion of representation, I think, that will distinguish intentionality from… other phenomena’ (Crane, 2008, p. 216). If so then we must call on our notion of representations to do important work in demarcating intentionality. But presumably we need a notion of content in order to distinguish representational phenomena from nonrepresentational phenomena. And, as we just noted, a notion of intentionality is needed to help us understand the notion of content. So it seems that trying to make sense of these notions in this way is to move in a rather tight, and seemingly incestuous, circle. And we must be on our guard here for another reason. Brentano’s understanding of intentionality is complex; it embeds more than one notion — he speaks not only of being directed at objects but also of such objects having intentional inexistence (see Menary, 2009, for an excellent exegesis). There are different ways of understanding these ideas in today’s context. Yet those who look to Brentano for a lead on the nature of intentionality typically end up making appeal to a notion of representation or content that is modelled directly on the kind of semantic content (whether truth conditional or referential) associated with linguistically mediated states of mind such as propositional attitudes. It is by this route that many of today’s philosophers come to endorse what I will call the thesis of semantic intentionality. Flanagan

54

D.D. HUTTO

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

(1991) supplies us with a neat reminder of how the standard thinking goes on this issue in his discussion of the central tenets of James’s philosophy of mind. The concept of intentionality is a medieval notion with philosophical roots in Aristotle and etymological roots in the Latin verb intendo, meaning ‘to aim at’ or ‘point toward’. The concept of intentionality was resurrected by and clarified by… Franz Brentano… Brentano distinguished between mental acts and mental contents. My belief that today is Monday has two components. There is my act of believing and there is the content of my belief, namely, that today is Monday… Beliefs are not alone in having meaningful intentional content… Language wears this fact on its sleeve. We say that people desire that [ ---- ], hope that [----], expect that [----], perceive that [----], and so on, where whatever fills the blank is the intentional content of the mental act. Intentionality refers to the widespread fact that mental acts have meaningful content…The fact that we are capable of having beliefs, desires, or opinions about non-existing things secures the thesis that the contents of mental states are mental representations, not the things themselves — since in the case of unicorns, ghosts, devils and our plans for the future there simply are not real things to be the contents of our mental states! On this interpretation, James is an advocate of what Jerry Fodor calls the representational theory of mind. (Flanagan, 1991, p. 28, second and third emphases added)

Going this way only takes us back to square one. Just as the notion of content is a term of art so too is representation as Millikan (1993) reminds us, ‘the name “representation” does not come from scripture’ (p. 103). The short exercise of this section is meant to remind us that, after all, as Matthen (2006) helpfully notes, representation is ‘a new and controversial concept… The natural home of this concept is in the study of communication between agents who possess intentions and goals. It is not immediately clear how it can be extended to states issued by automatic sub-personal systems’ (p. 147). Thus relying on our intuitive grasp of what these notions mean renders us none the wiser about the core nature of intentional, representational or contentful properties.

4. A Fresh Start It seems we need a different approach to these issues. I suggest putting aside our antecedent philosophical commitments and intuitions about the nature of ‘representation’ and ‘content’ and working forward from an agreed and non-controversial understanding of the nature of information, as the notion called upon in a variety of sciences. This should be the lowest common denominator in driving our thinking about what is needed to understand basic mentality and any advance beyond

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

55

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

it requires justification. The question is — does basic cognition or mentality require information processing? Before we get started — just what is the most basic kind of cognition or mentality? One suggestion is that ‘cognitive interactions are those in which sensory responses guide action and actions have consequences for subsequent sensory stimulation, subject to the constraint that the system maintain its viability. “Sensory response” and “action” are taken broadly to include, for example, a bacterium’s ability to sense the concentration of sucrose in its immediate environment and to move itself accordingly’ (Thompson, 2007, p. 125). There seems no good reason to rule this out as an instance of cognition or mentality, albeit basic, other than attachment to the idea that true cognition or mentality must involve symbols and concepts. But, as we saw in the previous section, even champions of a more restrictive understanding of cognition have apparently begun to waver on this point. If so, then as Thompson (2007) argues, the thesis of deep continuity between life and mind is secure, so long as we adopt a liberal understanding of autopoiesis as ‘internal self-production sufficient for constructive and interactive processes in relation to the environment’ (p. 127). Any living creature capable of this will need to be informationally sensitive (see Hutto, 1999, chapters 2 & 3; 2008, chapter 3). But it doesn’t follow that they need to process information — if this means that information is some sort of commodity that is in some way contentful, as such talk appears to suggest. Godfrey-Smith (2007) distinguishes two senses of information, a weaker and stronger one. The weak notion is the familiar one that derives from the work of Shannon and which has played a pivotal role in the development of communication technology. It assumes that informational relations are nothing more than covariance relations; they exist wherever correlations between facts, events or properties obtain. Let us call this the information-as-covariance notion. As Godfrey-Smith notes this conception of information is ‘unproblematic’ and does not require much philosophical attention.3 It has a richer cousin that is much more controversial. It is referred to as ‘semantic’ or ‘intentional’ information, the kind of contentful information — the message — that some communications convey.

[3]

This notion of information has two features that make it of great value to the naturalist. It doesn’t presuppose what it sets out to explain — i.e. the existence of semantic content — and it is resoundingly non-mysterious, having been put to good work in a number of hard sciences. Following Wheeler (2005) we can say that it is entirely muggle — there’s nothing magical about it.

56

D.D. HUTTO

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

That notion significantly adds to the basic Shannon notion. Let us call it the information-as-content notion. There is a real danger of conflating these two notions. Jacob (1997) tells us that: the relevant notion of information at stake in informational semantics is the notion involved in many areas of scientific investigation as when it is said that a footprint or a fingerprint carries information about the individual whose footprint or fingerprint it is. In this sense, it may also be said that a fossil carries information about a past organism. The number of tree rings in a tree trunk carries information about the age of the tree… In all of these cases, it is not unreasonable to assume that the informational relation holds between an indicator and what it indicates (or a source) independently of the presence of an agent with propositional attitudes. (Jacob, 1997, p. 45, emphasis added)

To stress this last point, he adds that ‘the information or indication relation is going to be a relation between states or facts… It is an “objective” relation’ (ibid., pp. 49–50, emphasis added). There is no doubt that information-as-covariance is an objective relation. And as the quotation suggests it has wide currency in a number of sciences. But to talk of informational semantics and to speak of indication when describing it, as Dretske (1981; 1988) does, courts confusion with its richer, sister notion of information-as-content. As Cummins et al. (2006) point out, ‘“indication” is just a semantic-sounding word for detection’ (p. 200). This being so, equating information and indication relations is doubly problematic — not only is there a risk of smuggling in a notion of semantic content where it does not belong, there is also the fact that the idea of detection undermines the idea that the information in question is a purely objective relation; it makes no sense to talk of detection in the absence of an agent (or equivalent) that does the detecting. This highlights something important. It suggests that if we only stick with the weak notion of information then there are no grounds for thinking that the world, standing apart from agentive systems, contains anything that could be called informational content. To see what’s at stake it helps to consider Cummins et al.’s (2006) attempt to expose a special problem for teleosemantics, by noting that we really have no choice but to believe in the existence of unexploited content, content of a sort that must exist independently and logically prior to the capacity of systems to make use of it. Apparently, this is a problem for teleosemantic theories because they insist that natural signs or signals lack representational, semantic content until a consuming response is selected for, one that governs a system’s reactions. The

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

57

worry raised is that cognitive systems are surely able to come to use previously unexploited content — either by individual learning or evolution in the species. But this apparently presents teleosemantics with a conundrum for it ‘requires content to pre-date selection and teleosemantics requires selection to pre-date content’ (Cummins et al., 2006, p. 199). Although interesting in its own right, it is not my purpose here to review the argument advanced by Cummins and co. against teleosemantics. Rather I want to highlight an observation they make at a crucial juncture when considering possible replies. For it is important to their argument that the content in question is that allegedly contained in representations used by cognitive systems. Thus they write: ‘A very natural response is to say that unexploited content isn’t really content. After all, there is a lot of unexploited information in the environment, information that cognitive systems must acquire abilities to exploit. We do not call that information content’ (ibid., p. 204, emphases original). But in cases of basic mentality — that of paramecia — this is exactly the kind of situation we have to deal with. Agents are interacting in reliable, informationally sensitive but non-contentful ways with their environments. Why non-contentful? Well, as we have just seen, if we stick to the information-as-covariance notion there is no content out there for them to interact with (let alone register, pick up, and so on). If there’s no objective content in the world, then perhaps content comes into being along with the activity of agents or consuming systems. This idea lies at the heart of teleosemantics. As Millikan (2006) says, ‘the content of a representation is determined, in a very important part, by the systems that interpret it’ (p. 100). But, once again, the metaphors can mislead. We should not think of agents as content-consuming systems — for this suggests that there is already pre-existing content to be consumed; and we have just ruled that idea out. Perhaps then we should speak of content-creating systems instead. That’s a step in the right direction, but note — now we are getting very close to the enactivist story. After all, Thompson (2007) maintains that ‘Cognition is behaviour in relation to meaning and norms that the system itself enacts or brings forth on the basis of its autonomy’ (p. 126, emphases added). There is doubtless much to be learned and perhaps salvaged from teleosemantic theories — which are widely regarded as the best, if still imperfect, attempts to naturalize representational content. Perhaps, with modifications, such accounts might help to augment autopoietic enactivism. Some contemporary

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

58

D.D. HUTTO

proponents of enactivism believe that the basic idea requires supplement by appeal to additional notions. We are told, ‘It is a mistake to take the theory of autopoiesis as originally formulated as a finished theory… autopoiesis leaves many questions unanswered. In particular, several essential issues that could serve as a bridge between mind and life (like a proper grounding of teleology and agency) are given scant or null treatment in the primary literature’ (Di Paolo, 2009, p. 12). Even if we accept this, it is unwise and unnecessary, for the reasons gestured at above, to buy into the teleosemanticists’ semantic ambitions. Indeed, despite initial optimism, many now doubt that attempts to naturalize semantic content have any chance of success. GodfreySmith (2006) provides an astute assessment: ‘there is a growing suspicion that we have been looking for the wrong kind of theory, in some big sense. Naturalistic treatments of semantic properties have somehow lost proper contact with the phenomena’ (p. 42). Nevertheless, he also acknowledges that the driving idea behind teleosemantics — that evolved structures can have a kind of ‘specificity’ or ‘directedness’ — is essentially correct: ‘there is an important kind of natural involvement relation that is picked out by selection-based concepts of function. But this relation is found in many cases that do not involve representation or anything close to it’ (ibid., p. 60). What should we make of this? To quote a famous Rolling Stones lyric, ‘You can’t always get what you want, but if you try sometimes, you just might find, you get what you need’. Teleosemantic accounts fail to provide an adequate basis for naturalizing semantic or intensional (with an ‘s’) content but they are proceeding along the right lines. Crucially, with adjustment they provide serviceable tools for making sense of something more modest — i.e. organismic responses involving intentionality (with a ‘t’). What if in the place of teleosemantics we put teleosemiotics? Teleosemiotics is an order of ‘teleosemantics — hold the semantics’. Teleosemiotics borrows what is best from teleosemantics and covariance accounts of information to provide a content-free naturalistic account of the determinate intentional directedness that organisms exhibit towards aspects of their environments (Hutto, 2008, chapter 3). Yet unlike teleosemantics, it does not seek to understand the most basic forms of directedness, such as registering, in semantic, contentful or representational terms. Such modes of responding are not to be understood as content-involving or even content-creating to the extent that these notions are understood in terms of reference or truth conditions. Compare this with Thompson’s discussion of virtual milieus, vital norms and meaning as essential features of cognition, as inspired by

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

59

Merleau-Ponty’s The Structure of Behaviour. As he stresses, even bacterial cells, the simplest life forms on earth (where life is understood autopoietically) have needs that are fulfilled by deriving nutrition from the environment. They achieve their ends by ingesting sucrose — an environmental feature. Following Merleau-Ponty, Thompson (2007) holds that the property of being a nutrient is a virtual property, not something found ‘objectively’ in the environment. Rather ‘it is enacted or brought forth by the way the organism, given its autonomy and the norms its autonomy brings about, couples with the environment’ (p. 74). Thus, ‘sucrose has meaning and value as food but only in the milieu that “the system itself brings into existence” or “constitutes for itself”’ (ibid.) The whole point is that ‘Behaviour… expresses meaning-constitution rather than information processing’ (ibid., p. 71). This way of talking will seem alien, or misplaced, to many analytic philosophers but Thompson is quite clear that he is concerned with norms in ‘the biological sense of norms’ (ibid., p. 75). Moreover, the ‘meaning’ and ‘value’ that are brought forth neither constitute nor depend on semantic content. Nor does it imply the existence of representations in anything like the standard sense. We are told ‘if we wish to continue using the term representation, then we need to be aware of what sense this term can have for the enactive approach… autonomous systems do not operate on the basis of internal representations… they enact an environment’ (ibid., pp. 58–9). Substantively I agree, but it will only breed confusion to use terms like ‘meaning’ and ‘representation’ to describe the cognitive antics of bacteria — hence I prefer the more austere teleosemiotic talk of informationally-sensitive responses to natural signs. But this comes to much the same thing. Can a modified teleosemantics — i.e. teleosemiotics — serve as a secure point of contact between what phenomenologically-inspired enactivists have to offer and some of the best work in the analytic tradition on theories of content? Making allowance for differences in language and connation I think the answer is clearly ‘Yes’. However, there are more important twists in this tale. Thompson (2007) rejects the idea of evolution as driven by external forces, such as natural selection. This is not to say he denies the existence or importance of natural selection, only that he objects to the standard interpretations of it. So, this is something that philosophers of biology need to debate — I say no more about it here. Rather I want to focus on another aspect of his view of ‘enactive evolution’, one that has a direct bearing on the issue at hand. It is his rejection of the idea

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

60

D.D. HUTTO

that organisms are ‘systems that have atomistic traits as their proper parts’ (ibid., p. 203). To accept this requires surrendering the idea that we should understand the intentionality of biologically basic cognitive systems as being a property of their individual mental states. While this will no doubt shock some — and goes against the grain of the standard way of talking in analytic philosophy of mind — there is much to be said for it. For on reflection there is every reason to think that the intentionality exhibited by basic cognizers differs significantly from that of beings whose thinking or perceiving takes the form of, or involves, propositional attitudes. To think otherwise would be to imagine intentionality as a property of individual mental states — i.e. states of mind that bear special kinds of mental content. This would be to model such states of mind on isolated words or sentences in the heads of thinkers, however weakly. To be wholly free of this idea we ought to think of the intentional directedness as simultaneously focused on both virtual and actual worldly targets and as involving the goaldirected activity of the whole organism. Consider that, in the style of David Attenborough, we say of baby, Sheba or Rover that they are trying to do this or achieve that. Moreover, we say that they succeed or fail because of what they know, think, notice or sense. Notice — it is the activity itself, and not some sub-part of it, that we can coherently regard as being successful or not. Deciding if it is, or not, requires appeal to some set of norms that specifies the goal in question. For this we must make appeals to the creature’s evolutionary history, individual learning or the norms of an established practice, and so on. Whether a bit of goal-directed organismic activity succeeds, or not, depends on whether certain facts obtain. Well-designed organisms have many (and often quite complex) means of responding to the natural signs of environmental correspondences that are important to them. Responding to such signs is meant to guide their behaviour with respect to the state of the world so they succeed in their activities. And, if they are well-built and conditions are normal, their activities non-accidentally succeed often enough to fulfil their needs. All of this can be true without it being the case that some sub-part of the organismic system — e.g. an internal mental state — contentfully represents some part of the external world correctly or incorrectly by saying that it stands thus or so. Indeed, in very basic cases there is no principled basis for picking out one segment, or part, of a much larger organismic response to some external natural sign as a discrete, contentful state of mind that represents some more distal state of affairs. In normal conditions it is the totality of an organism’s response

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

61

that ensures the non-accidental success of its activities. As such, it is the attitude of the whole organism engaged in such activities that exhibits intentional directedness. It is the response as a whole that targets certain aspects of the world and not some sub-part of the response. If so, it must be possible to be intentionally directed without having discrete mental states that possess any kind of mental content at all. I call such non-contentful but world-directed attitudes — intentional attitudes. They are to be contrasted with properly contentful, sententially-mediated propositional attitudes, such as truth-conditional beliefs and desires. The attitudes of the latter sort do possess semantic content and linguistic structure. Indeed, I have long held that our ‘ordinary concept of belief ranges over cases which, from the philosophical point of view, we should distinguish as instances of beliefs-as-propositional-attitudes and beliefs-as-intentional-attitudes’ (Hutto, 1999, pp. 109–10). To have a content-involving thought, it is not enough for an organism to be merely intentionally directed at a situation or state of affairs, even in the sorts of complex and systematic ways intimated above. A creature could engage in many highly sophisticated activities while only having attitudes of an intentionally directed sort that are to be understood in purely non-intensional (with an ‘s’) terms.

5. Conclusion Adequately responding to the Information-Processing Challenge requires acknowledging the special importance of the informational sensitivities of sentient and sapient systems and understanding these correctly. This requires resisting the temptation to think of information as a kind of contentful, object-like commodity, or to otherwise assume that basic mentality depends on the manipulation of content-bearing mental states. Where does this leave us? On the positive side, this conclusion is consistent with accepting that cognitive systems exploit the relations of covariance that hold between environmental states of affairs in various ways. This idea is consistent with — and, indeed, inspired by — the idea that well-fashioned organisms are responsive to natural signs and that these can guide actions successfully (in historically normal conditions), even if such signs do not supply the creatures’ cognitive mechanisms with contentful information and even if the signs are left semantically uninterpreted. That said, successful action requires informational sensitivity and a kind of responsiveness to natural signs that introduces asymmetries

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

62

D.D. HUTTO

(e.g. an organism’s sensitivity to one state of affairs enables it to respond appropriately to a more distal state of affairs — in historically normal circumstances). To be sure, organismic actions do not always succeed. But the mere possibility of worldly misalignment does not imply (and need not be explained in terms of) the existence of semantic relations of truth and reference. How does this help the prospects of enactivism? The appeal to teleosemiotics shows that the Information-Processing Challenge can be defused by enactivists without abandoning their core motivating insight. Now, it might be thought that even if contentful representations are not necessary for understanding basic mentality, surely this cannot be the whole story about cognition. True. But we can go a long way (even if not the whole way) in making sense of very complex, elaborate and sophisticated worldly engagements without assuming they are either constitutively contentful or contentfully mediated. The great bulk of living, thinking organisms act successfully by making appropriate responses to objects or states of affairs in ways mediated by their sensitivity to natural signs. But this does not involve contentfully representing those objects or states of affairs as such or even representing them non-conceptually. Basic forms of mentality depend on informational sensitivity and response, sometimes of a quite sophisticated variety, not processing informational content. Undoubtedly, some states of mind exhibit semantic intentionality — propositional attitudes, for instance. They are properly contentful. Nevertheless, a great deal of sophisticated, world-directed cognition exhibits intentional directedness that is not contentful in the sense just discriminated. Teleosemiotics understands on-line perceptual responding as informationally sensitive but it rejects the idea that this equates to a purely informational kind of nonconceptual representing. It denies that such responding constitutes ‘a way of representing X without representing it as anything’ (Fodor, 2008, p. 182). Radical Enactivism — of the sort that both Thompson and I promote — explicitly rejects the idea that content, whether informational or representational, is an inevitable ingredient in the process that enables basic mentality. In this, not only is this brand of enactivism wholly in line with the spirit of the original and most philosophically challenging conception of enactivism, it is independently wellmotivated. Surely, it will be objected, some behaviour is too off-line, ‘plastic and flexible’ to be explained without appeal to the manipulation of contentful propositional attitudes and symbolic representations. Just

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF MIND’S NEW LEASE ON LIFE

63

how far can we go? Much further than is commonly thought, I think, before we need to introduce anything like contentful states of mind into the picture (see Hutto, 2008, chapters 4 & 5). Still, there are obvious limit cases. Certain kinds of deliberative planning — such as acting on the basis of considerations that are explicitly represented — must be content-involving. We should accept that ‘the ability to think the kind of thoughts that have truth-values is, in the nature of the case, prior to the ability to plan a course of action. The reason is perfectly transparent: Acting on plans (as opposed to, say, merely behaving reflexively or just thrashing about) requires being able to think about the world’ (Fodor, 2008, p. 13). There are two things to note about this remark. First, if the above arguments hold then we should resist the idea that all worldly engagements that do not involve sophisticated, contentful deliberation and symbol-crunching are nothing but bits of non-cognitive reflex or ‘thrashing about’. Secondly, we must ask exactly who — i.e. which cognitive beings — are capable of reflective planning and deliberation? If that class contains only us — i.e. adult, linguistically competent and typically developing human beings — beings who have benefited from a wealth of specialized social scaffolding through engaging in communal practices — then we will want a story about how we get in a position to do so. Thompson-style enactivism promises to tell that story. It may be possible that we can account for the contentful or meaningful basis of such activities without gaps and without having to believe in the existence of underived mental contents of the sort that have resisted naturalistic explanation for so long. Perhaps, after all, we have reason to believe the enactivist credo, that practice logically precedes theory, and not — pace Fodor — the other way around. If so, philosophy of mind gets a new lease on life.

References Clark, A. (2008) Supersizing the Mind: Embodiment, Action, and Cognitive Extension, Oxford: Oxford University Press. Crane, T. (2008) Reply to Nes, Analysis, 68, pp. 215–218. Crane, T. (2009) Is perception a propositional attitude?, Philosophical Quarterly, 59, pp. 452–469. Cummins, R., Blackmon, J., Byrd, D., Lee, A. & Roth, M. (2006) Representation and unexploited content, in MacDonald, G. & Papineau, D. (eds.) Teleosemantics, pp. 195–207, Oxford: Oxford University Press. Di Paolo, E.A. (2009) Extended life, Topoi, 28, pp. 9–21. Dretske, F. (1981) Knowledge and the Flow of Information, Cambridge, MA: MIT Press. Dretske, F. (1988) Explaining Behaviour: Reasons in a World of Causes, Cambridge, MA: MIT Press.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

64

D.D. HUTTO

Flanagan, O. (1991) The Science of the Mind, Cambridge, MA: MIT Press. Fodor, J.A. (1986) Why paramecia don’t have mental representations, in French, P.A. (ed.) Midwest Studies in Philosophy, pp. 3–23, Minneapolis, MN: University of Minnesota Press. Fodor, J.A. (2008) LOT 2: The Language of Thought Revisited, Oxford: Oxford University Press. Fodor, J.A. (2009) Where is my mind?, London Review of Books, 31 (3). Godfrey-Smith, P. (2006) Mental representation and naturalism, in MacDonald, G. & Papineau, D. (eds.) Teleosemantics, pp. 42–68, Oxford: Oxford University Press. Godfrey-Smith, P. (2007) Information in biology, in Hull, D. & Ruse, M. (eds.) The Cambridge Companion to the Philosophy of Biology, pp. 103–119, Cambridge: Cambridge University Press. Gunther, Y.H. (2003) General introduction, in Gunther, Y.H. (ed.) Essays on Nonconceptual Content, pp. 1–19, Cambridge, MA: MIT Press. Hopkins, R. (2010) Factive pictorial experience: What’s special about photographs?, Noûs, DOI: 10.1111/j.1468-0068.2010.00800.x Hutto, D.D. (1999) The Presence of Mind, Amsterdam: John Benjamins. Hutto, D.D. (2008) Folk Psychological Narratives: The Socio-Cultural Basis of Understanding Reasons, Cambridge, MA: MIT Press. Jackson, F. & Pettit, P. (1993) Some content is narrow, in Heil, J. & Mele, A. (eds.) Mental Causation, pp. 259–282, Oxford: Oxford University Press. Jacob, P. (1997) What Minds Can Do: Intentionality in a Non-Intentional World, Cambridge: Cambridge University Press. Matthen, M. (2006) Teleosemantics and the consumer, in MacDonald, G. & Papineau, D. (eds.) Teleosemantics, pp. 146–166, Oxford: Oxford University Press. Menary, R. (2009) Intentionality, cognitive integration and the continuity thesis, Topoi, 28, pp. 31–43. McGinn, C. (1989) Mental Content, Oxford: Basil Blackwell. Millikan, R.G. (1993) White Queen Psychology and Other Essays for Alice, Cambridge, MA: MIT Press. Millikan, R.G. (2005) Language: A Biological Model, Oxford: Oxford University Press. Millikan, R.G. (2006) Useless content, in MacDonald, G. & Papineau, D. (eds.) Teleosemantics, pp. 100–114, Oxford: Oxford University Press. Nes, A. (2008) Are only mental phenomena intentional?, Analysis, 68, pp. 205–215. Oyama, S. (2000) The Ontogeny of Information: Developmental Systems and Evolution, Durham, NC: Duke University Press. Papineau, D. (1987) Reality and Representation, Oxford: Oxford University Press. Ramsey, W.M. (2007) Representation Reconsidered, Cambridge: Cambridge University Press. Seager, W. (2000) Theories of Consciousness, London: Routledge. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press. Varela, F.J., Thompson, E. & Rosch, E. (1991) The Embodied Mind: Cognitive Science and Human Experience, Cambridge, MA: MIT Press. Wheeler, M. (2005) Reconstructing the Cognitive World, Cambridge, MA: MIT Press.

Albert Newen

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Merits and Limits of a Philosophy of Autopoiesis I Thompson describes his main aims in his book, Mind in Life, as follows: On the one hand, I try to show that to be a living organism is physically to realize or instantiate a certain kind of self-organization — one that entails an autonomous and normative and cognitive mode of being in relation to the world. On the other hand, I try to show that certain features of the human mind, especially various structural features of conscious experience, are constituted by self-organizing processes of the human body engaged with its environment. (Thompson, 2011, p. 10)

Let us presuppose for the sake of argument that the biological background claim is true: life is constituted by autopoietic organization, i.e. by self-organization and self-maintenance of a system. One important question is: what can be derived from this claim? Thompson argues that being an autopoietic system entails being an autonomous and a cognitive system (while cognition is defined in a broad and a narrow sense as we will see); furthermore he wants to base his central claim of a deep continuity of life and mind on the observation that living beings are autopoietic systems (p. 128),1 especially, he wants to show that structural features of conscious experience are constituted by self-organizing processes. Although we can learn something by considering that living beings are autopoietic systems, his aims are much too ambitious. I try to show why this is so. Correspondence: Email: [email protected] [1]

References which include only a page number are all referring to Thompson’s book, Mind in Life (2007).

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 65–82

66

A. NEWEN

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

II. The principles of (minimal) autopoiesis are not sufficient for cognition: Adaptivity, and not autopoiesis, is the critical feature of cognitive systems Let me start by discussing whether it really follows that any autopoietic system is a cognitive system: it is clear that a basic understanding of autopoiesis is in no way sufficient to support such a claim since autopoiesis is realized by bacteria, by amoeba and paradigmatically by each cell. But although a primitive cell, and even more a plant, needs self-maintaining metabolic processes to persist, this does not involve any cognition per se, if cognition even in a broad sense involves at least some adaptivity in the reactions towards the environment. Autopoiesis in a minimal sense (which Thompson himself describes as ‘internal self-production of the minimal sort’, p. 126), as realized by a cell, may be adequate only according to one type of environment with the effect that the cell dies if the environment changes to another type. Such a cell can only survive as long as it just happens that the environment is very stable. This would be an autopoietic system without involving any cognition even in such a broad sense (let us call them from a biological perspective niche-cells, which immediately die when they leave or lose their niche-condition). Furthermore, Thompson (p. 125) accepts the case of tesselation automata (as described by Bourgine and Stewart, 2004) as a case of an autopoietic system that is neither a biological nor a cognitive system. Biological autopoietic systems can be very primitive: Craig Venter created a minimal bacterium genome by producing a reduced version of the genome Mycoplasma genitalium (Gibson et al., 2008). This new version is called M. laboratorium and it can only realize two activities: growth and reproduction (Boldt and Müller, 2008). This illustrates exactly the core idea of autopoiesis: self-persistence in an environment without ‘interacting’ with it (besides the minimal energy transfer which may be realized by a simple rigid mechanism). The positive result of this aspect is that even such a simple autopoietic system already establishes a self-environment differentiation. But this does not involve any cognition, a fortiori not any consciousness. If self-production and self-persistence is the core principle of autopoiesis then we acquire an important but very limited explanatory role of the concept of autopoiesis: we can only understand the basis of the self-environment differentiation. To allow for the claim of an entailment in the starting quote (‘…a certain kind of self-organization — one that entails an autonomous and normative and cognitive mode of being in relation to the world’),

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

67

Thompson introduces a more demanding reading of autopoiesis, which he calls a narrow reading of autopoiesis, ‘to mean internal self-production sufficient for constructive and interactive processes in relation to the environment’ (p. 127). Autopoiesis defined in that way now involves the property of being sensitive towards the environment; let us identify this with the property of being an adaptive system. If we add the feature of being an adaptive system into the definition of autopoiesis we get of course the result that the remaining types of autopoietic systems are cognitive systems in the broad sense of being able to react to the environmental conditions. But Thompson makes a stronger claim by maintaining ‘that any living system is both an autopoietic and a cognitive system’ (p. 127). Thompson does not constrain his claim to adaptive autopoietic systems but generalizes explicitly over all autopoietic systems. Furthermore, he introduces an even more demanding narrow sense of cognition and tries to argue that living systems are autopoietic and cognitive systems in the broad and narrow sense of cognition. This invites a criticism that helps to mark the limits of his project: first of all, as we have already seen, basic autopoietic systems do not essentially develop cognition in the broad sense of being able to react towards changes in the environment. Let us fix this point: (i)

Minimal autopoiesis neither includes adaptivity nor sense-making (see p. 159; quote next page).

It is important to notice that Thompson accepts that the principles of autopoiesis do not include adaptivity or more demanding properties of cognition. This shows us that we cannot derive any specific cognitive features from the fact that a system is an autopoietic system unless there is an additional argument that the necessary feature of being an autopoietic system is really a critical feature and not just a background condition. But this is not delivered. To develop an interesting view, Thompson first has to add the property of being an adaptive system. Of course, everyone agrees that this is an important biological feature of the enduring living system (those which do not have it can only exist very temporarily since typically all environments are continuously changing). We can accept the claim that systems that are autopoietic and adaptive are minimally cognitive systems. Then all the explanations for cognitive abilities seem to rely on the property of being an adaptive system, not on being an autopoietic system. The central role of adaptivity is in line with standard evolutionary explanations. Thompson seems to rely on the wrong feature to explain our

68

A. NEWEN

cognitive abilities (including consciousness) when he focuses on autopoiesis.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

III. Does autopoiesis plus adaptivity entail sense-making? Adaptivity in a wide sense does not, adaptivity in a narrow sense is the product of an over-intellectualized description of adaptivity Thompson then makes an additional move to argue that the property of being an autopoietic system can carry all the weight he wants to put on it: after he includes (by a new definition of autopoietic systems) the property of being an adaptive system as an essential property for any autopoietic system, he claims that the property of sense-making is essentially connected with it too. Sense-making is defined as a narrow sense of ‘cognition’ and characterized as follows: Cognition is behavior or conduct in relation to meaning and norms that the system itself enacts or brings forth on the basis of its autonomy. We have seen that sense-making requires more than minimal autopoiesis; it requires autopoiesis enhanced with a capacity for adaptivity or assimilation or accommodation. (p. 159)

This quote explicitly supports our analysis above about minimal autopoiesis and the missing emphasis on the property of adaptivity in Thompson’s general claim. Furthermore, we now deal with a narrow sense of cognition which he calls sense-making and the claim that autopoiesis plus adaptivity entails this property of sense-making: ‘According to the view I have been proposing, autopoiesis plus adaptivity entails sense-making, which is cognition in its minimal biological form’ (p. 159). From a logical point of view, the concept of adaptivity is the critical property that should allow us to infer the concept of sense-making. An unproblematic reading of adaptivity would be ‘being tolerant to changes in the environment by registering perturbations and compensating for them’. But such an unproblematic understanding would not justify speaking of sense-making at all if the latter includes a behaviour according to rules and norms and not just according to mechanisms. We do not contrast norms and mechanisms (norm following is in fact realized by special mechanisms) but presuppose that establishing or following a norm involves an internal representation of an expected state (e.g. as the efference copy in the comparator mechanism). There is no evidence that bacteria have such internal representations of expectations: this claim will be defended by an example. Recently it has been argued that bacteria make anticipatory changes in

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

69

their phenotypes in response to environmental events. For example, Mitchell et al. (2009) demonstrated that E. coli bacteria (as well as S. cerevisiae yeast) are sensitive to the order of presentation of lactose and maltose, undergoing gene regulatory changes that adaptively prepare the organism for an encounter with maltose after presentation of lactose (the normal order of presentation in the human gut). Other sugars (sucrose, etc.) do not induce this change. Furthermore, microbial strains exposed to lactose not followed by maltose, over multiple generations ceased to show the maltose-anticipatory response, although they retained the ability to respond adaptively to maltose itself. As sophisticated as this system is, however, its multi-generational nature and the fact that these microbes simply incur a fitness cost rather than being able to adapt individually when, in the experimental condition, they mistakenly gear up for maltose after encountering lactose, argues against the ability of these individual microbes to make sense of the situation, or to be guided by a norm that relates survival of the cell to perturbations in the maltose-lactose relationship. Let us have a look at what adaptivity means according to Thompson: ‘Adaptivity is a special way of being tolerant to challenges by actively monitoring perturbations and compensating for them in relation to the autopoietic identity taken as an internal norm’ (p. 148). Obvious questions are: what does it mean to ‘actively monitor’ perturbations and ‘to compensate for them in relation to the autopoietic identity taken as an internal norm’? An example Thompson gives is a bacterium that swims up a sucrose gradient (p. 147). This makes clear that his description is in danger of massive over-intellectualization: in the case of the bacterium, to ‘actively monitor’ just means to ‘register’ and ‘to compensate for them in relation to the autopoietic identity taken as an internal norm’ only means that the bacterium has a mechanism that directly and determinately influences the redirection of its movements that (in normal conditions) corresponds to the sucrose gradient thereby supporting its survival. In an elegant experiment conducted nearly 40 years ago, Macnab and Koshland (1972) demonstrated that a bacterium (Salmonella strain) does not, as one might have guessed, register a spatial difference in nutrient levels across its ‘body’, but registers a temporal difference in the nutrient levels and maintains its current direction of movement as long as an increase in nutrient level over time is detected. While this is a quite sophisticated mechanism, that might even be described as involving memory at the cellular level, it seems not very helpful to speak of a norm involved in this mechanism. Even if we account for the internal information-

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

70

A. NEWEN

processing in the bacterium registering the nutrient difference, there is no internal representation of an expected state. Thompson wants to speak of a norm here, the norm ‘implied by an organism actively seeking to improve the conditions of self-production’ (p. 147). This reference to a norm is at best superfluous because it just describes the established mechanism that supports or optimizes the survival conditions of the bacterium. The system behaves as if there were a norm but since there is no internal representation of expectations, we do not have any reason to speak of norms in such a system. The over-intellectualization seems to be a consequence of transforming the evolutionary process of the development of a mechanism for the whole species into an ‘internal norm’ for an individual autopoietic system, one bacterium. There is no clear evidence for describing the ability of a bacterium as compensating for perturbations ‘in relation to the autopoietic identity taken as an internal norm’ instead of simply claiming that it involves a mechanistic reaction supporting de facto its survival. The story how such a mechanism has been established can then be a standard evolutionary one. To summarize: adaptivity in an unproblematic sense as a product of mutation and selection processes does not imply sense-making. Adaptivity as understood by Thompson involving norms does, but without downplaying the surprising sophistication of single cells it is a massive over-intellectualization of the adaptive behaviour of cells (as paradigmatic cases of living beings). We should distinguish three kinds of autopoietic systems: (1) minimally autopoietic systems, (2) minimally autopoietic systems which in addition are adaptive, and (3) minimally autopoietic systems which are adaptive and involve sense-making where the latter means that their behaviour is (at least partly) relying on internal representations of expected states (feed-forward mechanisms). If one accepts these distinctions then it follows that the important basic properties to explain the behaviour of living beings are (1) adaptivity, and (2) feed-forward mechanisms, while autopoiesis is just a minimal background condition to explain the existence of the system by establishing a system-environment differentiation. This allows us to correct the picture Thompson prefers. We are especially lacking any convincing support for the claim that the principles of autopoiesis are intrinsically connected with sense-making, i.e. with cognitive abilities involving feed-forward mechanisms.

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

71

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

IV. The claim of a deep continuity of mind and life The central aim of Thompson’s discussion of autopoiesis and cognition is the claim about a continuity of mind and life: ‘For my purposes it is enough to maintain that any living system is both an autopoietic and a cognitive system, for this thesis is sufficient to establish a deep continuity of life and mind’ (p. 127). Let us have a closer look at what the claim about a deep continuity of mind and life means: again we find a weak and a strong interpretation in the book. The weak interpretation is again trivial and generally accepted while the strong interpretation is false. Let us start with a weak reading of the thesis of a deep continuity of life and mind. In a careful and epistemological interpretation Thompson writes: ‘Then the projects of understanding life and understanding mind are continuous’ (p. 128). This is a bit more concrete in the following sentence: ‘The usual way to express this idea is that the organizational properties of mind are an enriched version of those fundamental to life’ (p. 157). Even the second quote allows for a weak reading according to which the mind has its own principles, i.e. organizational properties, which are additional to the principles of life. Life is presupposed to establish a mind, but according to a weak reading, the principles of the mind are not very much constrained by the principles of life. The weak reading of Thompson’s main claim that there is the continuity of life and mind can be seen as illustrating the evolutionary process from living systems to cognitive/minded systems. This is a very plausible claim but it is neither original nor controversial: all scientists (who do not deny evolution) presuppose a more-or-less gradual development of the mind in the evolution of living beings. Furthermore, this does not justify the claim that all living systems are cognitive systems but only that some (by evolution) are transformed into cognitive systems by a gradual development. Explaining mind as a product of natural evolution on the basis of already established living beings is a basic claim of naturalism, and it is different from a strong interpretation according to which the principles of mind are essentially constituted by the principles of life (p. 128). The latter again has two readings: it is uncontroversial that (a) cognitive abilities of a living being are influenced by evolutionary conditions in general (including environmental conditions, the condition of living in groups in the case of humans, etc.) But it is very controversial that (b) cognitive abilities are essentially shaped by those principles that are constitutive for being an autopoietic system. So there are two main criticisms: (1) only some

72

A. NEWEN

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

living beings develop cognitive abilities in the sense of being adaptive systems with feed-forward mechanisms, and (2) there is no convincing argument that the autopoietic principles are at the same time essential for our cognitive abilities; they seem to be just general background conditions (compare: although it is important that I am a living being in order for me to learn mathematics — or e.g. to drive a car — my success and my progress in doing so cannot be understood or explained by this fact. The psychological principles of learning mathematics or driving a car are not entailed by the principles of being a living system).

V. Autopoiesis and life: What can we learn from artificial life? Thompson accepts that a tessellation automaton model can be seen in principle as an autopoietic system which is not a biological system, i.e. he accepts that autopoietic systems are not co-extensional with biological systems although the tessellation automaton is so far only a computational simulation (pp. 111–3). Despite some scepticism about embodied autopoietic systems being different from biological systems based on cells, he grants that autopoietic organization is a structure: ‘In an autopoietic framework, minimal life is not identified with any particular molecular structure, such as RNA/DNA, but rather with a bounded, self-producing concatenation of processes, which can in principle be structurally realized in different ways’ (p. 118). Given this view, Thompson is open-minded about artificial life because it is then only an engineering question whether we can construct autopoietic systems that are different from biological systems, which are the product of natural evolution (i.e. mutation and selection processes). There are two strategies actually applied in research programmes nowadays: the J. Craig Venter Institute successfully proposes a top-down strategy, i.e. they start with an existing biological entity and then substitute key natural parts of this entity by a synthetically created analogue (Glass et al., 2006; Gibson et al., 2008). The research question is whether all natural parts can be substituted by artificially created parts. If this can be done, nothing seems to prevent us from starting a direct construction of artificial life. This is already the strategy of Tadashi Sugawara and his group who work with a much more demanding bottom-up approach: they try to create the components that are necessary for a minimal autopoietic system. They try to construct an artificial cell using well-defined organic molecules and bio-polymers, and they have studied nonlinear dynamics exhibited by self-reproduction of giant vesicles and self-replication of

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

73

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

informational molecules. Coupling between these two dynamics is expected to lead to a primitive artificial cell (Toyota, Takakura and Sugawara, 2004; Shohda and Sugawara, 2006). Of course, it is an open question whether the bottom-up approach will work, but why should there be any principled barriers? Are there any radically new elements in biological systems, which would prevent the creation of life? Let us have a look at Thompson’s views. He presupposes the Kantian principle that life is intrinsically teleological. He then modernizes this claim as follows: In an autopoietic system, the continued existence of the membrane and internal reaction network is possible thanks to their mutual self-production. What is so startling about this self-production, from the standpoint of classical linear mechanics, is that a self-perpetuating whole emerges out of local processes while subsuming those processes so that they do no longer have a merely local and independent identity. It is precisely this view of circular causation and nonlinear emergence that was unavailable to Kant. (p. 138)

This explication is open to a naturalistic understanding which does not involve any non-scientific principles in the explication of life. So far, there is no important dispute. But then Thompson simply states that we have to adopt a transcendental stance to explain life: What is it that enables us, as scientists and philosophers, to cognize or grasp the phenomenon of autopoietic selfhood? What are the conditions of possibility for our recognition and comprehension of this form of existence? Would autopoietic selfhood be disclosable from some disembodied, objective standpoint? Or, rather, are we able to cognize this form of existence because it resembles the form of our own bodily selfhood and subjectivity, which we know firsthand? (pp. 162–3)

These questions lead Thompson to the claim that life can only be known by life (p. 162).

VI. Can life only be known by life? In the quote above, Thompson seems to discuss only one alternative view when he is actually posing (at least) two different questions: (1) What are the conditions of having knowledge of an autopoietic system? and (2) What are the conditions of having knowledge about the subjectivity of an autopoietic system, i.e. especially its first-person perspective?

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

74

A. NEWEN

Concerning question (1), we can come closer to an answer if we presuppose that knowledge is a representational state (or an ability) which we attribute to an entity to explain its behaviour (or reactions) — thereby we exclude metaphorical speech like ‘the universe knows that…’ Given this plausible presupposition we need a distinction between a representational system and its environment. The theory of autopoiesis gives us a plausible explanation of this self-environment differentiation such that we can say that autopoietic systems are systems that possess knowledge. But are they the only ones? Modern robots with complex behavioural abilities, e.g. in spatial navigation, have knowledge about their environment but they are so far not autopoietic systems. Therefore, autopoietic systems are not the only candidates for knowledge. With question (2), the focus is now on knowledge about a first-person perspective: do we need a first-person perspective to be able to have knowledge about it? Here the main criticism is that Thompson does not distinguish clearly enough between knowledge in general and empathic knowledge. It is undisputable that I need a first-person perspective to develop empathic knowledge of the subjectivity of another entity, but is that true for any knowledge? Thompson’s implicit premise is this: to have knowledge of a subjective phenomenon (e.g. first-person perspective) the knowing entity must be acquainted with this subjective phenomenon. Let us benefit from Nagel’s famous paper ‘What is it like to be a bat?’ — Nagel argues that we cannot know what it is like to navigate like a bat since we do not have an echo-based sensory system. This is true for empathic knowledge but not for knowledge in general: we have extended knowledge of the spatial orientation mechanism in bats and presupposing a non-dualistic world this is, at the same time, general knowledge about the subjective experience of the bat. Although this is not empathic knowledge, it is third-person knowledge about a subjective phenomenon. One reply could be that such a third-person knowledge is incomplete. Yes, it is, but most of the time we have incomplete knowledge about the world. This is also true for knowledge about other people: as healthy persons we can only have nonempathic knowledge of the subjective experience of a person suffering from persecution mania, but we can have a lot of third-person knowledge, and this allows psychiatrists to help them. Let me add one more quote concerning this question of knowing life: ‘The proposition that life can be known only by life is also a transcendental one in the phenomenological sense. It is about the conditions of

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

75

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

the possibility of knowing life, given that we do actually have life’ (p. 164). This quote makes clear that Thompson is presupposing the acceptance of a version of Kant’s transcendental paradigm. Let me remind here that there are core problems with a Kantian paradigm (for a detailed critique of the transcendental paradigm see Newen and Vogeley, 2008) that motivated modern philosophy of mind in its majority to leave a transcendental framework: (1) We have to presuppose an unknowable thing-in-itself that on the one hand should cause our sensory stimulations, but on the other hand we never can say anything about a thing-in-itself (including the impossibility of making any claim about the relation between it and our sensory stimulation, thereby self-defeating the motivation to introduce a thing-in-itself). (2) We have to presuppose a transcendental ‘I’ and an empirical ‘I’ without any possibility to determine the relation between them. The radically open question is: how is transcendental self-consciousness related to empirical selfconsciousness? These two levels involve a special variety of dualism. (3) Like any dualism such a transcendental dualism also does preclude any understanding of a mental causation of transcendental self-consciousness. Furthermore, if there would be no causal influence, it would be only an epiphenomenon. (If someone is an epiphenomenalist then there is no need to accept a transcendental paradigm at all.) Therefore both consequences are highly problematic. Since a transcendental paradigm leaves us with a lot of problematic presuppositions leading to unanswerable questions and into some kind of dualism, we should at least try to avoid such a loaded paradigm. Thus we lose the only foundation of the claim that life can only be known by life.

VII. Consciousness explained? How does Thompson try to explain phenomenal consciousness? He takes it to be immediately connected with self-regulation and thereby with autopoiesis: Our mental lives involve three permanent and intertwined modes of bodily activity — self-regulation, sensorimotor coupling, and intersubjective interaction (Thompson and Varela, 2001). Self-regulation is

76

A. NEWEN

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

essential to being alive and sentient. It is evident in emotion and feeling, and in conditions such as being awake or asleep, alert or fatigued, hungry or satiated. (p. 243)

In this quote, Thompson’s central claim again is that autopoiesis via the property of self-regulation is essential for phenomenal consciousness. Although it is true that, in humans, basic feelings are connected with realizing processes of self-regulation (Damasio, 1999), there is no evidence that this is essentially so: processes of self-regulation are realized in many engineered systems (starting from primitive heating control systems up to complex robots). If we grant that autopoiesis involves self-regulation, we simply can remind the reader that the principle of autopoiesis is only a necessary condition of consciousness but is by no means sufficient: a cell does not, even plants do not, enjoy consciousness, but they are paradigmatic cases of autopoietic systems. Since Thompson accepts this point at another place in his book, there remains at least some tension in his book and we have to search for additional properties that may constitute consciousness. The last quote about our ‘mental lives’ continues then by characterizing sensorimotor coupling and intersubjective interaction: ‘…hungry or satiated. Sensorimotor coupling with the world is expressed in perception, emotion, and action. Intersubjective interaction is the cognition and affectively charged experience of self and other’ (p. 243). If we focus on the property of a system to realize sensorimotor couplings we do not find any evidence that helps to explain consciousness: technical systems like modern robots realize sensorimotor couplings when they are programmed to realize complex tasks. Robots are standardly equipped with sensory systems and are able to learn to adjust their behaviour according to the sensory input, and readjust expected states on the basis of sensory feedback. Some robots are nowadays programmed on the paradigm of dynamical systems theory that is also presupposed by Thompson. Both new features, sensorimotor couplings and being a dynamic system, are not essentially connected with phenomenal consciousness. Although intersubjective interaction is an important aspect shaping the development of our mind, humans are already born with basic phenomenal consciousness that is active before intersubjective interaction can develop. Therefore these features are all not essential for phenomenal consciousness. What we additionally find in the book is that Thompson makes a transcendental move to explain consciousness: ‘Consciousness considered in this way is a condition of possibility for there being any appearances at all. To understand consciousness this way is to

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

77

understand it transcendentally and not merely empirically’ (p. 240). Here we would like to express the same kind of careful scepticism about the fruitfulness of the transcendental paradigm (see above). Thompson defends his view about consciousness by criticizing the zombie argument: he argues that it is relying on a Cartesian view of the mind which should be substituted by a view that the mind is essentially connected with life. We can accept this critique, the analysis and the consequence. But as we argued already, knowing that our mental abilities are relying on autopoiesis as a necessary background condition does not allow us to explain specific mental abilities (including consciousness). On the basis of Thompson’s criticism of the zombie scenario he presupposes that the only alternative is to accept the nonreducibility of consciousness. Let us have a closer look at the debate about zombies as presented by Thompson. He grants the logical possibility of a zombie scenario but denies its possibility given our laws of nature. We can accept this analysis. But it does not imply that consciousness is a necessary feature of autopoietic systems with self-regulation and sensorimotor couplings. Since these structural features can be observed at least in basic formats in technological systems, we are lacking support for grounding consciousness in these features. Furthermore, it is possible to develop a detailed argument against Thompson’s view. He claims that ‘consciousness as epistemic base is equivalent to the experiential acts by which those objects are disclosed to us’. I understand him such that this implies that consciousness is the essential feature responsible for unifying the information into object representations as part of a meaningful world. Although this sounds intuitively plausible, it neglects the knowledge we have acquired about unconscious representations of objects in humans that guide our behaviour: in another paper we have summarized many empirical studies which show that representations with the same content can be processed either consciously or unconsciously (Vosgerau, Schlicht and Newen, 2008). All these examples show that typical object features and object representations are unconsciously available and relevant for behaviour. Let us pick out three famous cases: the patient D.F. suffering from visual agnosia caused by a lesion in the ventral cortex is not able to consciously register the orientation of a turnable letter box. But if D.F. holds a letter in her hand and is asked to put it into the box, she is as good in doing that as any normal person. The information about the orientation of the letter box is registered by D.F., and it is available for the motor system guiding the grasp but it remains unconscious. The same information can be acquired

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

78

A. NEWEN

unconsciously or consciously and it can trigger the same type of grasping behaviour (Milner amd Goodale, 1995). There are lots of analogous cases concerning unconscious representations of typically conscious representations of objects or features of objects: in the case of patient P.S. suffering from hemineglect, the content that could not be attended to remained unconscious, although it reliably directed decisions and was therefore represented (Marshall and Halligan, 1988). Furthermore, people suffering from blindsight do not consciously perceive an object but are nevertheless able to grasp it and reliably make a ‘forced choice’ guess as to what object it is, whether its colour is A or B, or whether it is located in C or D (Weiskrantz, 1986; 2006). The lesson to draw from these examples is that the registration of objects is not essentially dependent on consciousness and that the registration of information that is seen as paradigmatically connected with consciousness, e.g. colours, can also be registered unconsciously and thereby influence our behaviour. There are several unconscious processes going on that are relevant to unify information as belonging to an object (sensorimotor integration, etc.) which are not connected with consciousness, such that conscious considerations are only one factor influencing the structuring of the world. The contrasting picture is this: we often do not rely on conscious experience to represent objects or features of objects (including phenomenal features like colour, etc.) Consciousness clearly makes a difference in our representation of the world but it is not the only factor disclosing objects and making them relevant for our behaviour. We may conclude this part by denying two strategies of explaining consciousness: (i)

Consciousness cannot be fruitfully analysed as a feature of life (even if we grant that life is a necessary condition of having consciousness). Consciousness may be described as a product of evolution but this is not sufficient for Thompson’s aims. (ii) Consciousness cannot be analysed as a special content (content and consciousness are orthogonal) or a principle of producing disclosed contents, since many contents can be registered consciously and unconsciously (Vosgerau, Schlicht and Newen, 2007) and the unification of information into meaningful units influencing our behaviour (disclosing objects) also takes place independently of consciousness.

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

79

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

VIII. Does enactment constitute phenomenal consciousness? Thompson stresses the role of enactment for mental phenomena and subscribes to Noë’s claim that the conscious experience of an object is constituted by sensorimotor contingencies. Since enactment is another core paradigm put forward by Thompson, I shortly present a central criticism of this strategy to explain consciousness as it is developed in detail in Noë’s work (see also the review by Schlicht and Pompe, 2007). The main claim of an enactment theory of consciousness is this: we are able to consciously see an object because we have a perspectival conscious experience and we have implicit knowledge of sensorimotor contingencies — this is implicit ‘knowledge of the way sensory stimulation varies as a function of movement’ (Noë, 2004, p. 117). Even if we grant that such knowledge is important in the unification of perception-based information to construct an object-representation, Noë’s strategy suffers from a problematic infinite regress. The strategy starts with a perspectival conscious experience, e.g. seeing a glass from one special perspective with an elliptical opening should be integrated by implicit knowledge about all the other perspectives into a three-dimensional perception of a glass with a round opening. But how are we able to have the initial conscious perspectival experience? Noë wants to generalize his theory such that every conscious experience is constituted by sensorimotor contingencies: ‘To experience a look (to see a look) is to make use, in experience, of a particular sensorimotor profile; it is to draw on our repertoire of sensorimotor skills’ (ibid., p. 87). He also notices the danger of an infinite regress: ‘After all, one would need to experience the looks of the looks in order to see how things are, and so on, ad infinitum’ (ibid., p. 87). But he never gives us any strategy to prevent it. Even if we grant that enactment plays a role in the constitution of three-dimensional object representations, what we find is only an explanation of a three-dimensional non-perspectival experience on the basis of a perspectival experience. The enactment strategy does not offer any explanation for the constitution of consciousness.

IX. Concluding remarks Despite the criticism, Thompson’s book makes an important contribution by putting a new research area on the agenda of the philosophy of mind: to connect the biological perspective of living systems with the principles of the mind (for other attempts to initiate such a project see

80

A. NEWEN

Fitch, 2007; Ginsburg and Jablonka, 2007a,b). A fruitful consideration is a paradigmatic move Thompson describes at the beginning:

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

A thorough investigation of behaviour at multiple levels of complexity reveals that even inorganic material processes, as well as living and mental ones, are structured unities rather than multiplicities of events external to each other and bound together by efficient causal relations. Specifically, behaviour is always a structured and dynamic whole, in which organism and milieu participate not as externally related stimulus and reaction, but as internally related situation and response. (Thompson, 2011, p. 11)

I accept that the investigation of behaviour at different levels of complexity is an important strategy for discovering insights about cognitive systems. It is also plausible that cognitive systems of a special kind develop representations that allow us to distinguish the internal and the external dimensions of the system itself. Here autopoiesis is one structural principle realizing such a self-environment differentiation. It is also an important insight to consider not only the influence of the environment on a living system but also the shaping of the environment by a living system. But I presented several observations and arguments to show that we cannot benefit much from a biological perspective focusing on autopoiesis to explain mental phenomena. What is really coming into play are several other structural elements which are also discussed by Thompson but which are not essentially connected with biological living systems:

· sensorimotor couplings · features of interaction with the physical environment · features of interaction with the social environment. Furthermore, we have at least to add:

· the informational organization of internal representations · the functional roles of internal representations. I argued that to account for these dimensions of cognition does not involve an acceptance of a transcendental paradigm or some axioms of phenomenology. What we need is a new sensibility for these structural dimensions when analysing the phenomena. Since I do not want to smuggle in any dualism, these structural elements can still best be seen as features of the functional organization of a cognitive system that are essentially realized in the case of human beings by the brain (or by the brain-environment interaction). But can such a view not easily be disputed by mentioning arguments from the extended and enacted mind debate? Such an alternative view can account for these

PHILOSOPHY OF AUTOPOIESIS: MERITS & LIMITS

81

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

arguments by distinguishing (i) the acquisition condition of a mental representation (with the weak externalist claim that the acquisition of mental representations presupposes an interaction with an external world), and (ii) the constituting condition (with the strong externalist claim that each instantiation of a mental representation involves an interaction with the external world). One can easily accept the weak and nevertheless deny the strong claim, e.g. by showing that a mental representation can be meaningful due to the internal organization of a cognitive system relative to the standard world even if this no longer exists. But this would be a different paper (for criticism concerning the extended mind thesis see Adams and Aizawa, 2001). A functionalist or an identity-theorist can still account for the phenomena discussed in Thompson’s book, especially since both lines of theories are compatible with dynamical systems theory. The only aspects we have to leave out are the transcendental paradigm and some axioms of phenomenology. But this is an advantage when aiming at a parsimonious explanation.2

References Adams, F. & Aizawa, K. (2001) The bounds of cognition, Philosophical Psychology, 14, pp. 43–64. Boldt, J. & Müller, O. (2008) Newtons of the leaves of grass, Nature Biotechnology, 4, pp. 387–389. Bourgine, P. & Stewart, J. (2004) Autopoiesis and cognition, Artificial Life, 20, pp. 327–345. Damasio, A.R. (1999) The Feeling of What Happens: Body and Emotion in the Making of Consciousness, New York: Harcourt Brace. Fitch, W.T. (2007) Nano-intentionality: A defense of intrinsic intentionality, Biology & Philosophy, 23, pp. 157–177. Gibson, D., et al. (2008) Complete chemical synthesis, assembly, and cloning of a mycoplasma genitalium genome, Science, 319, pp. 1215–1220. Ginsburg, S. & Jablonka, E. (2007a) The transition to experiencing: I. Limited learning and limited experiencing, Biological Theory, 2 (3), pp. 218–230. Ginsburg, S. & Jablonka, E. (2007b) The transition to experiencing: II. The evolution of associative learning based on feelings, Biological Theory, 2 (3), pp. 231–243. Glass, J.I. et al. (2006) Essential genes of a minimal bacterium, Proceedings of the National Academy of Sciences, 103, pp. 425–430. Macnab, R.M. & Koshland, D.E. (1972) The gradient-sensing mechanism in bacterial chemotaxis, Proceedings of the National Academy of Sciences, 69, pp. 2509–2512. Marshall, J.C. & Halligan, P.W. (1988) Blindsight and insight in visuo-spatial neglect, Nature, 336, pp. 766–767. Milner, A.D. & Goodale, M.A. (1995) The Visual Brain in Action, New York: Oxford University Press. [2]

I want to thank Colin Allen for fruitful comments and especially for the integration of the detailed description of the bacteria examples in section 3.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

82

A. NEWEN

Mitchell, A., Romano, G.H., Groisman, B., Yona, A., Dekel, E., Kupiec, M., Dahan, O. & Pilpel, Y. (2009) Adaptive prediction of environmental changes by microorganisms, Nature, 460, pp. 220–225. Nagel, T. (1974) What is it like to be a bat?, Philosophical Review, 83 (4), pp. 435–450. Newen, A. & Vogeley, K. (2008) Grundlegende paradigmen in der philosophie des geistes, in Spät, P. (ed.) Zur Zukunft der Philosophie des Geistes, pp. 93–124, Paderborn: Mentis. Noë, A. (2004) Action in Perception, Cambridge, MA: MIT Press. Schlicht, T. & Pompe, U. (2007) Rezension zu: A. Noë, Action in Perception, Zeitschrift für Philosophische Forschung, 61 (2), pp. 250–254. Shohda, K. & Sugawara, T. (2006) DNA polymerization on the inner surface of a giant liposome for synthesizing an artificial cell model, Soft Matter, 2, pp. 402–408. Thompson, E. & Varela, F. (2001) Radical embodiment: Neural dynamics and consciousness, Trends in Cognitive Sciences, 5, pp. 418–425. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press. Thompson, E. (2011) Précis of ‘Mind in Life: Biology, Phenomenology, and the Sciences of Mind’, Journal of Consciousness Studies, 18 (5–6). Toyota, T., Takakura, K. & Sugawara, T. (2004) Temporal emergence of giant vesicles accompanied by hydrolysis of ammonium amphiphiles with a Schiff-base segment, Chemistry Letters, 33 (11), pp. 1442–1443. Vosgerau, G., Schlicht, T. & Newen, A. (2008) Orthogonality of phenomenality and content, American Philosophical Quarterly, 45, pp. 329–348. Weiskrantz, L. (1986) Blindsight: A Case Study and Its Implications, New York: Oxford University Press. Weiskrantz, L. (2006) The case of blindsight, in Velmans, M. & Schneider, S. (eds.) The Blackwell Companion to Consciousness, pp. 175–180, Oxford: Oxford University Press.

Susan Oyama

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Life in Mind Commentary on Evan Thompson’s ‘Mind in Life’ I found Thompson’s recasting of the sciences and philosophy of mind to be timely, wide-ranging, and accessible, as audacious as it is meticulous. The author repeatedly confronts the Cartesian split between mind and body, as it appears in various guises, in various literatures. Each time he works through it by integrating methods and findings from more sources than cognitive science or, I take it, phenomenology typically draws on. Along the way, he takes on mental representations, as well as several favoured debating props of philosophers of mind, including creepy zombies and naked brains in vats, exposing their frequently odd assumptions and persuasively pressing his enactive alternative. At each explanatory gap he highlights the interdependence between insides and outsides, stressing immersion and intimate engagement: in stark contrast to much of the current literature, these are living, embodied minds open to their worlds. In short, I am impressed. I am also something of an interested party. First, of course, as a member of an intellectual community that benefits from judicious logjam-busting and well-wrought critique. Other commentators will surely address Thompson’s accomplishments on these fronts, and will assess his overall contributions to phenomenology and philosophy of mind. My own remarks will be rather more narrowly focused, on his utilization of the tradition in which I work, Developmental Systems Theory, or DST. My specific interests are in the ways in which kindred intellectual traditions can be brought into conjunction. Correspondence: Email: [email protected]

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 83–93

84

S. OYAMA

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

In DST, the effort has been to elaborate a nondichotomous, systems-based account of development and evolution, while Thompson aims to demonstrate the continuity of mind and life. These are related goals, and the general affinity between the two perspectives has long been evident. Both rely, for instance, on self-organization and emergent forms rather than trying to explain complexity by prime causes and preformation; they share a mistrust of representationism (whether mental or genetic).1 Both emphasize concrete embodiment and temporality rather than abstract information-processing, as well as the interdependence between living entities and their surrounds. And yet they descend from different conceptual forebears, have faced different challenges, and have developed different tools and tactics. Having been engaged in an ongoing conversation with proponents of autopoiesis/enaction over the years, and having had my moments of uncertainty, I welcome this opportunity to join our author in revisiting these issues.

Articulating Autopoiesis with DST Not long ago I wrote a short piece (2009b) on some aspects of interand intra-theory connections,2 distinguishing between ‘friends’ and ‘neighbours’. These were not separate groups, but kinds of relations. The former are cited for documentation or support, and are often friends in the sense of being enemies of my enemy. As such they can be mustered for limited duty against a common opponent. More often than not this is done despite substantial lack of agreement on many other matters: the concordance bar is fairly low. With ‘neighbours’, on the other hand, the bar is set higher. The focus is now on the precise relations between conceptual systems and styles, the lexicons, beliefs and habits that allow scholars to work comfortably beside each other, to share tools and chores, perhaps even to set up housekeeping together. Away from the polemical frays in which potential conflicts can be ignored in the interests of solidarity, I suggested, we face questions about degrees and kinds of agreement: ‘What are our sticking points, and are they resolvable or terminal? How important is it to work through our various disagreements? Do early commitments preclude our being any more than theoretical neighbors, harboring similar complaints and Genetic representations include the instructions, recipes, blueprints and other intentional, semantic ‘information’ said to reside in the DNA. ‘Genocentrism and computationalism thus run on the same conceptual fuel’ (Thompson, 2007, p. 174; unless otherwise noted, all page references are to this book). [2] The theory of autopoiesis was one of those treated, but at the time I had read only parts of Thompson’s book and was referring mainly to earlier works. [1]

LIFE IN MIND

85

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

irritations but kept apart by our histories, differences of academic culture and other impediments to mutual integration?’ (Oyama, 2009b, p. 148). In Mind in Life, Thompson periodically draws on DST arguments, especially in his Chapter 7, on development and evolution. Given our considerable common ground, we are certainly friends in the above sense. He seems to go beyond recognizing those overlaps, however, when he says he considers the theories of autopoiesis and DST to be complementary: the former, he says, gives an account of the internal dynamics of self-producing living beings and the latter, their coupling with their environments (p. 193). To me, complementarity implies something more than general kinship and ready citability, and it is these more finely articulated relations that take my interest here.3 How closely do these two approaches match up on the issues of boundaries, levels, causality, and control that have been so important to them both? Do significant mismatches limit the fruitfulness of their theoretical engagement? These are not necessarily matters that needed to be treated in a book that already takes on so much,4 but perhaps they bear some consideration now. My intent, then, is not to hedge my endorsement of Thompson’s book, but to explore the kinds of ties that can be forged between scholarly enterprises that have their own complicated backstories. To begin with, I have virtually no argument with the points Thompson supports with developmental systems references, whether with respect to genocentrism in developmental and evolutionary biology, nature-culture dualisms, or infotalk in biology and cognitive science. Something like the enactive approach seems just what is needed in the study of the mind. In addition, I was particularly struck (and gratified) by his deployment of my developmental-systems reformulation of nature and nurture (Oyama, 2000b). I have called them the product and process of ontogeny, thus removing the possibility of treating them as contrasting terms. In like manner, Thompson proposes dismantling the opposition between self-organization and natural selection in evolution: these, too, he asserts, can be seen, not as opposing forces, but as process and product to each other (p. 216). This seems to me both perspicacious and timely. While it will presumably be detailed and amplified, its immediate effect is to undermine the prevailing tendency to treat selection as a designing or shaping agent, an external force that can only be ‘constrained’ by internal selfAlthough the notion of complementarity can be used loosely, for the relationship between psychology and sociology, say, I have in mind something more exacting. [4] Though Thompson does devote some space to them (pp. 193–4). [3]

86

S. OYAMA

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

organization. At the same time, the move keeps faith with the thoroughgoing systems orientation of the book as a whole. Using a notion of complexity as ‘neither random nor ordered and predictable… exhibiting changing and unstable patterns’ (p. 40) for instance, pushes back against the unhelpful but ubiquitous insistence on depicting evolution as a combination of chance and necessity (see also Oyama, 2009a). My lingering hesitation, then, is less about the use Thompson has made of DST in this work than the precise articulations between the autopoietic and developmental systems that are so crucial to his arguments. Presumably we’re not just ‘saying the same thing’, for our aims are not identical, but I’m hoping for enough inter-translatability (or at least reciprocal comprehension) at the borders that we can engage in worthwhile trade across them, for near neighbours we certainly are. These matters may fall beyond the concerns of most readers, but I hope that my discussion will offer something even to those not seeking this kind of close-up scrutiny: not only do such negotiations go on all the time, among and within academic traditions, but they can be part of an individual scholar’s ratiocinations as well — aspects of the process of intellectual evaluation and integration. It seems to me that Thompson’s claim of complementarity implies a fairly full workingout of the sorts of questions mentioned earlier. His footnote (note 15, p. 458) to my past worries about internalism suggests that he has resolved them to his satisfaction. If so, I’m in luck, for I could use a little help in working out the details.

Worries over Internalism In the context of DST’s critique of the widespread privileging of internal causes (typically innateness, ‘the physical’, or DNA) over external ones (the environment, learning, culture, etc.), a certain kind of emphasis on interiors raises red flags. The privileging that worries DSTers often involves attributing asymmetrical causal roles to internal and external factors.5 Thus one kind of cause might be said to bear information or basic form, to control or determine some process, or to specify its possible outcomes, leaving the other class of causes to modulate or interfere with those outcomes, or to select from a preexisting array of potential ones. Thompson has been alert to this issue, [5]

The wariness extends to externalisms, such as environmental determinisms in psychology or the social sciences, or evolutionary adaptationism. Theorists of enaction seem to have as little patience with these as does DST. Indeed, the whole notion of bringing forth a world militates against things like pre-existing niches.

LIFE IN MIND

87

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

and remarks explicitly on it, providing us an entry point. In his footnote 15 he avers that his presentation is innocent of the kind of internalism that has concerned me (Oyama, 2000a,b). Indeed, Thompson describes ‘the dynamic co-emergence of interiority and exteriority’ (p. 79) pointing out (p. 118) that ‘autopoiesis always has to be ecologically embedded. “Self-producing” refers to the kind of circular organization that makes the cell an individual; it does not mean that the cell makes itself apart from its environment’. He starts, in fact, not from the traditional mind trapped in the head, but from a radically open, embodied one. ‘Thus bodily feelings are not self-enclosed’ (p. 23). Living interiority ‘comprises the self-production of an inside that specifies an outside to which that inside is constitutively and normatively related’ (p. 225). I take one of Thompson’s main points to be that the enclosed privacy of the brain-locked mind is part of a Cartesian legacy that must be superseded. His decision (p. 235) to start from ‘lived body’ is pivotal, signalling his departure from much of orthodox cognitive science. This openness extends to subjectivity (p. 36): ‘Generative phenomenology brings to the fore the intersubjective, social, and cultural aspects of our radical embodiment… Individual subjectivity is… intersubjectivity, originally engaged with and altered by others in specific geological and cultural environments.’ This all seems right to me, and important. Yet, having described the mutual generation of insides and outsides, he says (p. 79, quoting Varela) that there still appears to be an asymmetry, ‘for it is the internal self-production process that controls or regulates the system’s interaction with the outside environment’. On the same page is a reference to the ontological priority of insides, based in part on the control of interactions by those insides. The isolation that Thompson rejects is part of the complex I’m calling internalism, but the causal asymmetry sketched above is another, perhaps slipperier one. There may be intensive commerce with the outside, but what controls it? Thompson says (p. 370),6 ‘according to the enactive approach, sensorimotor processes modulate, but do not determine, an ongoing endogenous activity’. Interiors have a special say about their commerce with the outside. Someone with DSTish sensitivities pauses, wondering whether, if we shift our view, the outside can similarly be said to ‘determine’ that activity, while internal factors only modulate (and just what determine means). It is distributed causality and joint control that such a person sees, so the natural question [6]

Similar formulations are common in the autopoiesis literature.

88

S. OYAMA

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

is whether the openness and embeddedness Thompson describes goes with something like DST’s view of system causality, only with different terminology. Our hyper-alert reader might point out that the reasoning used to say that the inside ‘controls’ an interaction or ‘determines’ or ‘specifies’ an effect can just as well be applied to the external influence, which, if it has any effect at all (as it presumably must, or why call it an influence?), ‘controls’ the interaction as well. Indeed, what is deemed a control is to a non-trivial extent a function of what comparison is being made, and therefore what is being bracketed or taken for granted. Given this inside, the outside ‘determines’ the outcome; given this outside, the outcome is ‘determined’ by the inside. But in the end the phenomenon itself is controlled systemically. Not all contributions to an interaction are of the same kind, and a metabolizing, responsive organism or cell has different interactive possibilities than a stone, but for DST the outcome of any interaction is the product of a host of factors, very few of which are typically attended to but whose causal contributions can be made evident and are part of a richer story.

Pragmatic Shifting and Kinds of Systems Now Thompson’s point has to do not with just any old insides, but primarily with the interiors of living beings actively regulating and maintaining their boundaries. Such boundary-making and -keeping activities — such insides — are what autopoiesis was born to describe. A distinction (p. 43), furthermore, between self-governed systems and other-governed ones (autonomous and heteronomous, though perhaps not always under the same labels) has long been part of the autopoiesis literature (Varela, 1979). The autonomous ones have the qualities alluded to in the last paragraph, of internal regulation and endogenous control of interactions with the exterior, while the others are seen from the outside, as it were, as informationcontrolled, input-output mechanisms. This distinction is introduced by Thompson to contrast the enactive approach with traditional cognitivism. It seems that viewing the mind as a heteronomous system is a matter of ‘interpretive stance’, a felicitous phrase I take from his note 22 to Chapter 7 (p. 460), and that implies that the distinction is a matter of attitude rather than ‘the thing itself being interpreted’.7 This would be consistent with the pragmatic cast of both theories. [7]

In note 24 (p. 461), the distinction seems something closer to ontology: ‘organisms are different from heteronomous entities such as automobiles.’ The general impression, however, is of pragmatic relativity.

LIFE IN MIND

89

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Significantly, for Thompson (p. 39), the observer is present in the very definition of a system, a ‘collection of related entities or processes that stands out from a background as a single whole, as some observer sees and conceptualizes things’. That note 22 about interpretive stances is about the possibility of treating an organism not as an autonomous system, but as a designed object, ripe for the reverse engineering so dear to the heart of the evolutionary adaptationist. I was initially bewildered by the idea of treating entities either (or even alternately) as autonomous or as heteronomous, for in DST internal and external factors tend to be noted at the same time, to counteract the privileging of factors on one side of a boundary (and then only some of those factors). It is useful, for instance, to underscore the embeddedness of ‘internal’ processes in a developmental environment that is no mere support or modifier of features already specified from the inside, but that is part of the very dynamics of maintenance and change. The ambit of a developmental system thus extends beyond the focal entity itself. Yet by including developmentally relevant aspects of its surroundings, the system does not thereby exclude its internal relations. For me, living beings’ ‘coupling with their environments’ is part of their ‘internal dynamics of self-producing’. So I wondered how well a developmental system could fill the role of explicating organism-environment coupling as complementary to autopoiesis’s view from the inside. Would the organism then be a heteronomous system ‘defined by input-output information flow and external mechanisms of control’ (p. 43)? If not, are other characterizations available in the autopoietic framework besides the autonomous and heteronomous? If an autonomous system is coupled to an environment and/ or another organism in the mutually-influencing manner described by theorists of autopoiesis and developmental systems alike,8 is each heteronomous when viewed from the vantage point of the other? If ‘the state changes of an autonomous system result from its operational closure and structural coupling’, furthermore (which has a DSTish joint-determination sound to it), how does this square with the declaration (p. 182) that it is the autopoietic network ‘in its entirety that specifies the phenotypic characteristics of a cell’? Autonomy and heteronomy, according to Thompson (pp. 49–50), are ‘heuristic notions’, subject to choice and context. Crucially for the present enquiry, he says (p. 51, citing Varela) that a shift to the heteronomous perspective is simultaneously a shift up — to a ‘larger system [8]

According to Thompson (p. 45), such coupling occurs without exception.

90

S. OYAMA

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

of organism-plus-environment’. It is not, then, simply a movement between views at the same level, like alternating percepts of an ambiguous figure. Is that larger system, then, what I’m calling a developmental system? Must it, in turn, be an autonomous one? Much of the play here seems to be in the meanings of internality and externality, as well as the notion of causal symmetry. It has been suggested to me (by Gregory Mengel, personal communication, June 2009) that Thompson’s internal-external distinctions are less about spatial boundaries (which is how I have tended to read them) than about selfhood, organizational closure, and the context-dependence of causes. In just this issue of boundaries, in fact, he finds what he terms a ‘fruitful tension’ between the two theoretical traditions. This may be right. It would be consistent with what I said at the beginning of this commentary about our different pathways, and I return to this matter at the end. Before moving on, I’ll mention several related issues of enclosure and extent. If, as Thompson (p. 206) says, the ‘unit of evolution’ is the developmental system, which he describes as a ‘milieu-embedded, propagative unit’, is the developmental system the unit that is embedded in a milieu, or is the (relevant) milieu part of the developmental system? On such matters Thompson (e.g. p. 106) can be quite unflappable; he neither demands nor offers premature certainty. Arguably more readily flapped than he, I don’t demand immediate closure either, but do wonder whether any of these questions needs to be addressed, and what, if anything, hangs on them. Here is a last question. Thompson emphasizes ontogenetic emergence, on occasion noting (pp. 380–2) when other theorists have neglected development. Could there be any point at which the developmental transformations that captivate DST exert pressure on the idea of the self-sustaining, self-stabilizing dynamics of an autopoietic system?9 Arrested by these and other uncertainties, I go back to Thompson’s (p. 51) caveat that ‘there is no inconsistency between characterizing the nervous system and organism as autonomous and emphasizing their somatic and environmental embeddedness. We do, however, have to keep our logical and conceptual accounts clear, so that we know which explanatory heuristic is in play at any given time’. I admit that I did not fully assimilate that point about heteronomy entailing a move to a larger, inclusive system and so did not keep it in mind as I [9]

In developmental biology, regulation can refer to the production of a typical result despite perturbation. Perhaps development itself could be thought of as an indefinite series of ‘failures’ of steady-state system regulation.

LIFE IN MIND

91

moved through the book. And as suggested above, I may have been giving his discussions of internal and external relations an excessively concrete, spatial reading. As it is, I suspect the residual distance between us is neither large nor consequential for most purposes. Certainly these bobbles in the rhythm of reading were minor in comparison with the pleasures of following Thompson’s deeply informed analyses and seeing him show how notions of circular causality, embodiment and embeddedness play out in cognitive science.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Inter-System Relations and Lineage Histories The present commentary, then, is the reverse of many I’ve written. Too often I’ve wrestled with pieces that said many of the ‘right’ things but that were, from my point of view, misguided, even benighted. With Mind in Life, by contrast, I am in such sympathy with the general enterprise and so much in agreement with many aspects of the analysis that my efforts are instead directed at dispelling some of my own uncertainties about our apparent convergences and divergences (and so perhaps doing the same for others); increasing the mutual usability of our respective stores of concepts and terms; provoking reflection on the adequacy/suitability of those stores for the discourses we now face and may in the future; and last, raising a question I broached in my piece on theoretical friends and neighbours (Oyama, 2009b), where I compared the two traditions.10 There I speculated that as compatible as they are in many ways, the theories of autopoiesis and developmental systems may be destined to remain near neighbours, not because of any fundamental incompatibility, but simply because we started from different points, in somewhat different circumstances, engaging disparate problems. Rather than seeking to define life and identity, like the theorists of autopoiesis (and later, with the enaction perspective, to explore experience), I had begun with development.11 They found asymmetrical relationships across system boundaries indispensable to their accounts of self-making and self-maintaining, whereas I, struck by how often problematic nature-nurture distinctions were buttressed by asymmetrical language, stressed the importance of treating causal factors on both sides of those boundaries in the same way (and, if it was

[10] See also Oyama (1999), used by Thompson (p. 193) in his own comparison. [11] I use the first-person singular because there is, of course, diversity of personal history in

the DST camp.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

92

S. OYAMA

called for, at the same time): treating them symmetrically as causes.12 I am not able to say whether such symmetry in DST is the practical equivalent of the switch between autonomous and heteronomous views of a system (this is why I pose the question here) — whether alternating views are effectively the same as simultaneous ones.13 As noted earlier, it may well be that the internal and the external carry different meanings in the two literatures. It is fairly clear to me, though, that for one set of workers the integrity of boundaries was paramount, while for the other, it was their permeability. Yet, to talk about developmental interactions across a semipermeable membrane, say, there must be a focal entity in the first place. It is the nature of that entity itself that Thompson and his cohorts address. And in the end there need be no contradiction between integrity14 and openness, which may be simply to re-state Thompson’s remark several paragraphs back, about autonomy and embeddedness. If the two traditions are separated by a kind of sensitive dependence on initial conditions, by these prior choices of battles and implements, and by the entrenchment of certain styles of expression, it is an open question whether that separation can be reduced. It may even, as Mengel ventures, itself be the kind of fertile gap that spurs productive probing and exchange. What seems quite certain is that our associations are sufficiently close that we can enjoy the mutually enriching cooperation and understanding that characterizes the best of neighbours.

References Oyama, S. (1999) Locating development, locating developmental systems, in Scholnick, E.K., Nelson, K., Gelman, S.A. & Miller, P.H. (eds.) Conceptual Development: Piaget’s Legacy, pp. 185–208, Hillsdale, NJ: Lawrence Erlbaum Associates. Oyama, S. (2000a) Evolution’s Eye: A Systems View of the Biology-Culture Divide, Durham, NC: Duke University Press. Oyama, S. (2000b) The Ontogeny of Information: Developmental Systems and Evolution, 2nd ed., revised and expanded, Durham, NC: Duke University Press. [12] Not, as some have apparently thought, that all causes are the ‘same’, either in themselves

or in their effects, or that no distinctions can be made among them. Rather, that they should be judged by the same criteria, and not be deemed primary or secondary, formative or modifying, prescribed or accidental, just because they were on one or the other side of a boundary, or in- or outside the canonical set of ‘real causes’. This is the parity of reasoning that is a DST hallmark. [13] I think of cartoon characters rapidly looking at one thing and then another, whipping their heads back and forth until their heads blur into a single (albeit oddly shaped) outline. Do their objects merge too? [14] Though saying what integrity is will be a challenge once we move away from the closed sets of interacting chemicals of early autopoiesis.

LIFE IN MIND

93

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Oyama, S. (2009a) Compromising positions: The minding of matter, in Barberousse, A., Morange, M. & Pradeu, T. (eds.) Mapping the Future of Biology: Evolving Concepts and Theories, Boston Studies in the Philosophy of Science, Vol. 266, New York: Springer Verlag. Oyama, S. (2009b) Friends, neighbors, and boundaries, Ecological Psychology, Special Edition, 21, pp. 147–154. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press. Varela, F. (1979) Principles of Biological Autonomy, New York & Oxford: Elsevier North Holland.

John Protevi

Mind in Life, Mind in Process Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Toward a New Transcendental Aesthetic and a New Question of Panpsychism Introduction Philosophical linkages often fall into the ‘strange bedfellows’ category, and while for some the linking of cognitive science, biology and phenomenology in Evan Thompson’s Mind in Life (2007) is quite strange enough, this essay seeks to show a further unexpected connection between Thompson’s work and a post-structuralist classic, Gilles Deleuze’s Difference and Repetition (1994). In particular, we will examine the idea of ‘biological space and time’ found in Thompson (2007, pp. 154–7; see also Jonas, 2003, p. 86) and Deleuze (1994, pp. 70–9). As we will see, tracking down the ‘new Transcendental Aesthetic’ found in these thinkers will intersect new work done on a very old idea, panpsychism. Both Deleuze and Thompson can be fairly said to be biological panpsychists. That’s what ‘Mind in Life’ means: mind and life are coextensive; life is a sufficient condition for mind.1 Correspondence: John Protevi, Professor, Department of French Studies, Louisiana State University, Baton Rouge LA 70803 Email: [email protected] [1]

This is a contested reading, but against Wheeler (this volume), I read Thompson (2007) as upholding a coextensivity thesis regarding the relation of mind and life, rather than Wheeler’s enrichment thesis, which would move from life to mind. I can’t fully engage with Wheeler’s rich reading, but a key quote for me in defending the coextensivity thesis is the following: ‘any living system is both an autopoietic and a cognitive system… this thesis is sufficient to establish a deep continuity of life and mind’ (Thompson, 2007, p. 127). In other terms, autopoietic (cellular or multi-cellular) life is sufficient for cognition; where there is such life there is cognition. Leaving aside the ALife question and the ancient hylozoism question, which ask about non-cellular life, the panpsychism question will ask about non-living mind. Panpsychism asks whether autopoietic (cellular or multi-cellular) life is necessary for cognition (mind), or whether there is a defensible notion of mind not just in life, but in process. Is mind a genus of which enactive cognition or ‘Mind in Life’ is a species?

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 94–116

MIND IN LIFE, MIND IN PROCESS

95

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Deleuze is not just a biological panpsychist, however, so we’ll have to confront full-fledged panpsychism. At the end of the essay we’ll be able to pose the question whether or not we can supplement Thompson’s ‘Mind in Life’ position with a ‘Mind in Process’ position and if so, what that supplement means both for his work and for panpsychism. To begin, let’s notice that the Mind in Life position continues the twentieth-century trend of displacing human, language-expressed, top-level, reflective rational consciousness as the sole or prime or most basic or most important candidate for cognition, a position which would have corporeal practical engagement as a privative form, as sloppy or distorted or approximate theory. We see this displacement in the phenomenological tradition, in Husserl’s analyses of passive synthesis and the life world, as well as in Heidegger and MerleauPonty, where embodied practical engagement is primary. In this displacement of rational reflective conscious thought, we also see the connection with Deleuze, who picks up the post-Kantian demand, explicit in Solomon Maimon, for genesis rather than mere conditioning (see the new translation of Maimon’s Essay on Transcendental Philosophy, 2010; for commentary, Jones, 2009; and Smith, 2009). Rather than the Kantian deduction of conditions of human rational reflective consciousness, Deleuze holds we have to show the genesis of real experience from within experience. The post-Kantian challenge relayed by Deleuze is to show how space and time, categories, and Ideas develop in ‘dynamic genesis’, starting with the sheer atomic exteriority of sensations to one another (what Deleuze will call mens momentanea, 1994, p. 70) and moving to ‘virtual Ideas’.2 [2]

Deleuze’s terminology is idiosyncratic, but I believe the rewards of engaging his system are worth the effort. The following can serve as an all-too-brief sketch. In Difference and Repetition we find a tripartite ontological scheme, positing three interdependent registers: the virtual, the intensive, and the actual. Deleuze’s basic notion is thus a tripartite ‘ontological difference’: in all realms of being (1) intensive morphogenetic processes follow the structures inherent in (2) differential virtual multiplicities to produce (3) localized and individuated actual substances with extensive properties and differentiated qualities. Simply put, the actualization of the virtual, that is, the production of the actual things of the world, proceeds by way of intensive processes. In a fuller picture of Deleuze’s ontology, we see that the virtual field is composed of ‘Ideas’ or ‘multiplicities’, which are constituted by the progressive determination of differential elements, differential relations, and singularities; what are related are precisely intensive processes, thought as linked rates of change (Deleuze, 1994, pp. 182–91). Beneath the actual (any one state of a system), we find ‘impersonal individuations’ or intensive morphogenetic processes that produce system states and beneath these we find ‘pre-individual singularities’ (that is, the key elements in virtual fields, marking system thresholds that structure the intensive morphogenetic processes). We thus have to distinguish the intense ‘impersonal’ field of individuation and its processes from the virtual ‘pre-individual’ field of differential relations and

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

96

J. PROTEVI

The key concept shared by Deleuze and Thompson is that the sort of cognition for which Kant posited his transcendental conditions develops from a fundamentally biological cognition, what Thompson calls ‘sense-making’. The Mind in Life claim is that fully conceptual recollection and recognition, the active intellectual relation to past and future, is founded in ‘sense-making’ or ‘enactive cognition’ (what Deleuze will call the linkage of perceptual and organic syntheses). ‘Mind in Life’, read as coextensivity, means that life is fundamentally autopoiesis and cognition is fundamentally sense-making. Thus, ‘Mind in Life’ means ‘autopoietic sense-making’, or the control of action of an organism in its environment. Sense-making here is threefold: 1) sensibility as openness to environment; 2) signification as positive or negative valence of environmental features relative to the subjective norms of the organism; 3) direction or orientation the organism adopts in response to 1 and 2. This founding of human cognition in enactive cognition entails a new transcendental aesthetic, the a priori but always concrete genesis of organic time and space. The essential temporal structure of any organism is the rhythmic production of a living present synthesizing retentions and protentions, conserved conditions and expected needs. The essential spatiality of any organism comes from the necessity of a membrane to found the relation of an organism to its environment; there is an essential foundation of an inside and outside by the membrane, just as there is an essential foundation of past and future by the living present. We thus see the necessity of (at least) a notion of biological panpsychism: every organism has a subjective position, quite literally a ‘here and now’ created by its founding of organic time and space; on the basis of this subjective position an evaluative sense is produced which orients the organism in relation to relevant aspects of its environment (Protevi, forthcoming).

Mind in Life and Biological Space and Time In the de-centring of reflective consciousness we sketch above we see three moves, two familiar, the other being the novelty of the Mind in Life positions. First, the familiar phenomenological move of showing how high-level thought, exemplified in science, is a transformation of the life world or of Dasein’s practical involvement (depending on whether you prefer a Husserlian or Heideggerian vocabulary). This is a synchronic shift of position within adulthood: adults are not first and singularities that make up an Idea or multiplicity. For a more full discussion, see Bonta and Protevi (2004); Smith and Protevi (2008); Protevi (2010).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE, MIND IN PROCESS

97

foremost scientists in everyday life; they are instead practically and corporeally engaged with the world. In other words, we have to show how ‘know-that’ (science) is a transformation of corporeal space-time (a reformed Transcendental Aesthetic) and corporeal ‘know-how’ (a reformed Transcendental Analytic). As Donn Welton puts it, for Husserl: ‘the objects that we do find in Kant’s Analytic, full-blown objects of science, belong to a higher order and are not experientially basic. Constitution at this higher level must be understood not as elementary but as a transformation of what is elemental’ (Welton, 2000, p. 299). Second, another familiar move: genetic phenomenology. Or at least it’s familiar now thanks to the efforts of Donn Welton, Anthony Steinbock, and other ‘New Husserlians’ who have mined the archives containing Husserl’s manuscripts (Steinbock, 1995; Welton, 2000; 2003). Here we have to trace the development of corporeal space-time and corporeal know-how from embryo to adult, that is, along the developmental or ontogenetic time scale. This is where we get a first reformulation of the Transcendental Aesthetic. In The Other Husserl, Donn Welton shows how the Transcendental Aesthetic is renamed in Husserl: instead of the Kantian opposition of sensibility and understanding (judgment) we have Husserl’s opposition of experience and judgment (understanding). Because we have passive synthesis in what Kant would have as merely passive sensibility, there is a noematic sense in perception, prior to active understanding/judgment, and these passive syntheses include associative, kinaesthetic, and time-consciousness syntheses (Welton, 2000, p. 298). Directly addressing Husserl’s genetic undercutting of Kant’s Transcendental Aesthetic, Welton writes: At yet a deeper and final level of genetic analysis Husserl discovers that space and time themselves are not just ‘forms’ but are generated, on the one hand, by the interplay of position, motility, and place, and on the other, by the standing-streaming flow of the process of selftemporalization itself. Husserl’s studies of the self-generation of space and time are clearly the most difficult of all his genetic studies. (Welton, 2000, p. 254)

Our key question, then: is ‘dynamic genesis’ or Husserlian ‘genetic phenomenology’ restricted to the ontogenetic time scale, that is, the development from embryo to adult? If it is so restricted, then we need a new name for the third move, which is the key novelty of the Mind in Life position: we have to do ‘evolutionary’ genetic phenomenology (and not just ontogenetic). In Deleuze’s terms, we have to do dynamic genesis on the evolutionary

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

98

J. PROTEVI

time scale. That is to say, we have to show how single-celled organisms generate their own concrete space and time (a biological Transcendental Aesthetic) as well as display ‘sense-making’ (a biological Transcendental Analytic) AND how this develops along the evolutionary time scale into the potentials for what will develop along the human developmental time scale, that is, genetic phenomenology proper as the constitution of corporeal space-time and corporeal know-how, from embryo to adult. And then finally we can trace the synchronic transformation of corporeal space-time and categories/ Ideas into science or human ‘high reason’. Now if we can have a genetic phenomenology on the evolutionary time scale — if ‘evolutionary genetic phenomenology’ makes sense — then we have to talk about its basis, an empathy condition based on our living experience. To address what the sense-making or affective-cognitive ‘metabolic situation’ of the single-celled organism is, we ourselves have to be living beings. First, here is Jonas: On the strength of the immediate testimony of our own bodies we are able to say what no disembodied onlooker would have a cause for saying: that the mathematical God in his homogeneous analytical view misses the decisive point — the point of life itself: its being self-centered individuality, with an essential boundary dividing ‘inside’ and ‘outside’… (Jonas, 2003, p. 79, italics in original)

Thompson agrees: empathy is a precondition of our comprehension of the vital order, in particular of the organism as a sense-making being inhabiting an environment… [A] bodily empathy… widened beyond the human sphere to ground our comprehension of the organism and our recognition of the purposiveness of life [Thompson here refers to Husserl, Ideas II and Ideas III]. Empathy in this sense encompasses the coupling of our human lived bodies with the bodies of other beings we recognize as living, whether these be human, animal, or even — particularly for biologists with a ‘feeling for the organism’ [Thompson here refers to Evelyn Fox Keller’s biography of Barbara McClintock] — bacteria. (Thompson, 2007, p. 165)

Using this empathy condition to explore the experience of even the simplest living beings, Thompson and Jonas straightforwardly talk of a new transcendental aesthetic as ‘biological time and space’ (Thompson, 2007, p. 155; citing Jonas, 2003, p. 86). We find this expressed as a living present found in the simplest of organisms, a synthesis of retention and protention in the concrete form of metabolism and need (Jonas, 2003, p. 85–6).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE, MIND IN PROCESS

99

For Jonas, a physico-mathematical account misses the ontological emergence that makes of life an ‘ontological surprise’, and the organism a system, a ‘unity of a manifold’. (We will return to the question of emergence and panpsychism.) The organism is ‘whole’ as ‘self-integrating in active performance’; it is an ‘active self-integration of life’ (Jonas, 2003, p. 79). The ‘functional identity’ of organisms relative to the materials it metabolizes is constituted ‘in a dialectical relation of needful freedom to matter’ (ibid., p. 80, emphasis in original). Both elements, need and freedom, constitute the ‘transcendence’ of life, and this transcendence constitutes a living present, a metabolically founded transcendental aesthetic or a priori form of organic time: ‘self-concern, actuated by want, throws open… a horizon of time… the imminence of that future into which organic continuity is each moment about to extend by the satisfaction of that moment’s want’ (ibid., p. 85). For Jonas, echoing Heidegger’s grounding of Dasein’s spatiality in its temporality in Being and Time #70, organic space is founded by organic time: an organism ‘faces outward only because, by the necessity of its freedom, it faces forward: so that spatial presence is lighted up as it were by temporal imminence and both merge into past fulfillment (or its negative, disappointment)’ (Jonas, 2003, p. 85). Jonas then draws the consequences for the question of the adequacy of purely mathematical physics for the phenomenon of life; in other words, he shows the necessity of a dynamic genesis from instantaneity to the living present: ‘with respect to the organic sphere, the external linear time-pattern of antecedent and sequent, involving the causal dominance of the past, is inadequate.’ With life on the scene, ‘the extensive order of past and future is intensively reversed’, so that the determination of ‘mere externality’ by the past has to be supplemented by the recognition that ‘life is essentially also what is going to be and just becoming’ (ibid., p. 86).

Deleuze and Organic Syntheses: Biological Time While Thompson and Jonas discuss space and time together, we will have to first discuss biological time in Deleuze, then, to discuss spatiality, we will have to move to Gilbert Simondon, on whom Deleuze relies for the discussion of membranes. We will then be able to discuss the Deleuze-Simondon position on biological space-time. Chapter 2 of Difference and Repetition is devoted to Deleuze’s work on ‘repetition for itself’. The first step, on which we concentrate, is the discussion of the first passive synthesis of time, or habit, which

100

J. PROTEVI

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

produces the ‘living present’ as the a priori form of organic time.3 To begin his account of dynamic genesis, Deleuze must get down to the most basic synthesis; he must show how beneath active syntheses (thought) are passive syntheses (perception) and beneath passive perceptual syntheses are passive organic syntheses (metabolism).4 The challenge is to describe passive syntheses in differential terms, so as to avoid the ‘tracing’ of empirical identities back to transcendental identities; avoiding such ‘tracing’ is a basic principle of Deleuze’s thought (Smith and Protevi, 2008). Syntheses are needed to join together a disjointed matter or sensation, since in themselves, material or sensory instants fall outside each other, exhibiting ‘a perfect independence on the part of each presentation… one instant does not appear unless the other has disappeared — hence the status of matter as mens momentanea’ (Deleuze, 1994, p. 70). Deleuze goes on to distinguish three levels of synthesis of the zero-degree level of instantaneity: 0. 1.

2.

Instantaneous presentation and disappearance: ‘objectively’ as matter and ‘subjectively’ as sensation Passive syntheses (contraction or habit producing a living present) a. Organic syntheses (metabolism synthesizing matter) b. Perceptual synthesis (imagination synthesizing sensation) Active synthesis (memory as recollection and thought as representation synthesizing perceptions)

Deleuze will distinguish the organic and perceptual syntheses by showing that organic syntheses perform a contraction or induce a habit in their own, material, register. For Hume and Bergson, as Deleuze reads them, the psychological imagination moves from past particulars to future generalities, so that from a series of particulars we come to expect another of the same kind. Deleuze will abstract the process of ‘drawing a difference from repetition’ as the essence of contraction or habit and show that it occurs at the organic level as well as on the level of the passive perceptual imagination (Deleuze, 1994, p. 73). We should note that organic time, the synthesis of habit producing the living present, is only the ‘foundation’ of time. Deleuze’s full treatment of time in Difference and Repetition posits a second synthesis of memory producing the pure past as the ‘ground’ of time, while the third synthesis, producing the future as eternal return of difference, we might say unfounds and ungrounds time. [4] Many of the major commenters on Difference and Repetition — Hughes (2009); Bryant (2008); de Beistegui (2004); Williams (2003) — do not isolate the level of organic synthesis. The exceptions are Ansell-Pearson (1999) and DeLanda (2002). [3]

MIND IN LIFE, MIND IN PROCESS

101

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

In order to isolate organic syntheses as prior to perceptual syntheses (themselves prior to active intellectualist syntheses), Deleuze radicalizes Hume and Bergson, who by themselves leave us at the level of ‘sensible and perceptive syntheses’ (ibid., p. 72). But these perceptual syntheses refer back to ‘organic syntheses’, which are ‘a primary sensibility that we are’ (ibid., p. 73, emphasis in original). Such syntheses of the elements of ‘water, earth, light and air’ are not merely prior to the active synthesis that would recognize or represent them, but are also ‘prior to their being sensed’. So, each organism, not only in its receptivity and perception but in its ‘viscera’ (that is, its metabolism), is a ‘sum of contractions, of retentions and expectations’ (ibid.). Here we see the organic level of the living present of retention and expectation. Organic retention is the ‘cellular heritage’ of the organic history of life and organic expectation is the ‘faith’ that things will repeat in the ways we are used to. So Deleuze has isolated a ‘primary vital sensibility’ in which we have past and future synthesized in what is literally a ‘living present’. At this level, the future appears as need, as ‘the organic form of expectation’, and the retained past appears as ‘cellular heredity’ (ibid., p. 73).5 Now we must distinguish two genres of contraction: (1) contraction as activity in series as opposed to relaxation or dilation, and (2) contraction as fusion of succession of elements. With the second form of contraction, we come upon the notion of a ‘contemplative soul’ which must be ‘attributed to the heart, the muscles, nerves and cells’ (ibid., p. 74). Deleuze knows the notion of an organic ‘contemplative soul’ might strike his readers as a ‘mystical or barbarous hypothesis’, but he pushes on: passive organic synthesis is our ‘habit of life’, our expectation that life will continue. So we must attribute a ‘contemplative soul’ to the heart, the muscles, the nerves, the cells, whose role is to contract habits. This is just extending to ‘habit’ its full generality: habit in the organic syntheses that we are (ibid.) We cannot follow all the marvelous detail of Deleuze’s text in which he discusses ‘claims and satisfactions’ of organic life, going so far as to talk of the ‘beatitude of passive synthesis’ (ibid., p. 74). Nor can we follow his wonderful analyses of rhythm, which will reappear in the notion of the refrain in A Thousand Plateaus (Deleuze and

[5]

Deleuze cannot go directly to his key notion of organic synthesis because he must first free a notion of habit from the illusions of psychology, which fetishizes activity. For Deleuze, psychology, by fear of introspection, misses the element of passive ‘contemplation’. Indeed, current work in philosophical psychology says the self cannot contemplate itself due to fear of an infinite regress of active constituting selves (Zahavi, 2005).

102

J. PROTEVI

Guattari, 1987).6 We have to move to the question of membranes and organic spatiality.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Simondon and Membranes: Organic Space-Time As we have seen with Jonas, the essential spatiality of metabolism comes from the necessity of a membrane to found the relation of an organism to its environment; there is an essential foundation of an inside and outside by the membrane, just as there is an essential foundation of past and future by the living present. The interest of the new biological Transcendental Aesthetic is to see its intertwining of space and time in the relation of membrane and metabolism. Prior to the publication of A Thousand Plateaus in 1980, Deleuze only mentions biological space founded by membranes a few times, always with reference to Simondon. So let us turn to the section of Simondon’s L’Individu et sa genèse physico-biologique entitled ‘Topologie et ontogénèse’ (Simondon, 1995, pp. 222–7) in order to discover what he says about how membranes and metabolism entail a biological Transcendental Aesthetic.7 The basic concept of Simondon’s work is the process of individuation or ‘transduction’ starting from a metastable field.8 In Simondon’s work, a metastable field does not contain individuals; it is pre-individual, but poised for individuation. Simondon’s usual figure for transduction or individuation from a metastable field is crystallization: in the super-saturated field, there are gradients of density, but no crystalline forms nor crystals as individuated entities. There is the potential for crystallization, made actual when provoked by an external shock. From a metastable field, a process of individuation allows for the distinction of an ever-processual individual In Difference and Repetition, Deleuze claims that organic syntheses operate in series, and each series has a rhythm; organisms are polyrhythmic: ‘the duration of an organism’s present, or of its various presents, will vary according to the natural contractile range of its contemplative souls’ (Deleuze, 1994, p. 77). There are thousands of rhythmic periods that compose the organic being of humans: from the long periods of childhood, puberty, adulthood and menopause to monthly hormonal cycles to daily cycles (circadian rhythms) to heart beats, breathing cycles, all the way down to neural firing patterns. Everything has a period of repetition, everything is a habit, and each one of these repetitions forms a living present that synthesizes the retention of the past and the anticipation of the future as need. Now ‘need’ can be ‘lack’ relative to active syntheses, but ‘satiety’ relative to organic passive syntheses. Deleuze writes: ‘need marks the limits of the variable present. The present extends between two eruptions of need, and coincides with the duration of a contemplation’ (ibid.) [7] I will provide my own translation of the Simondon passages. [6]

[8]

Metastability is well-known in dynamical systems theory, serving as a key term in Kelso (1995), for instance.

MIND IN LIFE, MIND IN PROCESS

103

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

and milieu. Individuation as ‘transduction’ is thus an always-ongoing maintenance of metastability between individual and milieu.9 Let us follow Simondon’s treatment of biological space-time, the new biological Transcendental Aesthetic, in L’Individu. To establish the singularity of the living being (le vivant) ‘it would be necessary to exhibit [produire] the topology of the living being, its particular type of space, the relation between a milieu of interiority and a milieu of exteriority’ (Simondon, 1995, p. 223). The key is that the new biological Transcendental Aesthetic is topological, not Euclidean. We cannot be fooled by the seemingly Euclidean or ‘absolute’ inside-outside in single-celled organisms,10 for ‘the essence of the living being is perhaps a certain topological arrangement that cannot be known on the basis of physics and chemistry, which utilize in general a Euclidean space’ (ibid.). While it’s the case that there is an ‘absolute’ inside-outside of the single-celled organism, it’s not a Euclidean spatiality, but the dynamic and topological maintenance of metastability that counts: For this organism, the characteristic polarity of life is at the level of the membrane; it’s in this region [à cet endroit] that life exists in an essential manner as an aspect of a dynamic topology which itself maintains the metastability by which it exists. (ibid., p. 224)

So we see how it’s the dynamic topological process of individuation that constitutes biological space-time. The interior is the accumulated past, the exterior the forthcoming future. Concerning the relation of interiority and the past, Simondon writes: ‘the entire mass of living matter which is in the interior space is actively present to the exterior world at the limit of the living being: all the products of the past of the individuation [de l’individuation passée] are present without distance and without delay’ (ibid., p. 225). While interiority constitutes the past, exteriority constitutes the future: ‘The fact that a substance is in the milieu of exteriority means that that substance can come forth [peut advenir], be proposed for assimilation, [or] wound [léser] the living individual: the substance is to come, is futural [est à venir]’ (ibid.) Deleuze will call the pre-individual the realm of the virtual, and the individuation process the realm of the intensive. Using another of Deleuze’s idiosyncratic terms, staying in touch with the metastable fields surrounding your ongoing individuation (which can be psychic and social as much as organic) is keeping your ‘Body without Organs’ close by. Attaining your ‘BwO’ is not regression to a prior pre-individuation, but attaining your potential for transformation. [10] ‘The simplest organism, which we can call “elementary,” is that which does not possess a medial interior milieu, but only an absolute interior and exterior’ (Simondon, 1995, p. 225). [9]

104

J. PROTEVI

The full contours of the new, biological Transcendental Aesthetic come into focus as past and future combine in a living present constituted by the membrane:

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

At the level of the polarized membrane, the interior past and the exterior future face one another [s’affrontent]: this face off [affrontement] in the operation of selective assimilation is the present of the living being [le présent du vivant], which is made up of this polarity of passage and refusal, between substances which pass into the past [substances passées] and substances which come forth futurally [adviennent], [substances which are] present [présentes used here as an adjective] one to the other by means of [à travers] the operation of individuation… (ibid., p. 226)

However, we must never reify the membrane: it is the process of individuation maintaining a dynamic topology that constitutes the new Transcendental Aesthetic of living present as relation of interior and exterior, past and future: ‘the present is that metastability of the relation between interior and exterior, past and future; it’s in relation to this allagmatic11 activity of mutual presence that the exterior is exterior and the interior is interior’ (ibid.) To conclude, we can note that Simondon is quite clear that the new, biological Transcendental Aesthetic he articulates in his philosophy of transductive individuation is a departure from Kant: ‘Topology and chronology are not a priori forms of sensibility, but the very dimensionality of the living being as it individuates itself [la dimensionnalité du vivant s’individuant]’ (ibid.)

Life and Creativity The following section is something of a departure from a strict focus on space and time, but as the topic is so important, let us consider Simondon’s definition of life, which is quite close to that of autopoiesis. Life is self-maintenance [auto-entretien] of a metastability, but a metastability that requires a topological condition: structure and function are linked, for the most primitive and profound vital structure is topological. (Simondon, 1995, p. 224)

Simondon’s definition is quite close to the definition of life in autopoiesis, but there are some notable differences. The similarity comes [11] From the excellent resource on Simondon’s vocabulary (http://fractalontology.

wordpress.com/2007/11/28/a-short-list-of-gilbert-simondons-vocabulary/), we find this definition: ‘Allagmatic — The Greek word allagma can mean change or vicissitude, but it can also mean that which can be given or taken in exchange, which more genuinely captures the idea of energy exchange in Simondon’s usage.’

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE, MIND IN PROCESS

105

from the notion of self-maintenance of a topological dynamics in which structure and function are linked. But the ‘metastability’ thematized by Simondon is an interesting twist. The binary logic of autopoiesis — conservation or dissolution — had to be supplemented by the dynamic notion of ‘adaptivity’ developed by Ezequiel Di Paolo (2005), and explicated by Thompson in Mind in Life. The reason for this supplement is that autopoiesis is only sufficient for the maintenance of the organism’s self-identity. To account for sense-making, Thompson turns to Di Paolo’s notion of adaptivity: ‘A distinct capacity for “adaptivity” needs to be added to the minimal autopoietic organization so that the system can actively regulate itself with respect to its conditions of viability and thereby modify its milieu according to the internal norms of its activity’ (Thompson, 2007, p. 148). But what about Simondon’s ‘metastability’? Can this term, discussed in terms of virtuality by Deleuze, be covered by ‘adaptivity’? It would take more time than we’re able to devote to it here, but we can pose a few points for further development (Protevi, forthcoming). The key for us is to see that adaptivity requires a dynamic emergent self unifying a multiplicity of serial processes. We might say that autopoiesis entails synchronic emergence, whereas adaptivity entails diachronic emergence. Notice the dynamic monitoring of multiple processes Di Paolo isolates here as necessary for generating singular norms of each organism: Only if they are able to monitor and regulate their internal processes so that they can generate the necessary responses anticipating internal tendencies will they also be able to appreciate graded differences between otherwise equally viable states. Bacteria possessing this capability will be able to generate a normativity within their current set of viability conditions and for themselves. They will be capable of appreciating not just sugar as nutritive, but the direction where the concentration grows as useful, and swimming in that direction as the right thing to do in some circumstances. (Di Paolo, 2005, p. 437, emphasis in original)

The comparison of enactive biology and Deleuze is complicated, however, by Deleuze’s notion of intensive individuation processes. Deleuze is a process philosopher, one focused on creativity and novelty (Shaviro, 2009). We can truly say that autopoiesis is not a substance concept, at least in so far as substance is seen as reified, for what is conserved in autopoiesis is not the organism as stable thing, but the organism as self-maintaining membrane-metabolism recursive process. But what of the notion of creativity in life on which Deleuze focuses? Does the autopoietic organism help us think of life’s creativity? For Deleuze and Guattari, the answer is no; the organism is

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

106

J. PROTEVI

actually only the resting point between bursts of creativity: ‘the organism is that which life turns against itself to limit itself’ (Deleuze and Guattari, 1987, p. 503). So ‘life’ for them is most fully displayed in evolutionary creativity, even if in their more sober moments Deleuze and Guattari admit, perhaps even grudgingly, that the organism or autopoietic conservation is the condition for another round of biological creativity: ‘Dismantling the organism has never meant killing yourself… Staying stratified… is not the worst that can happen’ (ibid., pp. 160–1). After all, dead men tell no tales and dead organisms produce no creative variations. What are we to make of all this talk of creativity? Is this just another barbarous or mystical hypothesis? Far from being a vitalist fantasy, Deleuze’s emphasis on ontogenetic and evolutionary creativity echoes the notion of ‘developmental plasticity’ developed by Mary Jane West-Eberhard (2003; see also Pigliucci, 2010). Although I can’t show it in detail here, I would claim Deleuze is a multi-time scale thinker, an eco-evo-devo thinker: along with ‘involution’ (what Lynn Margulis calls ‘symbiogenesis’), 12 he would agree with WestEberhard that creativity in developmental plasticity provides the source of the variation with which evolution by natural selection works other than the canonical source, random mutation.13 But autopoiesis and adaptivity seem limited to the behavioural time scale. Even granted that autopoiesis is a (self-focused) process term, we might say that the notion of autonomous system overemphasizes stability, while a Deleuzean-Simondonian transductive individuation, even if it doesn’t emphasize creativity per se, at least provides the conditions for it. From this perspective, the embryo as paradigmatic ‘larval subject’ is merely a more intense site of individuation than the adult; however sclerotic and habitual, the adult is only the limit of the process of individuation. There’s always the chance for change, for [12] Margulis’s notion of symbiogenesis (Margulis and Sagan, 1995), is echoed in Deleuze

and Guattari (1987, p. 238). See Ansell-Pearson (1999, pp. 165–6) for a brief discussion.

[13] West-Eberhard (2003) does not deny natural selection, but claims it will favour the spread

of a particular environmentally-induced phenotypic variant when it has positive effects on individual fitness, that is, when it is adaptive. West-Eberhard denies this is Lamarckism, because there is no direct influence of environment on genotype. In other words, Lamarck thought that adaptive phenotypic changes were the source of variants that could be inherited (in contemporary terms, adaptive phenotypic changes produce genetic variation). But that’s not West-Eberhard’s scheme. What she says is that some adaptive phenotypic change is the result of developmental plasticity calling upon previously hidden, i.e. unexpressed, genetic variation. In other words, neither the phenotype nor the environment produces genetic variation. The above sketch needs to be made more precise through an analysis of West-Eberhard’s notion of ‘genetic accommodation’, but this is not the right venue for such a reading.

MIND IN LIFE, MIND IN PROCESS

107

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

development of new patterns. Of course they have to fit within limits of viability, as autopoiesis insists, but it’s a matter of emphasis: autopoiesis with its emphasis on conservation, and adaptivity with its emphasis on homeostasis versus Deleuze’s emphasis on creativity, for which Simondon’s notion of transductive maintenance of metastability serves as its condition. In terms of autopoietic synchronic emergence, then, we might say that enaction relegates the metastable field to coupled environment and limits transduction to metabolism, while in terms of adaptivity’s diachronic emergence, it neglects ontogenesis in favour of adult function and restricts transduction to homeostatic regulation.

Deleuze and ‘Spatio-Temporal Dynamisms’ To this point we have discussed the new, biological Transcendental Aesthetic. But Simondon’s notion of individuation extends below the organic level; transductive individuation is pre-biotic as well as biotic. There are important dynamic topological differences between crystallization and organic individuation, but ‘there might be an intermediary order of phenomena, between parcellary microphysics and the macrophysical unity of the organism; this order would be that of genetic processes, chronological and topological, that is to say, the processes of individuation, common to all orders of reality in which an ontogenesis operates’ (Simondon, 1995, p. 227). Let’s spend a minute on the fascinating difference between crystals and organisms as Simondon articulates it: vital individuation does not come after physico-chemical individuation, but during this individuation, before its completion, by suspending it at the moment when it has not attained its stable equilibrium… The living individual would be in some manner, at its most primitive levels, a crystal in the state of being born [à l’état naissant], amplifying itself without stabilizing itself. (ibid., p. 150, italics in original)

Simondon appeals to neoteny (slowing down) to explain this idea. So within the organic realm we also see individuation as the suspension of metastable processes. In a startling image, the animal is the ‘inchoate plant’: developing and organizing itself by conserving the motile, receptive, and reactive possibilities which appear in the reproduction of vegetative life [la reproduction des végétaux]… animal individuation feeds on [s’alimente] the most primitive phase of vegetative individuation, retaining in itself something anterior to the development of the adult

108

J. PROTEVI

plant [végétal adulte], and in particular maintaining, during a longer time, the capacity of receiving information. (ibid.)

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

These pre-biotic ‘genetic processes’, operating by means of a deferral of stability or maintenance of metastability, are what Deleuze calls ‘spatio-temporal dynamisms’. In his terms, they are intensive processes rather than virtual structures or actual products. Let’s turn to Deleuze’s essay ‘The Method of Dramatization’ (Deleuze, 2004) which has a somewhat more clear presentation than Difference and Repetition. Spatio-temporal dynamisms ‘create particular spaces and times’, in a non- or pre-biotic Transcendental Aesthetic. Beneath organization and specification [the actual], we discover nothing more than spatio-temporal dynamisms: that is to say, agitations of space, holes of time, pure syntheses of space, direction, and rhythms. (ibid., p. 96)

Spatio-temporal dynamisms, as intensive processes of impersonal individuation with their own space-time, entail a second new Transcendental Aesthetic, this time non- or pre-biotic. Although individuation is a general case, covering the pre-biotic, Deleuze finds biology a better model than Simondon’s crystallization. But biology is only a model for Deleuze’s notion of intensive processes which actualize a virtual Idea (Deleuze, 1994, p. 220–1). So, when he unleashes one of his most infamous gnomic utterances, ‘the whole world is an egg’ (Deleuze, 2004, p. 96; see also Deleuze, 1994, p. 251), we cannot restrict the extension of spatio-temporal dynamisms to the biological realm. In other words, it’s not the ‘egg’ we should concentrate on, but ‘the whole world’. For transductive individuation in all registers, organic, physical, and social, you need a pre-individual field with virtual potentials that are not individuated, that do not ‘resemble’ the products produced by intensive individuation processes structured by those potentials (Toscano, 2006). Making the connection to the new Transcendental Aesthetic pursued by our thinkers, we see that Deleuze will claim that ‘What I am calling a drama [another term for ‘spatio-temporal dynamism’] particularly resembles the Kantian schema’ (Deleuze, 2004, p. 99; see also Deleuze, 1994, pp. 216–7). Seeing spatio-temporal dynamisms as the analogue to schematisms is linked to the post-Kantian demand for genesis of the Transcendental Aesthetic, that is, space and time as generated rather than posited as conditions: ‘We would have to distinguish what belongs to space and what to time in these dynamisms, and in each case, the particular space-time combination. Whenever an Idea is actualized, there is a space and a time of actualization’

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE, MIND IN PROCESS

109

(Deleuze, 2004, p. 111). To locate the space and time of actualization we must first distinguish three registers; virtual, intensive, actual (DeLanda, 2002; Bonta and Protevi, 2004). The intensive is the space-time of individuation processes, that is, actualization of the virtual (see note 2 above for a brief sketch). In the virtual register, we have virtual space: the meshed continuum of Ideas with zones of indiscernibility between them. And we have virtual time as progressive determination of Ideas, the ‘movement’ from determinable but undetermined differential elements, their reciprocal determination in differential relations, and the ideal of complete determination in the singularities these relations generate.14 Deleuze calls virtual space ‘depth’ or spatium and virtual time ‘Aion’. In the intensive register, we find intensive spatial processes: foldings, cascades, convection currents, etc. With intensive time, we see the time of thresholds and critical points, the time of kairos. Finally, extensive or actual space has universal measurements: millimetres, metres, etc. while extensive/actual time is similarly universally measurable with same units: the time of Chronos or clock/ calendar time. The difference of intensive space-time and extensive space-time is the existence of ‘critical’ points and moments in the former: the moment of process reaching a threshold that produces a qualitatively new pattern is not just any old moment or ‘time T1’ just as the point at which currents bend is not just any old spatial point at specific values of co-ordinates x, y. Rather, critical times and spaces are immanently determined as critical in the intensive process that unfolds with its own concrete space and time; it cannot be compared in a universal framework to some other moment or point. Deleuze’s ‘spatio-temporal dynamisms’ are found in multiple registers: 1) in the physical register, the spatial density gradients and temporal critical points are reciprocally determining in crystallization; 2) in the ontogenetic organic register, cellular displacement and temporality of gene expression networks are linked in embryonic development; 3) and in the evolutionary organic register, the distribution of plastic [14] We can compare this to Donn Welton’s notion of ‘transcendental space’ of constitutive

phenomenology, and ‘transcendental time’ of genetic phenomenology: ‘This gives us yet another interesting way of understanding the difference between constitutive and genetic analysis. We can say that constitutive phenomenology schematizes the structural transformations making phenomenal fields possible according to transcendental space. They are framed as layers or strata beneath each field, providing it with its supporting ground. Genetic phenomenology schematizes those transformations in terms of transcendental time, and thus as a process of development in which the earlier gives rise to the later, and in which the later draws and gives direction to the now’ (Welton, 2003, p. 254, italics in original).

110

J. PROTEVI

developmental systems (multiplicity of concrete space and time of ontogenesis in a population) provides the variation for the temporality of genetic accommodation in Mary Jane West-Eberhard’s work. The contrast then of concrete intensive space-time dynamisms and abstract universal extensive time is the contrast between the new Transcendental Aesthetic of Deleuze and Simondon and that of Kant in which universal space and time are the a priori forms of intuition.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

The Question of Panpsychism When we realize that each spatio-temporal dynamism for Deleuze has a larval subject, we’re forced to tackle the question of panpsychism. Although he uses many biological examples in Difference and Repetition, they are only examples of spatio-temporal dynamisms and larval subjects. As we will see, rocks and islands are spatio-temporal dynamisms too, so they too will have a ‘larval subject’! Deleuze writes: Dynamisms are not absolutely subjectless, though the subjects they sustain are still only rough drafts, not yet qualified or composed, rather patients than agents, only able to endure the pressure of an internal resonance or the amplitude of an inevitable movement. A composed, qualified adult would perish in such an environment. The truth of embryology is that there are movements which the embryo alone can endure: in this instance, the only subject is larval. (Deleuze, 2004, p. 97)

One of the great advantages of the Mind in Life position is that it enables us to escape from the badly-posed Cartesian problem of the relation of the mental and the physical. But then we have a problem with emergence of life and mind: the move from the abiotic to the biotic, from the non-cognitive to the cognitive. And with this move, we come upon the question of panpsychism. Recent work has gone back to the problem of panpsychism (Skrbina, 2005; 2009). We will very briefly touch on two of the contemporary approaches to panpsychism Skbina identifies, information theory of the ‘cybernetic mind’ type and process philosophy (Skrbina, 2005, p. 246). We’ll begin with process philosophy. To Cartesian mechanists, panpsychism is laughable if not maddening, the abject of thought. Panpsychism is inconceivable: extended substance is dead, inert, unconscious, non-sentient, bereft of experience. But Thompson rejects the Cartesian extended substance picture of nature in favour of a radical processualism: In the context of contemporary science… ‘nature’ does not consist of basic particulars, but fields and processes… there is no bottom level of basic particulars with intrinsic properties that upwardly determines

MIND IN LIFE, MIND IN PROCESS

111

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

everything else. Everything is process all the way ‘down’ and all the way ‘up,’ and processes are irreducibly relational — they exist only in patterns, networks, organizations, configurations, or webs… There is no base level of elementary entities to serve as the ultimate ‘emergence base’ on which to ground everything. (Thompson, 2007, pp. 440–1)

In so far as the major process philosopher of the twentieth century, Whitehead, was a panpsychist, we have at least a prima facie invitation to pursue the connection of Thompson’s radical processualism and panpsychism (for a recent piece on Whitehead and panpsychism, see Basile, 2009). With the coextension of mind and life we come to the question by the panpsychists: is ‘Mind in Life’ too restrictive with its definition of mind? If there’s no mind prior to life, then mind must radically emerge with life. But with the process view there’s no radical emergence from a baseline of elementary particles. There’s emergence in the sense of new structures generating new capacities, but the panpsychist would say these new capacities are the development of potentials in the ‘lower level’. When cognitive capacities are at stake, a panpsychist would say that mind gets more complex as we find life, but it doesn’t radically emerge with life. A Cartesian radical emergentist would say there’s dead unmindful matter that when properly arranged becomes living and minded. But is that really less strange than the panpsychist position? In fact, Strawson (2009) will say radical emergence is no better than ‘magic’, so it’s actually a more rigorous position to be a panpsychist. Let us concentrate on Thompson. In Mind in Life Thompson examines several cases at the borderline of autopoietic cognition, starting with Stuart Kauffman’s autocatalytic loops (Thompson, 2007, pp. 104–5). Recall the Mind in Life position that strongly links autopoiesis and cognition: ‘cognition is behavior or conduct in relation to meaning and norms that the system itself enacts or brings forth on the basis of its autonomy’ (ibid., p. 126). On this reading, you need the physical instantiation of metabolism-membrane-metabolism recursivity to have an autonomous subjectivity such that life and organismic sense-making are linked. Thus Thompson will rule out Kauffman’s autocatalytic loops as the basis or minimal example of life: they don’t have a recursive membrane-metabolism structure, so they don’t have autonomy and don’t enact a subjective position (ibid., p. 105). Later he examines other borderline cases, the tessellation automaton of Bourgine and Stewart, and the autocatalytic miscelles and viscelles of Bitbol and Luisi (ibid., p. 125). While Thompson leaves undecided whether these systems are minimal cases of autopoiesis or only protoautopoietic systems, on the strong autopoietic definition of cognition

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

112

J. PROTEVI

given above, they too fail to qualify as autonomous autopoietic systems because they do not produce a physically realized membranemetabolism recursivity and hence an autonomous subject position (ibid., pp. 125–7). But what about a simpler definition of cognition? There is information transfer and self-organization in autocatalytic loops, and this fits the cybernetic definition of mind offered by Gregory Bateson when he identifies ‘mind as synonymous with cybernetic system — the relevant total information-processing, trial-and-error completing unit’ (Bateson, 1972, p. 460).15 Deleuze will have such a cybernetic outlook: in self-organizing systems or ‘spatio-temporal dynamisms’ there’s a ‘dark precursor’, then ‘internal resonance’ and then ‘forced movement’ (Deleuze, 1994, pp. 126, 277–8). The first term, ‘dark precursor’, indicates information transfer between heterogeneous series (DeLanda, 2002, p. 80). Since intensity is difference, differences in intensity must enter into communication. Something like a ‘difference operator’ is required, to relate difference to difference. This role is filled by what is called an obscure precursor [précurseur sombre; Patton translates as ‘dark precursor’ in Deleuze, 1994]. A lightning bolt flashes between different intensities, but it is preceded by an obscure precursor, invisible, imperceptible, which determines in advance the inverted path as in negative relief, because this path is first the agent of communication between series of differences. (Deleuze, 2004, p. 97)

Just like with ‘the whole world is an egg’, Deleuze seems to be unnecessarily gnomic with his term ‘dark precursor’. But in fact it is plain old meteorology. Actually, there are a number of steps here in lightning genesis: 1) the formation of ionized air called ‘plasma’ which is much more highly conductive than normal air; 2) formation of ‘step leaders’ or channels of ionized air which propagate from cloud to ground in stages; 3) ‘positive streamers’ which reach from objects on the ground to cloud; 4) the meeting of positive streamer and step leader, which allows the current to pass. So ‘dark precursor’ could be either step 2 or step 3. In fact, the term précurseur is used straightforwardly in any number of French-language websites on lightning. If we push it, we can see a total panpsychism in Difference and Repetition that surpasses the biological to the level of ‘spatio-temporal dynamisms’ or the self-organizing cybernetic mind level. Deleuze notes that the mathematical and biological notions of differentiation and differenciation employed in Difference and Repetition are only a [15] As Skrbina notes, Bateson later backs away from this cybernetic mind position (Skrbina,

2005, pp. 196–8).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE, MIND IN PROCESS

113

‘technical model’ (Deleuze, 1994, p. 220). Now if ‘the entire world is an egg’ (ibid., p. 216), then every individuation is ‘embryonic’, we might say, even rocks: ‘On the scale of millions of years which constitutes the time of their actualization, the hardest rocks in turn are fluid matters which flow under the weak constraints exercised on their singularities’ (ibid., p. 219). Now if rocks and islands as individuation processes are embryonic, then they too have a psyche: ‘every spatio-temporal dynamism is accompanied by the emergence of an elementary consciousness’ (ibid., p. 220). By the time of AntiOedipus and A Thousand Plateaus, Deleuze and Guattari explicitly thematize that the syntheses they investigate are fully material syntheses, syntheses of nature in geological as well as biological, social, and psychological registers (Welchman, 2009). Not just organic syntheses, but inorganic ones as well, are ‘spatio-temporal dynamisms’. With this full naturalization of syntheses, the question of panpsychism is brought into full relief, since material syntheses are as much syntheses of experience as they are syntheses of things, as we see in the title of Chapter 3 of A Thousand Plateaus: ‘The Geology of Morals: Who does the earth think it is?’ (‘La géologie de la morale: pour qui elle se prend, la terre?’) We thus have a second new Transcendental Aesthetic here with Deleuzean ‘spatio-temporal dynamisms’. It’s the Transcendental Aesthetic of larval subjects, of mind in physical self-organizing processes, echoing Bateson’s cybernetic mind. So the question is: how do we relate this to Mind in Life? Can we have a coherent defensible notion of mind that’s broader than that of sense-making of an autopoietic organism, one based on information transfer and selforganization in physical processes (crystallization, convection currents, lightning, hurricanes, etc.)? Then the question of emergence of mind is pushed down below the emergence of life. How far down? Is there a point of emergence we can locate? That’s what the panpsychists deny. It’s mind all the way down. Thompson will say it’s process all the way down, but doesn’t say whether there’s a nonautopoietic notion of mind that accompanies process. Is there a ‘Mind in Process’ position we need to think about? Thompson’s subtitle is ‘Biology, Phenomenology, and the Sciences of Mind’. Is there a ‘Physics, Phenomenology, and the Sciences of Mind’ book to be written? To move toward a conclusion, let us note that a classic objection to panpsychism is based on a worry about the overuse of the principle of parsimony: we can’t push parsimony too far, because the fewer principles we have, the more we risk stretching them beyond their useful extension. So we have to worry that a definition of mind as mere

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

114

J. PROTEVI

information transfer involved in self-organization is so broad as to be meaningless: if convection currents in a pot of boiling water are mind, what good is such a broad definition? But on the other hand, what’s exciting about dynamic systems modelling is that it shows self-organizing processes in an extremely wide range of registers, from convection currents through neurodynamics. So if self-organization is a univocal concept, that is, if there is a non-trivial shared structure between convection currents and neurodynamics, then we have identified a fundamental principle that links the inorganic and organic registers. So we’re back to the cybernetic challenge: is information transfer and self-organization capable of being called ‘mind’ in a defensible fashion? It wouldn’t be autopoietic cognition, because it doesn’t involve a membrane-metabolism recursive process and hence an autonomous subject position. But wouldn’t it be ‘Mind in Process’, even if it’s not ‘Mind in Life’? To conclude somewhat abruptly: if there is ‘Mind in Process’, that is, mind all the way down just as there is process all the way down, that means we really have our work cut out for us in discussing this second new Transcendental Aesthetic, the non- or pre-biotic one. It’s not that we don’t have enough to talk about with a biological Transcendental Aesthetic, but if we want to follow Deleuze all the way, we’ll have to go not only ‘beyond the turn’ in (human) experience as Bergson puts it (1988, p. 185), but ‘beyond the turn’ of (living) experience out into the ‘plane of consistency’ we find posited in A Thousand Plateaus.

References Ansell-Pearson, K. (1999) Germinal Life: The Difference and Repetition of Deleuze, London: Routledge. Basile, P. (2009) Back to Whitehead? Galen Strawson and the rediscovery of panpsychism, in Skrbina, D. (ed.) Mind that Abides: Panpsychism in the New Millennium, Amsterdam: John Benjamins. Bateson, G. (1972) Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology, Chicago, IL: University of Chicago Press. de Beistegui, M. (2004) Truth and Genesis: Philosophy as Differential Ontology, Bloomington, IN: Indiana University Press. Bergson, H. (1988) Matter and Memory, Paul, N.M. & Palmer, W.S. (trans.), New York: Zone Books. Bonta, M. & Protevi, J. (2004) Deleuze and Geophilosophy: A Guide and Glossary, Edinburgh: Edinburgh University Press. Bryant, L. (2008) Difference and Givenness: Deleuze’s Transcendental Empiricism and the Ontology of Immanence, Evanston, IL: Northwestern University Press. DeLanda, M. (2002) Intensive Science and Virtual Philosophy, London: Continuum.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE, MIND IN PROCESS

115

Deleuze, G. (1994) Difference and Repetition, Patton, P. (trans.), New York: Columbia University Press. Deleuze, G. (2004) The method of dramatization, in Lapoujade, D. (ed.) Desert Islands and Other Texts, Taormina, M. (trans.), New York: Semiotext(e). Deleuze, G. & Guattari, F. (1987) A Thousand Plateaus, Massumi, B. (trans.), Minneapolis, MN: University of Minnesota Press. Di Paolo, E. (2005) Autopoiesis, adaptivity, teleology, agency, Phenomenology and the Cognitive Sciences, 4, pp. 429–452. Hughes, J. (2009) Deleuze’s Difference and Repetition, Edinburgh: Edinburgh University Press. Jonas, H. (2003) The Phenomenon of Life: Toward a Philosophical Biology, Evanston, IL: Northwestern University Press. Jones, G. (2009) Solomon Maimon, in Jones, G. & Roffe, J. (eds.) Deleuze’s Philosophical Lineage, Edinburgh: Edinburgh University Press. Kelso, J.S. (1995) Dynamic Patterns: The Self-Organization of Brain and Behavior, Cambridge, MA: MIT Press. Maimon, S. (2010) Essay on Transcendental Philosophy, Welchman, A., Somers-Hall, H., Reglitz, M. & Midgley, N. (trans.), London: Continuum. Margulis, L. & Sagan, D. (1995) What is Life?, New York: Simon and Schuster. Pigliucci, M. (2010) Phenotypic plasticity, in Pigliucci, M. & Müller, G. (eds.) Evolution — The Extended Synthesis, Cambridge, MA: MIT Press. Protevi, J. (2010) Adding Deleuze to the mix, Phenomenology and the Cognitive Sciences, 9 (3), pp. 417–436. Protevi, J. (forthcoming) Larval subjects, autonomous systems, and E. coli chemotaxis, in Guillaume, L. & Hughes, J. (eds.) Deleuze and the Body, Edinburgh: Edinburgh University Press. Shaviro, S. (2009) Without Criteria: Kant, Whitehead, Deleuze and Aesthetics, Cambridge, MA: MIT Press. Simondon, G. (1995) L’individu et sa genèse physico-biologique, Grenoble: Million. Skrbina, D. (2005) Panpsychism in the West, Cambridge, MA: MIT Press. Skrbina, D. (ed.) (2009) Mind that Abides: Panpsychism in the New Millennium, Amsterdam: John Benjamins. Smith, D. (2009) Genesis and difference: Deleuze, Maimon, and the post-Kantian reading of Leibniz, in van Tuinen, S. & McDonnell, N. (eds.) Deleuze and the Fold: A Critical Reader, London: Palgrave Macmillan. Smith, D. & Protevi, J. (2008) Gilles Deleuze, Stanford Encyclopedia of Philosophy, [Online], http://plato.stanford.edu/entries/deleuze/ [23 October 2010]. Steinbock, A. (1995) Home and Beyond: Generative Phenomenology after Husserl, Evanston, IL: Northwestern University Press. Strawson, G. (2009) Realistic monism: Why physicalism entails panpsychism, in Skrbina, D. (ed.) Mind that Abides: Panpsychism in the New Millennium, Amsterdam: John Benjamins. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press. Toscano, A. (2006) The Theatre of Production: Philosophy and Individuation between Kant and Deleuze, London: Palgrave Macmillan. Welchman, A. (2009) Deleuze’s post-critical metaphysics, Symposium, 13 (2), pp. 25–54. Welton, D. (2000) The Other Husserl: The Horizons of Transcendental Phenomenology, Bloomington, IN: Indiana University Press. Welton, D. (ed.) (2003) The New Husserl: A Critical Reader, Bloomington, IN: Indiana University Press.

116

J. PROTEVI

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

West-Eberhard, M.J. (2003) Developmental Plasticity and Evolution, New York: Oxford University Press. Williams, J. (2003) Gilles Deleuze’s Difference and Repetition: A Critical Introduction and Guide, Edinburgh: Edinburgh University Press. Zahavi, D. (2005) Subjectivity and Selfhood: Investigating the First-Person Perspective, Cambridge, MA: MIT Press.

Charles Siewert

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Embodied Consciousness and the Explanatory Gap 1. The Problem of the Explanatory Gap Evan Thompson’s Mind in Life brings together a diverse array of themes and ideas drawn from classical phenomenology, theoretical biology, and dynamical systems theory. It is remarkable how Thompson synthesizes these into a coherent whole. I will not (and I can’t) try to survey this whole, or match the breadth of his discussion in my comments. Instead I will focus on some pivotal aspects of his manner of meeting the challenge that frames the discussion — the challenge of the so-called ‘explanatory gap’ of consciousness and nature. Thompson does not purport to have solved the problem of the explanatory gap; there is no grand claim that consciousness has been explained. But he aims to make a substantial start on a new approach, to ‘provide new resources for addressing the explanatory gap’, as he says in his Précis. I am sympathetic to the thrust of what I take him to be saying, and I want to make more explicit what is involved in a view of the type Thompson wants to advance (or at least, to nudge him into being more explicit about this). For I want to see better how a view of this kind might be elaborated and defended. What I am going to say about the problem of the gap clearly owes much to Joseph Levine’s and David Chalmers’ treatments of these issues — though, naturally, my exposition will be much rougher than theirs. I only hope it will be clear enough for my main purpose here: to help us to appreciate and begin to critically consider the kind of approach Thompson is suggesting. First, let me give some orientation to what I will say by indicating briefly some of my assumptions. We are capable of ‘first-person Correspondence: Email: [email protected]

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 117–38

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

118

C. SIEWERT

reflection on experience’ — that is, we have a way of forming thoughts in the first-person about our own experience that differs significantly in kind from that we have of forming thoughts about others’ experience. Experience here is taken to include, at a minimum, certain kinds of sensory states, such as appearances of colour, tastes, smells, and bodily feelings — but much richer conceptions of the range of experience, including conceptual thought and higher emotion, are by no means excluded. To have experience is to have conscious states — conscious in the ‘phenomenal’ sense. We may call first-person reflection on experience ‘introspection’ — as long as our use of that term leaves open the question of what exactly is distinctive about this ‘first-person perspective’. And an approach to enquiry into the general character of experience that relies in a fundamental way on such first-person reflection, I will call — and I don’t think this is at odds with Thompson’s usage — ‘phenomenological’. Now let me sketch a way of thinking about the problem of explaining consciousness which, as Thompson remarks (pp. 222–3), has become influential. First-person reflection on our own experience involves the use of certain concepts (which we may call ‘phenomenal’ concepts) — concepts we have, for instance, of ways things look or feel to us, or more generally, of ‘ways of seeming’. And these concepts we have of our own experience — of forms of ‘phenomenal consciousness’ — are distinct from the concepts we deploy in theories of our brain activities and the functional or causal roles these activities play. This is evident in that there seems to be no entailment leading us from application of the latter (physical/functional) concepts to application of the former (‘phenomenal’) concepts. One way of trying to make this failure of entailment particularly vivid is to invite us to conceive of a world type-identical to our own at the level of fundamental physics, but from which the forms of experience we understand first-personally (ways of looking, feeling, tasting) have gone missing. Paraphrasing Chalmers: in such a conceivable world the same types of fundamental physical particles that are found in our world are distributed in fields of force in the same manner, conforming to the same physical laws as our own. Nonetheless, these particles are imagined not to make up feeling beings such as we ourselves are (and such as we normally suppose our monkeys, dogs and kangaroos to be). We are to suppose there is nothing it’s like to be our analogues in this imagined world, nothing it is like for them, any more than we usually suppose there actually is something it’s like to be a coffee machine or a pile of hair. This is the idea of a world type identical to our own at the microphysical level, but totally bereft of feeling,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

119

and of phenomenal experience generally — a world full of movement, but dead to consciousness: a ‘zombie world’. Use of the word ‘zombie’ in this context can invite an easy derision. But we shouldn’t allow the underlying point to be dismissed. That our first-person understanding of experience employs concepts that permit such reflections points to a distinctive challenge facing efforts to construct a scientific theory of consciousness. For this may lead us to think that our usual strategy for the theoretical explanation of nature cannot be applied here. Usually (so the story goes), we satisfy our scientific ambitions by conceiving of what is to be accounted for in terms of what it does (the role it plays, its function), and then seeking out and finding physical mechanisms that do that job, play that role, fulfil that function. Thus it is with boiling and freezing, raining and burning, respiration, digestion and reproduction. Once we see how — given the character of the underlying physical structures and their activity — the function in terms of which the explanandum is identified can’t help but get carried out, we will have explained what we wanted to explain. The problem is, when it comes to consciousness, our understanding of what is to be explained seems to accord it no comfortable place in such a model. That is what the fantasy of a zombie world is supposed to make vivid: you have the structures and you have whatever functions can’t help but get carried out when such structures are operative — but must you then have consciousness? There is no evident necessity that you must, since our subjective understanding of consciousness seems to make a consciousness-deprived physical analogue of our world conceivable. Thus it seems we will inevitably be drawn up short in our efforts at explanation, at least as long as we hope to make our explanation of consciousness follow the same general pattern we find elsewhere in natural science. For we can neither see how justifiably to identify types of conscious experience with forms of activity peculiar to specific physical structures, nor can we honestly see how it must arise, as long as those structures are fulfilling certain functions in a larger picture. The challenge of responding to considerations such as these is, I take it, the challenge of the explanatory gap. Certain general forms of response to the challenge, so understood, can be distinguished. Before considering Thompson’s, let’s review three main options with which it can, I think, be usefully contrasted. Option (1). We have no concept we are warranted in applying to our own experience that would allow us to conceive of a ‘zombie’ world in contrast with our own. For any concept we have a

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

120

C. SIEWERT

right to apply to our own experience really marks nothing but a purely functional difference — and there would be no such difference between our world and a microphysical twin world. Since there is ultimately no conceptual difference of the sort alleged, neither is there an explanatory gap — and, once the confusions and illusions are cleared away, there are no obstacles to explaining consciousness in pretty much the same way science successfully explains other things. (In rather different ways, this is the sort of response found in Daniel Dennett, 1991, and implicit in conceptual functionalists like David Armstrong, 1968.) Option (2). One can (by contrast) accept there is indeed a disparity in our (legitimately applied) concepts (perhaps made vivid by various thought experiments), but deny that it inevitably brings with it an explanatory gap of the form alleged. It would be a mistake to think that since a legitimate first-person understanding of our own experience does not conceive of it in terms of ‘structure’ or ‘function’, there is something more than structure and function in reality for us to conceive of. (Rather different forms of this response can be seen in Michael Tye, 2000, and Ned Block and Robert Stalnaker, 1999.) Option (3). One can accept that consciousness should be seen differently than other targets of scientific explanation to which some had sought to assimilate it. To do this we needn’t abandon a broadly scientific, ‘naturalistic’ view of the world. We just have to acknowledge that the psychophysical principles governing consciousness, while open to scientific discovery, are at bottom contingent laws relating the phenomenal and the physical, and their truth does not admit of deeper explanation by derivation from yet more fundamental principles. (A form of this view has been defended by Chalmers, 1996.) This vision of the challenge of explaining consciousness (and of our options in response to it) can seem compelling, even inescapable. Part of what is interesting about Thompson’s book is that, while he accepts there is a deep and important scientific challenge in accounting for consciousness in its various forms, and he wants to make a start on meeting it, he rejects the now fairly familiar types of responses to it that I have just sketched — and wants to scout out another. The sort of view he is putting forward is not entirely unprecedented, since it is avowedly inspired in part by the philosophies of Husserl, MerleauPonty and others. But it seems to represent a distinctive option missing from the usual maps of this territory.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

121

In my opinion the effort to articulate some such alternative deserves careful and sympathetic attention. For the other options seem to me to varying degrees problematic. First, as I have spelled out elsewhere (Siewert, 1998; 2010), there are reasons I find convincing for rejecting option (1). We can — even without contemplating ‘zombie worlds’ — see that our understanding of our own experience, and of how things appear to us, is not to be construed purely in terms of functional differences that are ultimately characterizable in ways that eliminate all talk of appearance. Or rather, if we do insist on such construals, we end up eliminating phenomenal consciousness from our conception of reality altogether (and that, I think, would be a false economy, to put it mildly). Second, I can’t see how option (2) ultimately manages to avoid trying to get deep, more than nomological (metaphysical) necessities ‘for free’ so to speak. If there is nothing about our conception of the physical substratum of consciousness whence experience and its phenomenal character can be derived, and various functional notions fail to capture such contrasts as our phenomenal concepts allow, it’s hard for me to see what entitles us to the relevant claims of strong necessitation here, or how justifiably to determine exactly what’s minimally necessarily sufficient for it in the required, strong sense. So, in the absence of palatable alternatives I would incline towards option (3). But (as I will try to bring out momentarily) I worry that something has gone wrong in this strict and exhaustive dualist division between a purely phenomenal realm (answering to our phenomenal concepts), and a purely physical realm (which is logically determined by the sorts of microstructure described in physics). What I fear this leaves out is what Thompson in his Précis (following Merleau-Ponty) calls ‘comportment’, or what I will call here ‘embodied consciousness’. I will devote the rest of this essay to exploring how recognition of this category might give Thompson a distinctive take on the explanatory gap.

2. Thompson’s Response to the Gap: Embodied Consciousness There may be aspects of Thompson’s approach that are not yet sufficiently clear to me. However, I think it will be helpful to start by seeing him as advancing the following three ideas. (a) Part of what we need to do to understand how consciousness arises is to show why the idea of a ‘zombie world’ that is functionally just

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

122

C. SIEWERT

like our own is a misconceived legacy of the Cartesian dissociation of mind and life (pp. 222–34). (b) However, we do this not by showing that a proper understanding of experience consists in purely functional notions, which will then leave us free to apply some general structure-and-function model for explaining consciousness (see pp. 224–5). Nor do we diminish the role of phenomenology. Rather, we show that a proper phenomenological understanding of perceptual experience finds it to be so bound to embodied activity that, in a nontrivial sense, the same perceptual functions that occur in our world would not occur in a world in which there were no experience. Even if it’s not correct to say that phenomenal differences are ‘nothing but’ functional differences, we can see that there is, after all, in a sense, no possible truly functionally equivalent world without consciousness. For these types of embodied activity must be missing, if consciousness is missing. This is the upshot of Thompson’s ‘phenomenological critique of zombies’, which draws on Husserl’s (phenomenologically-based) view that ‘…it is through one’s movement and bodily self-experience in movement that an object presents itself as a unified series of appearances’ (p. 232). Thompson argues that if this is right, ‘…then bodily experience is constitutive of the perceptual function of individuating continuous objects in space through a manifold of sensory appearances. So any being that was capable of the same perceptual function would need to have experience of its own body and hence could not be a zombie’ (p. 233). (c) Once we have banished the specter of a functionally equivalent zombie world, we have opened a way around the mind–body impasse. For we no longer have to regard a personal level, phenomenological understanding of experience as radically discontinuous with the forms of understanding proper to the natural sciences — particularly the life sciences. For we can now see that biological systems exhibit a kind of self-organizing, normatively intelligible activity that, at some level of sensorimotor engagement at least, deserves to be considered a form of (and literally to embody) full-blown phenomenal consciousness. (This is a view for which Thompson lays the groundwork mainly in Chapter 6.) For ease of exposition, I need a label for the view characterized by (a)–(c). But while I do think this view can be attributed to Thompson, I hesitate to employ his own doctrinal labels here (such as ‘the enactive approach’ or ‘embodied dynamicism’) since these include

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

123

additional aspects of his position which, while important, I need to leave aside to make my discussion manageable. Let me then call the position roughed out in (a)–(c) the ‘Embodied Consciousness’ View. There are of course many questions we might raise about this view. But let’s start by noting how it diverges from perspectives on the explanatory gap earlier mentioned. One might at first mistake Embodied Consciousness for a version of option (1). For in a sense both would hold that a proper understanding of the concepts we rightfully apply to our own experience (even in ‘introspection’) will not leave room for us to conceive of a world truly functionally equivalent to our own, from which all experience has been expunged. However (as I read Thompson) there is an important difference. On his view, while we cannot properly conceive of an experience-free world where everything ‘functions’ just the same, this is not because (as friends of (1) would maintain) our only legitimate concept of experiences would make them nothing but functional role fillers of some sort. Rather it is because our forms of bodily engagement with the environment cannot be properly understood but as ways of experiencing oneself and things in one’s environment. Neither our ways of experiencing, nor our forms of bodily interaction are rightly understood — even phenomenologically — independently of one another. Thus any world that had in it the very same types of bodily engagement as we have (and in that sense was ‘functionally’ the same) could not help but be a world in which there was consciousness. So it couldn’t be a zombie world. Now we can see that this view also differs from responses (2) and (3) to the alleged gap problem. Like (2), Embodied Consciousness maintains there is no possible world functionally like ours but missing consciousness, even while allowing (unlike (1)) that we can legitimately understand what sorts of experiences we have in first-person reflection in ways that cannot be exhaustively ‘functionalized’. However, the Thompsonian view gives such a phenomenological understanding a more robust role in conceiving of the nature of experience and its necessary connection to bodily activity. According to (2), we cannot rule out, on conceptual grounds, a functional twin world without consciousness — the point is just that this fact lacks ontological import. By contrast, Embodied Consciousness, like (3), takes a phenomenological conception of experience more seriously as a guide to what experience is. But, unlike (3), it finds this perspective on experience does not support, but ultimately casts doubt on the possibility of a zombie world — at least on a certain construal. That last qualification — ‘on a certain construal’ — needs a little more attention. There is a sense in which the kind of connection

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

124

C. SIEWERT

Thompson would make between a phenomenological understanding of perceptual consciousness and embodied activity apparently would not preclude the conceivability (or even the possibility) of a zombie world. If a zombie world is simply taken to be a world the same as this (our consciousness-endowed) world with respect to the types of fundamental particles and their distribution — but just minus all consciousness — I don’t yet see how Thompson’s view would make it any less conceivable. All it would render inconceivable is the idea that vibrating swirls of particles in such a world would constitute the very same types of embodied activity as are to be found in our world. There would be in this experience-deprived world no ‘zombie twins’ doing just what we are doing. While a zombie world could have in it the same types of mere motion as are found in our world (in virtue of having the same distribution of particles in fields of force), it would not — absent consciousness — contain (to speak with Merleau-Ponty) the same (or any) forms of ‘comportment’. Nothing in the zombie world would have our being-in-the-world. Though entities located in that world would move around in space, none would have a world at all. Now that we start to situate a view like Thompson’s vis-à-vis options (1)–(3), we can start to see what difficulties need addressing. At this point friends of (3) may begin to suspect that no real alternative is being offered to their view after all. It is true that we could stipulatively define types of bodily activity in such a way that tied these, with conceptual necessity, to phenomenal experience. But that idle exercise ultimately does nothing to remove the brute contingency of the psychophysical relationship. Consider: we might, for example, define a category of ‘painfultoestubbing’ so that it involved both a kind of movement (for example, jamming one’s toe into the bed post) and a kind of (painful) feeling — a form of phenomenal consciousness. And we may then say that painfultoestubbing is a form of behaviour not to be found in the zombie world. But, we may then note: that is just for the trivial reason that feelings of pain would not be found there — the other part of this contrived category of ‘painfultoestubbing’ (the collision of toe with bedpost) would be there in the zombie world. Thus, ultimately, we are still left with a brutely contingent causal relationship between movement and feeling. The lesson is, until we have explained why these forms of bodily engagement with the world mentioned in (b) above (comportment or what have you) cannot be factored into a ‘phenomenal, experiential bit’ and a ‘bodily movement bit’ with no more than a contingent causal link between them, then the Embodied Consciousness view has not successfully articulated a real alternative to option (3). Once more we are ultimately left with a

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

125

picture of some (perhaps interestingly systematic, but entirely contingent) linkage between (on the one hand) phenomenal features and (on the other) features logically fixed by the distribution of physical particles — an ontological picture which, seen from a certain distance, is not so very different from Descartes’. Now if this constitutes a genuine challenge to Thompson’s view, I cannot yet find a clear answer to it in his ‘phenomenological critique of zombies’. Consider Thompson’s gloss on Husserl’s account of the ‘perceptual function’ that would be missing in a zombie world: that of ‘individuating continuous objects in space through a manifold of sensory appearances’ through ‘movement and bodily self-experience’. It seems that we are merely defining the function in a way that packs in phenomenal consciousness (the manifold of sensory appearance) — and so we risk trivializing the claim that the function could not be found in zombiedom. And while it may be that both movement and bodily self-experience play a role in executing this function, it is unclear why the linkages between experience and movement in this account need be regarded as anything but causal relations among separate existences. Why not suppose zombies to make the very same types of movements, just without their being hooked up to manifolds of sensory appearance and bodily self-experience? The challenge that emerges is this. In order to show why his view really is an alternative to (3), and why it is preferable, Thompson needs to show why, from a phenomenological viewpoint, we distort the character of fundamental cognitive activities (such as object constancy in perception) if we analyse them into quite separate ‘mental’ (or ‘phenomenal’) features and mere bodily movement types. Maybe Thompson’s position is that the burden of proof lies on anyone who would claim they can be thus analysed. But given how entrenched the quasi-Cartesian ‘factoring’ view is in the theoretical imagination, it seems we should try to do more. I want now to suggest a way of meeting this need which, while it departs from direct engagement with the Husserlian (and Merleau-Pontyan) ideas on which Thompson draws, nonetheless concerns the same crucial phenomena, having to do with how the experienced (‘lived’) body figures in our perception of constant objects through the ‘flux of appearance’. But I will try here to develop some basic ideas in this area without reference to their writings or terminology, and using (fairly) plain English. Of course, readers are free to try to map what I say onto Husserlian talk of ‘horizons’, ‘the indeterminately given’, ‘der Leib’, and ‘anticipation’ as they will. But what I say is intended to stand on its own. I will leave

126

C. SIEWERT

it to Thompson to explain whether he thinks my suggestions are helpful, mistaken, or just unnecessary.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

3. A Defence of Embodied Consciousness Let us consider two concepts, pervasively applicable in our lives, in which experience and movement are intertwined, but not in some apparently readily decomposable way (as in the artificial ‘painfultoestubbing’ example). What I have in mind are concepts of looking and touching. We may readily enough understand talk of our looking at something or looking for something, so that this entails that something then looks somehow to us (we may also then say: something visually appears somehow to us, we have visual experience of some kind, some kind of visual, phenomenal consciousness). And while it’s true that we may say that x is touching y, even when x feels nothing (e.g. ‘the fork is touching the plate’), nonetheless, where the ‘touching’ is understood as asymmetrical (‘x touches y’ does not entail ‘y touches x’) — as in ‘I touched the lamp’ — the subject of the touching feels something somehow. At least, we may easily interpret ‘touch’ in that way. And looking and touching, so understood, are activities we engage in nearly constantly while awake, and through which we experience space. But while these are thought of not only as involving experience, but also bodily movement (e.g. of eyes, head, limb and hand) — it is not so clear the understanding I have of looking and touching from the first-person point of view, hence a phenomenological understanding, can be readily segregated into a purely experiential part, and a mere movement part, seen as only contingently, perhaps causally related. In fact, on examination, it seems that my understanding of changes in the character of the appearances I enjoy through looking and touching does not detach these from the activities involved. And likewise, my understanding of the types of movements I take to be involved in looking and touching does not tease them apart from the experiences they afford. Let me elaborate on this idea, concentrating on the visual case. Again, I am looking at something or other most of the time I am awake, and what I am looking at and how much I am looking at it undergoes constant transformation. And in thinking of myself as looking at things, I think of myself as moving somehow — my eyes, my head, my entire body. Further, I can think of the variations in how things look to me by thinking of them as those variations involved in differently looking at things. The question is whether I can, in

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

127

first-person refection, factor out variations in the experiential part of ‘looking at’ (the ‘how things look’ part) and understand it quite separately from a movement part also essential to ‘looking at’. Can I completely capture in first-person thought how appearances differ for me depending on whether and in what way I am looking at things, but without thinking of this at all in those terms? The challenge here is to think of how the appearance of the area before me differs as I look at it differently, but without thinking of myself as looking at things, and thus as moving somehow. Let’s consider how this might apply to an example. If I am looking at something — the plate on the table at which I sit — it looks somehow to me. But also, then, typically: the area around what I am looking at (e.g. where the fork and spoon, and the table surface are), and areas within what I am looking at (e.g. that of the designs around the edge of the plate), also look somehow to me — even while I am not looking at anything in those areas. And as what I am looking at changes — as it does nearly constantly — the way the area before me looks changes. As my gaze shifts from the plate to the fork then to some chipped bit of the plate’s rim on the opposite side, appearances are changing. These changes are evident to first-person reflection. Now can I conceive of the area’s appearance altering in just the way it does as I look at it, but without thinking of myself as looking at it, or at anything in that area? What I need here is some finely discriminating grasp of how things look to me, quite detachable from my understanding of myself as a motor agent. I need, it seems, some way of thinking of looking at as two quite separable things: a movement, and a series of representations that this movement causes (which series then in turn may perhaps be used to control the movement), where the manner of visual appearance corresponds to the representations. Perhaps I can do this by finding some verbally expressible content in vision, which content possession I can understand independently of how it happens to be generated by motor activity. Or perhaps how things look to me can be made out as some visual ‘picture’ or ‘image’ of the scene, which I can think of as the product of my movements, but whose character I can adequately conceptualize in isolation from them. Perhaps. Let’s examine this more closely. What grasp do I really have conceptually of these changes in appearance? First, let me try to characterize them by attributions of distinct verbally expressible content to my experience over time, using common nouns, adjectives, demonstratives of ordinary language (it looks to me as if there is a fork; that looks to me green; that (design) looks to me this shape). But

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

128

C. SIEWERT

this does not exhaust the alterations in the character of my experience of which I am aware as I look around in ordinary circumstances. In that still apparent region where the fork lies waiting for my gaze to light upon it, it either does not yet appear to me ‘as a fork’ until I get a better look (and then, when I do, what is there appears to me differently than it did), or else if I have got a good enough look at it for it to appear to me as a fork, there are still many subjectively discernibly different ways a fork can appear to me ‘as a fork’ depending on how I am looking at it. The point is that the variations in appearance either resist such verbalized content attributions, or exceed what is thereby differentiated. I would extend the point to whatever general terms one might find suitable for reporting how things look. For whatever lies in a visually apparent region, and whatever general ‘F’ one cares to consider, either it is not (yet) correct to report any manner of appearance I enjoy by saying ‘…looks to me… F’ (for I haven’t looked at something well enough), or if it is correct, I will be able to find discernibly different appearances that may both be so reported, because of differences in how much or in what way I am looking at something. Here it may be suggested that I can always capture this ‘fine grained’ alteration in visual appearance by beefing up my linguistic resources with resort to demonstrative predicates (it looks to me ‘that shade’, ‘that shape’). So, every change in visual appearance — as I look from the plate’s edge to the fork, from its tines to its handle, to the fork as a whole, to the interior of the plate — can be marked: ‘that looks this shape to me, that looks this shape to me, that looks this shape to me’ where the different ‘thats’ and ‘thises’ pick out distinct localizations of distinct shapes apparent to me, and this tracks in a conceptually independent way exactly all the ongoing variations in appearance of which I am aware in reflection. But now: can you really succeed in capturing all the differences in appearance of which you are aware by means of such demonstrative iterations? That is, might you really manage to identify, with a suitably populous crowd of ‘thises’ or ‘thats’, a set of distinct specific shapes that appear to modify a specific area before you in a way that completely coincides with every change in how the scene appears to you? There is no reason to think so, if you do not have even a fleeting capacity to recognize and distinguish the various ‘thats’ and ‘thises’ allegedly discriminated. Do you? Consider any reasonably complex and variegated visual field — such as a page of ordinary text seen with normal acuity. As your gaze moves across a line of text and down the page, how will you capture the continual variation in appearance, by saying (even demonstratively) which

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

129

shapes appear to you when, both in the area where you are looking and where you are not? Singling out a set of specific shapes and locales individually that will take in the continual dynamic transformation of the entire visual field seems obviously hopeless. Maybe you will say, in one grand gesture encompassing an extended episode of looking: ‘these (various items) look to me these (various) shapes at these times.’ But you must not pretend that mouthing such words somehow guarantees you know what you are talking about here — specifically which various shapes, or items, or times are in the class of ‘these’. Certainly here it will be clear that you have no capacity to recognize and distinguish all the varying shapes you would allegedly be picking out with this characterization. This ‘grand gesture’ turns out to be an empty one. And from the fact that the way the shapes of what is before me appear varies over time it doesn’t follow that there simply must be some specific set of distinct shapes that maps onto this variation such that just these shapes appear to me at just these times in specific places. At this point one might turn back to the idea that seeing is, after all, more like getting a ‘picture in one’s head’ than having words in it. So perhaps specifically what spatial details are apparent when can be expressed and identified by means of an image, in a way that detaches my understanding of the changing appearances from any notion of myself as engaged as looking at this or that. But any image also seems to fail at capturing the manner of appearance, in a way similar to that we just discovered. Either the candidate image will have more represented in it of the apparent region than was actually apparent to me, or the same image would also be offered to capture how it appeared to me, even where the manner of appearance was not the same (since, again, I am looking at the scene before me somehow differently). Once more, either way one has not found a manner of conceptualizing the appearance that is commensurate with its variation, and which detaches one’s understanding of that variation from one’s understanding of oneself as a looker. So, in these cases I can judge that changes in appearance are the sort that occur when I am looking at something (or not looking at it, or looking at it more, or looking at it less). I can tell from a first-person point of view what I am looking at, whether I am looking at it more or less, and I can understand these changes in how something looks to me relative to this. But I think of looking at something as involving movement. What I still do not have then is an understanding of this variation in appearance that is entirely purified of any association with the activity of looking, and thus detached from my understanding of

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

130

C. SIEWERT

myself as motor agent. When I concern myself in first-person reflection with the ordinary changes that occur in the manner of visual appearance as my gaze shifts around and explores the area before me, what I find is not something whose character I can conceive of quite independently of the activity that affords it — a ‘picture in my head’ whose contents and details alter as a result of movements in the visual apparatus. When I try to isolate a purely experiential component of ‘looking at’ — a ‘receptivity’ quite distinct from the ‘spontaneity’ of visuomotor activity — I do not succeed. For phenomenologically I do not succeed in thinking of the variations in appearance in all their fleeting and fluctuating subtlety, while holding quite apart any notion of myself as looking at things. But perhaps here it will be pointed out that, even if this shows us that we have no phenomenological grasp of the subtle dynamics of appearance independent of exercise of visuomotor skill that comprises ‘getting a look at’ something — its shape, location, texture, size — this does not yet yield the result that our understanding of the type of motor activity in question cannot be pulled apart from our understanding of the kind of experience it makes available. Perhaps when we think of what we’re doing as getting a look at the fork, or the design on the plate, we do have a concept of that type of motor activity, which would allow us to attribute the very same type of activity to a zombie or a robot, to whom nothing looks any way at all. It’s just that we think of the experience we have in terms of its being (in us at least) caused by movement of that type. But will this work? Note the question is not whether there is some way, in principle, of conceptualizing the history of movements of your eyes, head and so on, that imputes no experience to you, and which could therefore correctly be attributed to some imaginary ‘zombie twin’, or perhaps some robot imitation of you, to whom nothing looked any way at all. The question is whether the way you have of conceptualizing your activity when you think of yourself as looking at something involves such concepts. Well, consider the movements you have made in reading this page of text. Do you have some grasp of just what movements you made, independently of understanding them as the movements required to look at (and thereby read) the words found on the page, from left to right, top to bottom one line at a time? If you had some such grasp, it seems you should be able to use it to follow instructions like these: close your eyes and repeat the same sequence of eye movements you just made reading this page. But this seems quite hopeless. On the other hand, you can, in a sense, make ‘the same movements’ by re-reading the page if you’re allowed to look at it. This

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

131

seems to indicate that your understanding of what you’re doing when looking depends on things looking to you as they do when you look at them, and in the absence of this you have no conception of just what you are doing when you are looking. You conceive of the movements as, irreducibly, those required to read this text by looking at it. Maybe someone will be tempted by the thought: ‘Of course I have a conception of that way a region differently appears to me, as my gaze is drawn to it, as I look at it, quite detached from any notion of the direction of my gaze or my activity of looking. For there, I just expressed it — by saying “that way a region differently appears to me”.’ But nothing about this form of expression shows that the claimed independence of conception is achieved. This deployment of a demonstrative device does not in and of itself indicate the presence of a totally separate conceptual purchase on what is spoken of. (Consider the anaphoric use of ‘this’ at the beginning of the previous sentence.) And notice that the demonstrative phrase at issue was first introduced with essential support from a conceptualization in terms of the direction of the gaze: ‘that way it appears differently as my gaze is drawn to it.’ Another objection may arise here. ‘Haven’t you read the Meditations? Descartes shows how I can conceive of myself having experience of the very same character as I am and have been having, even as I suppose myself to have no body at all — and to be a demon’s plaything. Or, leaving Descartes aside, equally apropos for current purposes: I may suppose myself to be a brain in a vat, not moving at all — though “from the inside” the experience is the same.’ But these sorts of considerations do not seem to me to show that we can, in first-person reflection, conceive of the variation in our visual appearance, without thinking of ourselves as looking at things, thus as moving. What they suggest to me instead is that, if I were (for example) a brain in a vat of the sort I am asked to envisage, I would be defective not only in my thoughts about my surroundings and my body, but also — at least to a significant extent — in my thoughts about the character of my own experience. And perhaps it reveals that when I consider the matter more closely I see that I do not succeed in thinking of my own current and actual situation as that of a disembodied brain, unless I could stop thinking of myself as looking at (and touching) things. But I cannot do that without losing a grip on the character of my own experience. It should perhaps be no great surprise that, when it comes down to it, I cannot really think of my current situation as that of the brain in the philosophical fable.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

132

C. SIEWERT

But even if I cannot make out the character of my own experience, while consistently conceiving of myself as a disembodied being, may I not conceive of some disembodied being (brain in vat or Cartesian ego) having experiences of the same character as mine? I can consider what it’s like for me to look at things before me, and entertain the notion of a disembodied being having experience such that what it’s like for it to have its experience is none other than what it’s like for me to have mine. And then: may I not in this fashion imagine that I have all the phenomenal features this entity has? And isn’t this to say I have conceived of my phenomenal features as purely phenomenal, in isolation from my comportment? Well, perhaps I can conceive of many of my phenomenal features in this manner, purified of movement. But I have done this in a way that is parasitic on conceiving of my phenomenal life that is not free of a notion of myself as an embodied mover — a way that presupposes what we might now call a ‘comportmental’ phenomenal concept: ‘looking at’. And we still have not analysed that concept into a purely phenomenal and purely motor factors. So the point remains that looking at things, understood phenomenologically, is not to be decomposed into a purely sensory or purely ‘mental’ part, and a purely ‘behavioural, physical’ part. Looking at things is understood as a form of behaviour. However, this is not ‘behaviour’ as Descartes, Skinner or Quine would see it, but essentially, experiential (‘lived’) behaviour, Merleau-Pontyan comportment. Paraphrasing Merleau-Ponty, I might say: looking at things is, for me, indissoluably both consciousness and movement. Something similar emerges, I believe, when we shift from the case of looking to examine the case of touch and feeling. Much as more of what you see appears to you than what you’re looking at, so also you feel more of what you touch than the area with which your body makes contact. But we have no conceptual, phenomenological grasp of this ‘what one feels’ and its variation as one moves that thoroughly prises this apart from one’s tactile engagement with one’s surroundings — that detaches it from one’s awareness of how one touches the world.

4. Does this Help with the Gap? The activities with which we have just been concerned are pervasive and fundamental aspects of cognition. For we have been speaking here of that through which object constancy in perception arises. Or (to revert to Kantian language) we are concerned with that through which objects are given to the understanding. For it is by looking at things in such a way as to constitute, though looking, over time,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

133

coherent appearances of their shape, size and position, that there are (visible) objects ‘for us’. Thus adequately developing these points about looking and touching could be consequential for our understanding of mind. A thorough phenomenological treatment of this, including an examination of the complex parallels and differences between vision and touch would be an involved affair. All I want to do here is to show in a preliminary fashion how we might go about filling the lacuna I find in Thompson’s argument, by dwelling on the phenomenology, while remaining true to the sources of his inspiration, and his general orientation. I also want to bring out how this offers an importantly different way of breaking the Cartesian spell that radically estranges mind and body than is to be found in the behaviouristmaterialist tradition that has dominated Anglophone philosophy of mind. But this last point is not yet satisfactorily clear. To make it clearer, let’s return to what prompted these reflections about ‘looking’ — the challenge of showing that the Thompsonian picture as characterized by (a)–(c) above really can be elaborated and defended so as to distinguish itself not only from options (1) and (2), but also from (3). We saw that there was a difficulty in making good the claim that a proper phenomenological understanding of perceptual experience finds it so bound to embodied activity that, in a non-trivial sense, the same perceptual functions that occur in our world would not occur in a putative ‘zombie world’. The worry was that phenomenologically there would be none but a trivial (because merely stipulative) sense in which the character of experience was necessarily bound to embodied activity. Any consciousness-involving form of motor activity we found in our own case, which could not be projected into the consciousness-free zombie world, could really just be seen as a conjunction of distinctly conceivable experience types and movement types, no more than contingently causally related. So the view on offer would provide no real alternative to (3). Once again we would ultimately come back to the proposal that we think of the mind–body relationship in terms of fundamental psychophysical principles joining consciousness (as phenomenologically conceived) to such physical properties as would be strictly guaranteed by the microstructure, and could obtain in a world without consciousness. But I hope to have given support to the prospect of a genuine alternative here, by suggesting a way to defend Thompson’s view. Understood phenomenologically, looking (and, I would argue, touching) involve movement whose unity of kind is conceived of by reference to the manner of experience it affords, which experience in turn cannot be conceived of entirely in detachment from

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

134

C. SIEWERT

that kind of movement. Basic phenomenal concepts of experience are comportmental phenomenal concepts. So, from a phenomenological perspective, the relation between experience and motor activity cannot be correctly represented as a contingent connection between pure experience and mere movement. In other words, a phenomenological understanding of the relation of experience and movement does not leave us with brute psychophysical contingencies, as does option (3) — for it does not divide the world into the purely phenomenal and the purely physical. But now, even if all this is granted, it still is not exactly clear where this puts us with respect to the challenge of the explanatory gap. Perhaps we cannot explain why any world with the very same microstructure as ours must have consciousness in it. If that is required to close the explanatory gap, no prospect of closure has been offered. Nor is there any proposal in the offing as to why purely mental and purely physical domains should be linked in a particular manner, and not some other. But still we may discern some significant explanatory aim that requires neither of these. The basic goal should instead be to get a precise and general understanding of how an organism’s embodied nervous system, in interaction with the environment, can generate the sort of movements that constitute the appearance to it of things in that environment — and thus how comportment comes to be. But once again: how is this supposed to help? It might seem that, even if Embodied Consciousness is distinguishable from (3), it is really not all that different. Whereas Descartes and his descendants leave us with a brute contingency linking the domain of physical properties with that of mental properties, the phenomenologist leaves us with a brute contingency between the physical universe (what’s fixed by fundamental physics) and the ‘life-world’ of our comportment, in which we look at and touch things. In either case we find no evident necessity joining the universe of natural science with the world of experience. In the end, aren’t we still simply left with this naked contingency, and a question about our intellectual comfort with this failure of necessitation? Thompson is sensitive to this problem. It is what he addresses when he concedes that he might be taken to have simply replaced the old mind–body problem with a new ‘body–body’ problem (pp. 236–7; for now instead of the problem of how to relate mind to body, we have the problem of how to relate the experiential body to the physiological body). His response is that such a replacement is progress, since the second problem is in principle more tractable, as now there is at least clearly a ‘common term’ viewed under different aspects to deal with.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

EMBODIED CONSCIOUSNESS

135

Let me try to say why it seems to me that the Thompsonian alternative does indeed give us an interestingly different approach to the issue of the explanatory gap. Perhaps then Thompson can explain how or whether he agrees. Maybe this will help (or again, maybe Thompson can explain why it won’t, or isn’t needed). First, we may note that the activity of the brain makes the organism move — in saying this, as yet we reveal no deep ‘in principle’ gap, even if there are enormous theoretical challenges in understanding how it works. But now the question is, how should we conceptualize that movement, if we are to make sense of it — as we should if we want a realistic theory of cognition and behaviour — as a creature’s interaction with an environment? Well, in principle we could try to conceptualize it in the Cartesian/Behaviourist fashion as ‘bare movement’. But we could also follow our natural inclinations to think of movement from the outset in terms (like looking and touching) that are, from a phenomenological point of view, tied conceptually both to consciousness and to movement, inseparably. We could conceptualize movement as comportment. And if this second framework helps us find intelligible patterns in our own and other animals’ interaction with their surroundings we have no reason to abandon it. (Anyway, it is doubtful that we really ever manage to sustain for long the intellectually induced autism of behaviourist methodology in a natural setting.) The burden of proof then would fall on someone who claimed we should attribute only some zombified analogue of looking and touching to ourselves and other animals. But now, from this perspective, we do not have the usual problem of how to put consciousness back into a world of matter in motion once it has been exiled from our conception of that world. For the conception of motion that matters, for the purpose of making good biological sense of ourselves and other creatures in the actual world, is the phenomenological one, the comportmental one, from which consciousness is not to be exiled in the first place — since it is inextricable from that conception. If, nevertheless, there remains a sense in which the zombie world is conceivable (and even in a broad sense (metaphysically) possible), because we cannot conceptually derive comportment from microphysics, this should not trouble us. For whatever way of conceptualizing movement and mid-sized entities’ interaction with their surroundings would be suitable to some world from which we conceive all the consciousness (sadly) drained away, this should serve as no guide for how we make sense of this, our own world — inhabited by billions of conscious beings, looking at and touching things. And if there remains a sense in which a disembodied consciousness is conceivable,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

136

C. SIEWERT

this doesn’t affect the fact that the basic conception of visual consciousness, in terms of which we understand ourselves, is one in which movement and consciousness are indissoluably fused, and that concept is the concept of comportment in terms of which we should couch our explanatory project. Now it does seem that even if all this is embraced it would not put to rest all worries that we really don’t know just how widely to extend comportment. Do salamanders have comportment in a sense that incorporates consciousness? Do dung beetles? What about the artefacts that roboticists dream of constructing? Perhaps these and similar questions are containable epistemic problems that do not really threaten our prospects of explaining the comportmental consciousness to be found in ‘central’, allegedly less problematic cases (to be found in humans, and most (all?) mammals). But this is a challenge that needs to be faced. In any case, what’s different (and, I think, refreshing) about the Thompsonian picture is that, while it can accept the modern idea that there is a way of conceiving of physical reality as a law-governed self-contained whole without thinking of it as inhabited by experiencing beings, unlike other approaches it refuses then to see the challenge of explaining consciousness as that of how (either causally or epistemically) to reintroduce consciousness into the universe thus conceived. For if we’re trying make sense of our interaction with our environment, the terms in which we do this should not be stripped of consciousness, as subjectively understood, to begin with. And the task of explaining consciousness is, most basically, the task of explaining that interaction so conceived. We find we can oppose the Cartesian estrangement of mind and body not (like a behaviourist) by spurning or belittling the reflective turn to consciousness, but by engaging in it more deeply. The Cartesian, we might say, turns to consciousness in a way that alienates us from our bodies, while the behaviourist (and many modern materialists) lock us into a way of thinking about our bodies that alienates us from consciousness. We can overcome both forms of alienation by discovering the unity of body and consciousness in sensorimotor life. What Descartes (and others) have put asunder we can rejoin — from the preferred Cartesian standpoint, that of consciousness, i.e. phenomenologically, ‘from the inside’ if you like — which after all will be the only truly satisfying way to heal the split. This little exhortation I hope is faithful to the spirit of Thompson’s view. But perhaps I have strayed too far from this. Even if I have, then maybe setting me right will afford him a good opportunity to make his views yet clearer.

EMBODIED CONSCIOUSNESS

137

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Perhaps I should say something about my fears of having strayed. In spite of my generally supportive response to Thompson’s vision, I do have some misgivings that are perhaps not trivial, relating to aspects of his account that I have so far neglected, and which may ultimately show that my attempts to bolster the phenomenological side to the story are really not fully in line with his approach. A key aspect of Thompson’s account has to do with how he finds a kind of ‘inwardness’ constituted by a living thing’s activity of self-organization — a kind of ‘immanent purposiveness’ relative to which features of its environment have ‘valence’ for it, and with respect to which its activity is normatively assessible. This kind of inwardness, he says, already goes beyond what can be captured in an ‘external’ conception of ‘structure and function’ — and is a ‘precursor’ to phenomenal consciousness (pp. 128–62, 222–5). One thing that is not clear to me is how this kind of ‘inwardness’ is supposed to relate to the sort of subjectivity one might think integral to phenomenal consciousness itself (and not merely its ‘precursors’). Thompson connects consciousness with the notion of empathy (see pp. 162–5). Perhaps the idea here is that phenomenal consciousness is that which by its very nature somehow demands a kind of empathetic understanding, an understanding ‘from the inside’ — in the sense that forms of consciousness need to be understood by sharing in them oneself, if not actually, then in imagination. Real consciousness is present just where that kind of empathetic regard is called for. This seems to me on the right track — though of course the devil is in the details. But now I worry that there is some gap not fully acknowledged here that threatens to become sharper and wider as we are clearer about these two ways in which something may have ‘inwardness’ — namely: (1) being such as to maintain (either relatively well or badly) a boundary between itself (inside) and others (outside); (2) being such as to serve as a suitable focus of empathetic regard (something that can be understood ‘from the inside’). I get the idea that the first is somehow a necessary but not-quite-sufficient condition of the second, but the details are hazy for me. Does Thompson mean to show us just how (1) can be built up to get us to (2)? Or is that perhaps a gap he does not aim to close? Do we need to close that gap to explain consciousness?1

References Armstrong, D. (1968) A Materialist Theory of Mind, London: Routledge. [1]

I wish to thank Evan Thompson for discussions of his views, and Jeff Yoshimi for much helpful feedback on an early draft of this essay.

138

C. SIEWERT

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Block, N. & Stalnaker, R. (1999) Conceptual analysis, dualism, and the explanatory gap, Philosophical Review, 108, pp. 1–46. Chalmers, D. (1996) The Conscious Mind, Oxford: Oxford University Press. Dennett, D. (1991) Consciousness Explained, Boston, MA: Little, Brown. Siewert, C. (1998) The Significance of Consciousness, Princeton, NJ: Princeton University Press. Siewert, C. (2010) Saving appearances: A dilemma for physicalists, in Bealer, G. & Koons, R. (eds.) The Waning of Materialism, Oxford: Oxford University Press. Tye, M. (2000) Consciousness, Color and Content, Cambridge, MA: MIT Press.

Robert Van Gulick

Life, Holism and Emergence: Converging Themes Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

1

Evan Thompson’s Mind in Life adds a welcome voice to the current conversation about consciousness and how it should be understood and studied. Consciousness is a complex and many-sided phenomenon, and understanding it will surely require a wealth of perspectives, models and frameworks. A spirit of theoretical and conceptual pluralism is clearly in order: ‘Let a thousand flowers bloom; we will need that many and more to understand consciousness.’ In that regard Mind in Life merits our attention and adds useful new resources to our mental toolbox. It aims to shift the way we look at mind and self and consciousness, and from those altered viewpoints let us see more clearly things that might otherwise escape attention. Thompson’s intent is to re-centre our view of consciousness by situating it within the twin traditions of Husserlian phenomenology and the biological theory of autopoietic self-organizing systems, two approaches that find their synthetic overlap in notions such as the subjectivity of lived existence. The rhetoric of Mind in Life is often that of the radical or iconoclast, as intent on showing the errors and mistakes of the supposed mainstream tradition as on presenting alternative perspectives. However, I think the value of Mind in Life can be best appreciated by viewing it as a matter of convergence rather than divergence, i.e. as a matter of important themes that unfold in their respective but parallel ways Correspondence: Email: [email protected] [1]

I dedicate these comments to the memory of John Haugeland who I knew and admired since our days as graduate students together many decades ago. John died recently (June 2010) following a heart attack that ironically occurred during an event honouring him on the occasion of his retirement from the University of Chicago where he completed his career as the David B. and Clara E. Stern Professor of Philosophy.

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 139–47

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

140

R. VAN GULICK

within different research programmes and literatures. In particular, many of the most important elements of Thompson’s view have their analogues within the mainstream literature he attacks. I had the good fortune many years ago (1980) to spend a few months in Berkeley, California at a summer institute on ‘Continental and Analytic Perspectives on Intentionality’ funded by the National Endowment for the Humanities. The institute was created by Hubert Dreyfus and the late John Haugeland with the express aim of getting philosophers like myself who had trained in the analytic tradition to spend the summer studying the phenomenological literature. (Given that it was not long after the North Vietnamese had set up ‘re-education camps’ in the South following the fall of Saigon, some of us jokingly referred to the summer institute as ‘Camp Dasein’.) The Dreyfus-Haugeland view, ably supported by a diversity of lecturers that summer, was that all the really important disputes about the nature of mind and intentionality crosscut the divide between the analytic and phenomenological traditions. As they presented the history, the key disagreements were not between the two traditions but between parallel groups within each tradition. On the big issues about intentionality and meaning — such as the role of the social versus the individual or of the practical versus the intellectual or sensory — the partisans of each view within the analytic tradition had their close analogues with the phenomenological tradition and vice versa. I do not know enough about the history of phenomenology to say with confidence that Dreyfus and Haugeland were right, but they made a plausible case and I have found that way of viewing things quite helpful ever since. Thus I would like to offer some comments on Thompson’s Mind in Life in that same spirit. There are many things Thompson says about life, mind and consciousness that I find quite plausible and helpful, but in most cases I believe they have close analogues within the mainstream literature he attacks. The ‘mainstream’ has a complicated flow, and no doubt Thompson swims against some of its currents, but other major flows move more in parallel with his views and end up close to where he does himself. My aim in noting those similarities is not to diminish Thompson’s contribution but to enhance the plausibility and value of his ideas by showing their convergence with similar developments in other literatures including some at the core of the mainstream itself. Moreover, even when different research programmes arrive at similar conclusions, their full import depends upon their context, and seeing how those implications play out within each framework can

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

LIFE, HOLISM & EMERGENCE

141

give us a better understanding of their overall value. This is hopefully a case in which one plus one adds up to more than two. I will discuss this general theme with respect to just three specific topics addressed by Thompson in Mind in Life: the biological nature of mind, the holism of the mental including the interdependence of mind and environment, and the nature of emergence. However, the general point could be applied as well to many other proposals made in the book. As its title suggests, Mind in Life views minds as a biological feature of the world. Indeed, in its opening sentence Thompson announces that his theme is ‘the deep continuity of life and mind’ (p. ix). Surely he is right, but few today hold otherwise. Descartes long ago regarded living things as mindless mere machines and restricted consciousness and reason to a distinct realm of immaterial souls whose whole essence was to think. Though some latent Cartesian intuitions may still lurk implicitly in our thinking about the mind, few if anyone today explicitly denies the tight link between mind and life. There is nearly universal agreement that all known natural minds are biological, and that no genuine artificial minds as yet exist. Thus actual minds all fall within the domain of living things. Opinions may vary about which organisms count as having minds, but that debate takes place against a general view of living things as essentially engaged in informationally-sensitive purposive (or goal-directed) behaviour. Their capacities to do so vary widely and come in many forms and degrees, so perhaps it is better to talk in term of mindedness rather than of minds as things. There need be no sharp boundary between mind and non-mind along that spectrum. As Thompson says, the continuity between life and mind is deep, but in that regard his view is solidly in the mainstream. For example, many contemporary functionalist theories of mind are explicitly committed to reading the notion of ‘function’ in a bio-based teleological sense (Lycan, 1987; Sober, 1990; Van Gulick, 1980). So-called teleo-functionalism specifies its defining mental roles in terms of the purposes they serve in guiding the organism’s successful informationally-sensitive goaldirected behaviour. Of course, multiple theories might agree that mind is deeply biological in nature and still draw very different implications from that fact. But again much of what Mind in Life has to say on those issues accords with mainstream theories. For example, according to Thompson the basic autopoietic nature of life as a self-organizing and self-replicating system or process is the source and ground of meaning. It is the autopoietic system itself that creates significance and difference. It is

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

142

R. VAN GULICK

only in relation to that autopoietic order that anything comes to have meaning or valence, and thus already in its basic form it gives rise to a primitive type of subjectivity. Even for the unicellular organism, there is some subjectivity in so far the world exists for the organism in a way defined by its needs and its possible modes of mutual engagement with that world, what Thompson might regard as its enactive embedding in its world. It’s a plausible and attractive position, but at its core it is not at odds with mainline teleo-functionalism or the many biosemantics theories of meaning that have been developed in considerable detail over the past twenty-five or so years (Millikan, 1984; 1989; Neander, 1995). Thompson might disagree with some of the specifics of those biosemantic models — e.g. consider his stated disagreement with Dennett’s biological account of the origin of meaning (Thompson, 2007, p. 160) — and some of those differences might be important — e.g. Thompson might fault some as being too adaptationist. But he and they are nonetheless in agreement on the core idea that meaning and purpose are grounded in the engagement between selforganizing, self-replicating systems and their world. Their shared commitments far outweigh any disputes between them. With specific regard to consciousness, Thompson argues that viewing the mind in a biological context helps narrow the explanatory gap problem — the puzzle of explaining how consciousness might result from a non-conscious substrate. According to Thompson, his biological approach does not in itself solve the problem, but it nonetheless transforms it into the potentially tractable ‘body–body problem’ of explaining how the ‘lived body’ can be produced by the ‘living body’. He writes, ‘First, the gap is no longer between two radically different ontologies (mental and physical), but between two types within one typology of embodiment. Second, the gap is no longer absolute because in order to formulate it we need to make common reference to life or living being’ (p. 237). He contrasts this situation with traditional or mainstream approaches: In the hard problem, the explanatory gap is absolute because there is no common factor between the mental and the physical (and there can be none given how they are defined). Hence the main options are to accept the gap as a brute ontological fact (dualism), to close the gap by reduction (materialism or idealism) or bridge the gap introducing some third and speculative ‘extra ingredient’ (for which there is no scientific evidence or motivation). (p. 237)

This list of mainstream options seems strangely incomplete in at least two respects. First, non-reductive physicalism is not mentioned despite the fact that it has been a widely held view at least since the

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

LIFE, HOLISM & EMERGENCE

143

mid-1970s. Such theories aim to combine a commitment to the physical as the ultimate underlying basis of the mental with non-reductive views about the autonomy of the special sciences and a pluralistic denial that the physical sciences and physical concepts can suffice for all our legitimate explanatory and descriptive purposes (Fodor, 1974; Boyd, 1980; Van Gulick, 1992). Non-reductive physicalism is not without its critics (Kim, 1989) and its problems, but it is surely an option that is taken seriously by many philosophers. Secondly, most mainstream physicalists hope to close the explanatory gap not in a single step but by building multiple theoretical links from both directions, and biologically-based models are typically prominent among those proposed to play a mediating role. For example, teleo-functionalist theories of the sort mentioned above aim to narrow the gap by explaining how the mental might be explained in terms of the teleological, where the latter notions are to be interpreted and understood in biological terms. William Lycan, for example, argued in Consciousness, ‘Our job as philosophers of mind was to explicate the mental in a reductive (and non-circular) way, and this I am doing by reducing mental characterization to homuncular institutional ones, which are teleological characterizations… I hope, and am inclined to believe, that the teleological characterizations that Homunctionalism requires can be independently explained in evolutionary terms’ (Lycan, 1987, pp. 42–3). Lycan’s approach may not succeed, and perhaps there are good reasons for preferring Thompson’s specific model, but they share a common belief that closing the gap requires having biologically-based notions on both sides of the divide. Again it is a matter more of convergence than of divergence. Let us turn to the second and third topics on our list: the holism of the mental and the nature of emergence. The two are interconnected so it will be apt to consider them together, and doing so will reveal at least as many parallels with mainstream views as disagreements. Thompson stresses the holistic nature both of mental systems and of the mind-world complexes they form. He argues that the parts composing such systems often do not have any reality or existence independently of the wholes of which they are parts. The parts of such systems depend upon their holistic context for their existence and character just as much as such systemic wholes depend on their parts. The dependence runs equally in both directions. He applies that insight both to minds and to the environments within which they exist. Thompson stresses that the world with which enactive minds engage cannot be adequately characterized from a purely objective third-person point of view. The organism lives within an umwelt, a domain of

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

144

R. VAN GULICK

meaning or valence and possible engagement defined by the organism’s needs and sensorimotor capacities. Thompson’s holism seems quite plausible, but once again it is in accord with many mainstream views. Indeed one of the central ideas behind functionalist theories of mind has been the recognition that the essential nature of mental states and processes depends upon the interactive roles they play with other mental states within a larger system that constitutes a mind. Those individual states could not be the states they are in isolation but only within the context of a mind with all its other similarly context-dependent parts. This multiply ‘tail-biting’ interdependence of state on state, as well of state on whole and whole on state, was one of the central motivating ideas behind functionalism which is about as close to a mainstream view as one can get in the contemporary philosophy of mind. Nor is the functionalist’s definitional circle restricted to the domain of mental states and process. In many cases, it extends out into the world just as on Thompson’s view. This is readily evident in the case of so called ‘common sense functionalism’ which draws upon the norms and patterns of everyday ‘folk psychology’ to supply the relevant defining functional relations. Folk psychology clearly characterizes the world with which the agent interacts in highly intentional ways that make sense only in relation to the minds that engage them. Folk platitudes for example tell us that a person is likely to express offence if insulted, make a purchase if offered favourable terms, or decline an invitation extended with a hint of insincerity. The world of folk psychological interactions is a highly intentional domain suffused throughout with just the sort of meaning and valence Thompson invokes. Moreover, teleo-functionalists will make the same point about organism-environment interactions in general; the taxonomy of situations, stimuli and response that the teleo-functionalist invokes to characterize the environmental pole of the relevant functional relations will be highly sensitive to the interests, needs and capacities of the organism. That is just a standard part of the model. Thompson will of course agree, and my intent is not to raise an objection to his view but merely to note the affinities between what he says and widely accepted mainstream positions, which hopefully increases the plausibility of each. Just as functionalists categorize the environment in ways that are sensitive to mental facts, so too in the other direction they often embrace externalist views about the nature of mental content. Many models individuate mental states and their intentional contents partly on the basis of external factors about the cognitive agent’s environment,

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

LIFE, HOLISM & EMERGENCE

145

including physical, historical and social factors. The relevant external features might be those of the individual organism’s immediate context or of the external milieu associated with its past, its social group or its evolutionary predecessors. From a functionalist perspective, externalism provides a variable parameter that can be set in diverse ways depending upon one’s explanatory goals and interests. Given a spirit of pragmatic pluralism, there need be no single right way to individuate content or to factor in the role of external features. Thompson emphasizes the inter-definitional link between mind and world, but for some purposes a more internalist mode of characterizing mental kinds and contents may be best, while for other purposes some more externalist scheme would serve better (Van Gulick, 2004). A similar spirit of pluralism is in order with regard to emergence, which is of course closely related to holism since both concern the sorts of dependence relations that hold between parts and wholes. As noted, Thompson argues that both in minds, as well as in other nondecomposable systems, wholes may have parts that have no reality independent of the larger whole of which they are parts. At some points, Thompson is cautious to assert emergent holism only as an epistemic (or conceptual) claim rather than as an ontological one. He writes, ‘“Nondecomposability” and “decomposability” are heuristic epistemological categories not ontological ones. It is not my intention to argue for a metaphysical thesis of ontological holism on the basis of nondecomposability’ (p. 423). However, he parenthetically adds, ‘which is not to say that nondecomposability is merely epistemological in the sense an ontological reductionist would assert’ (p. 423). In contrast, at the very end of Mind in Life, he makes a more radically ontological claim. In the final sentence of the book (Appendix B) he writes, ‘Phenomena at all scales are not entities or substances, but relatively stable processes, and since such processes achieve stability at different levels of complexity, while still interacting with other levels, all are equally real and none has absolute ontological primacy’. It is perhaps with this very last claim that Thompson swims most directly against the mainstream currents, whose flow more closely follows the moderate path he rejects, that of combining an epistemic form of holism and emergence with a commitment to ontological physicalism. Thompson rejects the physicalist’s ontological reductionism. He does so in part because he interprets it in an implausibly strong way. He writes, ‘According to the ontological thesis of part/whole reductionism (“mereological supervenience”) all the properties of a whole are determined by the intrinsic (nonrelational) properties of its

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

146

R. VAN GULICK

fundamental parts’. Given the qualifier ‘mereological’ the definition is reasonable, but few if any ontological reductionists read it as a mereological thesis which explicitly excludes any relational facts. In Thompson’s sense, reductionism would be as false of a ballpoint pen or of a snowflake as of a mind or brain. What clearly matters are not just the intrinsic properties of the parts, but how they are combined and organized. Thus what ontological physicalists assert is that the properties of wholes are fully determined by the intrinsic properties of their basic parts plus their mode of composition and arrangement. One can be an ontological physicalist in that sense while still being a pluralistic non-reductionist, one who holds that the physical sciences do not come close to providing all the conceptual and theoretical resources that we as situated agents require to adequately understand the world in all its complexity. The fact that all those higher-level facts and properties ultimately derive from their underlying basic physical substrate does not mean that physics provides all the cognitive tools we need to understand those complex properties and guide our practical engagement with them. That at least is the standard mainline view of non-reductive physicalists. Thompson rejects that view. On his alternative picture there is no privileged base, no level of reality on which all else depends. In this respect Mind in Life does indeed end on a radical and revolutionary note. I am not persuaded of its truth, but if it could be shown it certainly would qualify as re-centring our view of world and mind.

References Boyd, R. (1980) Materialism without reductionism: What physicalism does not entail, in Block, N. (ed.) Readings in Philosophy of Psychology, Vol. 1, Cambridge, MA: Harvard University Press. Fodor, J. (1974) Special sciences, or the disunity of science as a working hypothesis, Synthese, 28, pp. 77–115. Kim, J. (1989) The myth of nonreductive physicalism, Proceedings and Addresses of the American Philosophical Association, 63, pp. 31–47. Lycan, W. (1987) Consciousness, Cambridge, MA: MIT Press. Millikan, R. (1984) Language, Thought and Other Biological Categories, Cambridge, MA: MIT Press. Millikan, R. (1989) Biosemantics, Journal of Philosophy, 86, pp. 281–297. Neander, K. (1995) Malfunctioning and misrepresenting, Philosophical Studies, 79, pp. 109–141. Sober, E. (1990) Putting the function back in functionalism, in Lycan, W. (ed.) Mind and Cognition, Oxford: Blackwell. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of the Mind, Cambridge, MA: Harvard University Press. Van Gulick, R. (1980) Functionalism, information and content, Nature and System, 2, pp. 139–162.

LIFE, HOLISM & EMERGENCE

147

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Van Gulick, R. (1992) Nonreductive materialism and intertheoretical constraint, in Beckermann, A., Flohr, H. & Kim, J. (eds.) Emergence and Reduction, pp. 157–179, Berlin & New York: De Gruyter. Van Gulick, R. (2004) Outing the mind, in Schantz, R. (ed.) The Externalist Challenge: New Studies on Cognition and Intentionality, pp. 255–284, Berlin & New York: De Gruyter.

Michael Wheeler

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Mind in Life or Life in Mind? Making Sense of Deep Continuity 1. The Deep Continuity Thesis One of the many ground-breaking themes in Evan Thompson’s rich and thought-provoking book Mind in Life: Biology, Phenomenology, and the Sciences of Mind is his distinctive development and defence of an idea that he calls the deep continuity of life and mind — henceforth just deep continuity. Thompson introduces this idea as follows: ‘life and mind share a set of basic organizational principles, and the organizational properties distinctive of mind are an enriched version of those fundamental to life. Mind is life-like and life is mind-like’ (p. 128).1 In this initial characterization, deep continuity is (as Thompson notes) tantamount to what others (e.g. Godfrey-Smith, 1994; Wheeler, 1997) have called the strong continuity thesis of life and mind.2 Thompson claims, however, that these other theorists, in concentrating on organizational, functional or behavioural properties, have ignored a crucial aspect of life-mind continuity, namely its phenomenological dimension. The corrective, then, which recruits an insight that Thompson traces back to the work of Hans Jonas (1966), is to recognize that ‘certain basic concepts needed to understand human experience turn out to be applicable to life itself’ (p. 129). Such concepts (more on which below) include needful freedom, self-transcendence, Correspondence: Email: [email protected] [1]

Unless otherwise indicated, all page numbers refer to Thompson (2007).

[2]

For example, Godfrey-Smith (1994, p. 83) characterizes strong continuity as the view that ‘[life] and mind have a common abstract pattern or set of basic organizational principles… Mind is literally life-like’.

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 148–68

MIND IN LIFE & DEEP CONTINUITY

149

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

and immanent purposiveness. In other words, ‘certain existential structures of human life are an enriched version of those constitutive of all life’ (p. 157). In what follows I shall offer an all-to-brief response to Thompson’s account of deep continuity. The bulk of the paper will be devoted to a critical analysis of the key elements out of which Thompson constructs his account, an analysis during which I highlight and discuss a number of internal problems and unresolved issues for the view. That done, I shall argue that Thompson’s specification of the relationship between autopoiesis, cognition and life has the unpalatable effect of closing off what is arguably the most plausible form of deep continuity. I shall end by sketching (but no more than sketching) the thought that the most plausible form of deep continuity may yet be established within the kind of generic conceptual framework to which Thompson doesn’t exactly warm; a framework, that is, which holds that the fundamental structures of life and mind are to be understood in terms of notions such as information and representation.

2. Autonomy and Cognition What is the precise content of Thompson’s deep continuity thesis? To make progress on this question, we need to reflect on various theoretical concepts that Thompson deploys as he explicates his version of the so-called enactive approach to life and mind. For the most part, these concepts are drawn (sometimes in a revised form) from the theory of autopoiesis — that powerful, but difficult to unravel, web of ideas concerning life and mind on which Thompson’s enactivism is built.3 Perhaps the place to begin is with the least specialized of the concepts at work, namely that of self-organization. A self-organizing system is one in which certain intrasystemic components, on the basis of purely local rules (i.e. without the direction of some global executive control process), interact with each other in nonlinear ways so as to produce the emergence and maintenance of structured global order. Self-organization is now recognized as being a widespread phenomenon. Regularly cited examples in the literature include the BeloussovZhabotinsky chemical reaction, lasers, slime moulds, foraging by ants, and flocking behaviour in creatures such as birds. Now, although self-organization is commonplace, only some self-organizing systems are what Thompson (following Varela) calls autonomous. For a [3]

See, especially, Chapter 5 of Mind in Life. For the canonical presentation of autopoietic theory, see Maturana and Varela (1980). For other autopoiesis-based statements of the enactive approach, see e.g. Varela et al. (1991), and Di Paolo (2005).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

150

M. WHEELER

self-organizing system also to be an autonomous system, the constituent processes of that system must ‘(i) recursively depend on each other for their generation and their realization as a network, (ii) constitute the system as a unity in whatever domain they exist, and (iii) determine a domain of possible interactions with the environment’ (p. 44; citing Varela, 1979, p. 55). Put slightly differently, an autonomous system is a network of interdependent processes whose recurrent activity (a) produces and maintains the very boundary that determines the identity of that network as a unitary system (such that what counts as inside the system and what counts as outside the system co-emerge as functions of that endogenous activity), and concurrently (b) defines the ways in which that system may encounter perturbations from what is outside it while maintaining its organization and thus its viability, that is, while not disintegrating. Where successive environmentally induced perturbations to the autonomous system trigger only states changes that remain within that system’s bounds of viability (such that organization is not lost), phenomena (a) and (b) together produce a history of structural coupling between that autonomous system and its environment. Enaction is the process by which significance or relevance is brought forth through structural coupling. The core idea here is that the process of autonomous organization, in establishing the distinction between the self-maintenance and the collapse of a system as a unity, institutes a norm of survival, and thereby the significance or relevance of certain environmental perturbations as either leaving the system organizationally intact or resulting in its disintegration. However, the bare norm of survival to which I have just given expression generates only an impoverished kind of significance, because it is based on no more than the robustness or conservation of systemic organization in the face of environmental perturbation. The system either survives any perturbations it experiences, or it doesn’t. As Di Paolo (2005) usefully puts it, the norm of survival in play so far is an all-or-nothing affair. What seems clear, however, is that domains of significance enacted on the basis of structural coupling are regularly more complex in nature, in that they are places ‘of valence, of attraction and repulsion, approach or escape’ (p. 158). To illustrate this with the kind of example that Thompson himself often uses (e.g. p. 158), bacteria swim towards the area containing the greatest concentration of glucose molecules. Thus, as a consequence of the specific metabolically realized autonomy of the bacteria, glucose emerges as — is brought forth as — significant for those organisms as food. The normative structure enacted here is not an all-or-nothing affair, but rather

MIND IN LIFE & DEEP CONTINUITY

151

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

a gradient of better or worse outcomes for the autonomous system. Navigating this gradient requires the system to be able to alter its behaviour in response to changes in its environment, so as to exhibit a sensitivity to differences between states. In some way (perhaps implicitly), the system monitors how it is doing with respect to maintaining its viability, and thus regulates its behaviour accordingly, in order to improve its situation. In other words, the autonomous system is now also an adaptive system. The enactive process of generating graded significance on the basis of adaptive autonomy is what Thompson, following others (e.g. Varela, 1991; Di Paolo, 2005), calls sense-making. The next idea from Thompson’s conceptual toolkit that we need in order to carry out our analysis of deep continuity is the aforementioned notion of autopoiesis (Maturana and Varela, 1980). Put crudely, autopoiesis is autonomy plus materiality. More specifically, to be autopoietic, an autonomous system must, through its own endogenous self-organizing dynamics, produce and maintain a material (or physical) boundary which distinguishes that system as a material (or physical) unity in the space in which it exists. The paradigm example of an autopoietic system is the single living cell, understood as a network of chemical reactions that produces its own membrane. The autopoietic organization of the living cell is, as Thompson describes it, ‘autonomy in the biochemical domain’ (p. 44). The precise relationship between autopoiesis and autonomy will concern us later, but one thing is worth noting immediately: autopoiesis is sufficient for autonomy.4 So now, what does all this tell us about life, mind, and the relationship between them? Let’s begin with the notion of cognition. Thompson explicitly defines cognition as sense-making. Thus he says ‘cognition, in the present context, means the activity of sense-making’ (p. 159). Given that autonomy and adaptivity are individually necessary and jointly sufficient for sense-making (see above), the identification of cognition with sense-making makes autonomy and adaptivity individually necessary and jointly sufficient for cognition. However, we need to know how to interpret the phrase ‘in the present context’. Thompson’s text strongly suggests that this qualification is there to highlight the fact that we are concerned, in the first instance, with cognition ‘in its minimal biological form’ (p. 159), where the minimal biological [4]

Although this accurately reflects the way Thompson introduces the term ‘autopoiesis’, he later gravitates toward using it in a ‘wide’ sense, so as to include adaptivity (e.g. p. 158). It will suit my purposes to retain the narrower usage, and that’s what should be assumed in the discussion that follows.

152

M. WHEELER

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

form of cognition is the first (simplest, most basic) instance of an enacted domain of graded significance in nature. Thus Thompson directs our attention towards what he calls ‘the natural roots of intentionality’, where what is at stake is an ‘analogue of the phenomenological notion of the disclosure of the world’, understood as activity that brings forth a domain of graded significance on the model of our glucose-sensitive adaptive bacteria (p. 159). In spite of this drive towards the minimal and the basic, one shouldn’t be tempted to think that anything less than sense-making through adaptive autonomy will be adequate for genuine cognition. As Thompson puts it: Cognition is behavior or conduct in relation to meaning and norms that the system itself enacts or brings forth on the basis of its autonomy. We have seen that sense-making requires more than minimal autopoiesis: it requires autopoiesis enhanced with a capacity for adaptivity… [A]utopoiesis plus adaptivity entails sense-making, which is cognition in its minimal biological form. (p. 159)

The attentive reader will no doubt have noticed that Thompson here talks of enhancing minimal autopoiesis rather than, as one might have expected from the momentum of my exposition, enhancing minimal autonomy. If we are concerned with the sufficient conditions for cognition, this is not an overly significant variation. As we have seen already, autopoiesis is sufficient for autonomy, so autopoiesis plus adaptivity will be sufficient for sense-making and thus for cognition in its minimal biological form. And it is ‘on the basis of its autonomy’ that an autopoietic system of the right kind (i.e. one whose autonomy is adaptive in form) is able to generate cognitive activity. However, if we are concerned with the necessary conditions for cognition, Thompson’s use of the term ‘autopoietic’ at a point where we might have expected him to use ‘autonomous’ immediately raises the question of whether there could be a non-autopoietic autonomous system. In other words, it raises the question of whether, for Thompson, autopoiesis is necessary for autonomy. It might seem that Thompson’s answer to this question is blindingly obvious. Indeed, he states that to ‘qualify as autonomous… a system does not have to be autopoietic in the strict sense (a self-producing bounded molecular system)’ (p. 44) and that ‘there can be autonomous systems that are not autopoietic if their constituent processes exhibit organizational closure in their domain of operation’ (pp. 106–7). We can understand these statements better if we consider the examples Thompson gives of autonomous systems, which, alongside the living

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE & DEEP CONTINUITY

153

cell, include the immune system (pp. 49, 65), the nervous system (pp. 46–7, 65), insect colonies (pp. 44, 65) and animal societies (p. 65). As we have seen, the living cell is an autopoietic system, but, as Thompson points out, these other autonomous systems are not. In each case the categorization of the system as autopoietic or not turns on the nature of (a) the relevant system-nonsystem boundary and (b) the domain in which the system in question exists, as a network of constituent processes that generates a unity. For the living cell, the system-nonsystem boundary (the cell membrane) and the domain of existence (the biochemical) are material. This specific kind of dual materiality is definitional of autopoiesis. We can contrast this with, for example, the nervous system and insect colonies. For the nervous system, the fundamental logic of which is to couple sensing and movement, the relevant system-nonsystem boundary and the domain of existence are fixed at the level of behaviour and intentional action (p. 49); and for insect colonies the system-nonsystem boundary and the domain of existence are fixed at the level of sociality and territoriality (p. 44). The issue, then, seems to be cut and dried: Thompson’s view is that autopoiesis is not necessary for autonomy. There is, however, an exegetical snag. There are passages in Mind in Life which strongly suggest that the interpretation just given must be wrong, and that, for Thompson, autopoiesis is necessary for autonomy. Consider, for example, the following sequence of claims: Agency and meaning require autonomy; minimal agency and meaning require minimal autonomy. Minimal autonomy depends on macromolecules but requires that those macromolecules be organized in a particular way, namely, in the autopoietic way. (p. 160)

It is hard to read this passage without taking away the message that autopoiesis is necessary for autonomy. If autonomy, agency and meaning in their minimal forms require autopoiesis, then the most natural thought is that all forms of agency, meaning and autonomy do. After all, ‘necessary for the minimal form of X’ sounds achingly close to ‘minimally necessary for X’. What breeds confidence that we must put any such thought aside as an interpretation of Thompson is that it clashes unhelpfully with his claim that the immune system, the nervous system, insect colonies and animal societies are autonomous systems. Recall the dual materiality of self-generated systemic boundary and domain of existence that, as we have seen, is definitional of autopoiesis. If autopoiesis is indeed necessary for autonomy, then the systems currently under consideration, by virtue of the natures of their self-generated systemic boundaries and domains of existence (see

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

154

M. WHEELER

above), would not count as autonomous. The most prominent difficulty would then be that, if autonomy is necessary for sense-making and thus for cognition, the nervous system (no longer an autonomous system) could not be a basis for cognition. And that really would upset the applecart. In the preceding paragraphs I have suggested that there is an inconsistency in Thompson’s treatment of the relationship between autopoiesis and autonomy, but that we can be confident that he would resolve that inconsistency in a particular way. Nevertheless, we should pause for a moment to wonder whether there is an alternative reading of the anomalous passage that would avoid plunging him into the difficulty in the first place. So let’s try something out. If we unpack ‘minimal’ as ‘minimal biological form of’, in line with Thompson’s analysis of cognition (see above), we might interpret him as holding that although the minimal biological form of autonomy, and thus the minimal biological forms of agency and meaning, require autopoiesis, more advanced forms of autonomy, agency and meaning are able to divest themselves of any such dependence. Attributing this view to Thompson would avoid the troublesome inconsistency with which we are presently concerned (and, incidentally, explain why he is comfortable substituting ‘autopoiesis’ for ‘autonomy’ when characterizing cognition in its minimal biological form — see earlier discussion). That said, it would saddle him with a new burden, by focusing critical attention on the contribution made by the materiality requirements that distinguish autopoiesis as a phenomenon over and above autonomy. In identifying autopoiesis as necessary for minimal autonomy, Thompson’s goal is to reveal ‘the ground from which the seeds of intentional action grow’ (pp. 160–1). But the only thing that autopoiesis adds to the concept of autonomy is the dual materiality of the systemic boundary and the systemic domain of existence. And it is genuinely hard to see what special substantive contribution is made to our explanation of the genesis of intentional action by recognizing that the minimal biological form of autonomy exhibits such dual materiality, given that such materiality is apparently expendable by the time that biology gets as far as a nervous system. After all, although the distinctive kind of materiality that autopoiesis signals is an interesting and important feature of the cell, the idea of minimal autonomy already provides for a self-producing unity capable of instantiating a history of structural coupling, and it is an enrichment of that very capacity for structural coupling, through the addition of adaptivity, that accounts finally for sense-making and thus for cognition. So the dual materiality of the systemic boundary and the

MIND IN LIFE & DEEP CONTINUITY

155

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

domain of existence is not required for the realization of these phenomena. In response to the preceding analysis it might be suggested that although the boundary and domain of existence of the autonomous system in question need not be material in the relevant sense, nevertheless there must be autopoietic materiality somehow at the centre of any system that genuinely has its own perspective on the world.5 Of course, some account would need to be given of what ‘at the centre’ means here, but in any case I think the considerations that figured in the original analysis are very likely to continue to apply. Consider: as it happens, biological cells are not merely autonomous, they are autopoietic. But let’s imagine that the physical building blocks of organisms were somehow autonomous but not autopoietic. Given that autonomy is sufficient to establish a self-producing unity, and that adaptiveness is a modulation of that organization, it seems that despite this assumed lack of any autopoietic dimension at the centre of things, cognition and intentionality would still enjoy an impeccable naturalistic grounding.

3. Autopoiesis and Life After some clarificationary discussion, we now have a grip on Thompson’s view of cognition: cognition is sense-making and so requires adaptive autonomy but not autopoiesis. Now, what about life? Thompson’s striking claim is that life is autopoiesis plus cognition (p. 158). Since cognition is sense-making as established by adaptive autonomy, this is formally equivalent to claiming that life is adaptive autopoiesis (adaptive autonomy in the biochemical domain). Thompson’s development of this view of life is complicated, but in the present context we can concentrate on two things — the neo-Kantian role played by autopoiesis and the neo-Jonasian role played by the combination of needful freedom and self-transcendence. Thompson uses the concepts of self-organization and autopoiesis to give a contemporary reinterpretation of Kant’s (1790) claim that organisms are natural purposes (pp. 129–49). In brief, the picture is that organisms, like artefacts, are systems in which each component exists for the sake of the others in the context of the organized whole, in such a way that the cause-effect relations that characterize the workings of the system are also means-end relations. That means that [5]

Something like this thought seems to be suggested by Di Paolo’s (2005) claim that autopoiesis ‘provides a self-distinct physical system that can be at the centre of a perspective on the world’.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

156

M. WHEELER

teleological (purposive) language is needed to characterize the system. However, whereas in the case of artefacts the purposiveness in question is extrinsic (e.g. established by the pre-existing plan of a human designer), in the case of organisms the purposiveness is intrinsic or, as Thompson prefers to say, immanent. A property is immanent if it is neither a ‘nonrelational [and so unanalysable, see p. 146] property of something internal to the system… nor a property determined by something outside the system… [but rather] a constitutive property the whole system possesses because of the way the system is organized’ (p. 146). The puposiveness of an organism is necessarily immanent because the means-ends relations that characterize its parts are established by the distinctive self-organizing and self-producing activity of the system according to which ‘each of its parts is both a product and a producer of the other parts, so that the system is a self-organizing whole’ (pp. 145–6). And Thompson takes it that if organisms were recognized to be autopoietic in nature, that would be sufficient to explain their distinctive self-organizing and self-producing character and thus the immanence of their purposiveness (p. 146). What are we to make of Thompson’s account of immanent purposiveness? It is worth just noting that there is a dispute to be settled here over whether or not any multicellular organism is itself strictly autopoietic, as opposed to being built out of autopoietic elements (see pp. 105–7 for discussion). Fortunately, however, having noted this issue, we can simply ignore it, since it seems open to Thompson to hold that even if multicellular organisms are not themselves strictly autopoietic, the immanent purposiveness of any multicellular organism may be inherited from the cells that together make it up, given that each of those cells will be an autopoietic system and thus will individually realize immanent purposiveness. So let’s agree that Thompson can make good on the claim that autopoiesis is sufficient for immanent purposiveness. A further question immediately suggests itself: for Thompson, is autopoiesis necessary for immanent purposiveness? To answer this question we need to dig just below the surface of the text. During his summary of Di Paolo’s (2005) argument that sense-making requires adaptivity (see above), Thompson notes that, for Di Paolo, ‘[a]daptivity needs to be established on the basis of autopoiesis; otherwise sense-making is not original to the system but merely attributed from the outside’ (p. 148). What drives this claim is the thought that, without the connection to the specific self-distinguishing process of autopoiesis, the teleological structures of sense-making would not be original to the activity of the system (that is, immanent), but would merely be attributable to the system by some external observer. The

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE & DEEP CONTINUITY

157

same dependency would presumably hold for the teleology appropriate to any bare (all-or-nothing) norm of survival based on conservation rather than adaptivity, although of course a minimally autopoietic, and thus non-adaptive, system would not be cognitive and so, for Thompson, could not count as being alive. If all this is right, then, on Di Paolo’s view, autopoiesis is not only sufficient for immanent purposiveness, it is necessary too. And given that Thompson greets Di Paolo’s argument with approval, I take that to be a view which Thompson shares. At this point it is tempting to resurrect the kind of worry that I raised earlier in connection with the claim that autopoiesis is necessary for the genesis of intentionality, and ask exactly why, in addition to the self-maintaining profile of autonomy, the dual materiality of autopoiesis is necessary to establish the immanence of the relevant teleology. One response that Thompson might conceivably pursue here would be that it is the specific kind of materiality enshrined in autopoiesis (in which a system-defining material boundary is endogenously produced and maintained) that is needed to give content to the idea (introduced earlier) that each part of a system displaying immanent purposiveness must be both a product and a producer of the other parts. That would explain why autopoiesis, and not merely autonomy, is necessary for immanent purposiveness and thus for life. So how compelling is the idea that the kind of productive interdependence at issue requires autopoietic materiality? That depends largely on how the notion of a ‘systemic part’ is to be understood. As we have seen, within Thompson’s framework, the domain of existence of an autonomous system — the ontological level at which its identity is fixed — need not be one of first-order materiality; it might, for example, be a domain of social relations rather than biochemistry. Presumably, then, there is a relevant notion of a ‘systemic part’ that is fixed at an ontological level other than first-order materiality. For example, a part of an animal social system may be fixed with respect to a level in a dominance hierarchy. But if this is right, then it is significant that Thompson’s approach already contains the idea that the constituent processes of an autonomous system ‘recursively depend on each other for their generation and their realization as a network’ (see above). For if, as seems likely, the constituent processes of a system count as systemic parts, or if, as also seems likely, the idea of recursive dependence stretches to systemic parts that ontologically speaking aren’t processes (if any such parts there be), then it seems the notion of autonomy already gives us a picture in which each systemic part is both a product and a producer of the other systemic parts. And that is a

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

158

M. WHEELER

picture of immanent purposiveness. But if autonomy is sufficient for immanent purposiveness, then even though autopoietic systems will exhibit immanent purposiveness in virtue of being autonomous systems, autopoiesis is not needed for immanent purposiveness. Whatever problems Thompson faces in carving out a distinctive role for autopoiesis in explaining the phenomenon of life, the fundamental structure of his own final position is, I think, clear enough: cognition is sense-making, while life is autopoiesis plus cognition, which is tantamount to saying that cognition is adaptive autonomy, while life is adaptive autopoiesis. However, this structural description of things, although accurate, does not do justice to the full richness of the cognitive dimension of life, as Thompson sees it. According to Thompson, living systems display the intertwined properties of needful freedom and self-transcendence. Both spring from the character of metabolism. Although, at any particular moment, an organism is realized by a finite collection of material elements, the distinctive form of the organism cannot be identified with that matter, precisely because metabolism involves a constant turnover of material elements. Life as a phenomenon thus exhibits a kind of freedom from material constitution, simultaneous with a dependence on, a need for, the material exchanges with the environment (roughly, eating and excretion) that metabolism encompasses. Self-transcendence emphasizes the fact that such metabolically grounded needful freedom may be interpreted in terms of a kind of projective organization in which organismic being involves a relentless going beyond of current state, a transcending (surpassing) of present self-identity. These properties of needful freedom and self-transcendence are explicitly introduced by Thompson as analogues of existential structures that may be revealed by a phenomenological analysis of conscious human experience. For example, and very roughly, according to Heidegger (1927/1962), I confront every concrete situation in which I find myself as a range of possibilities for the future realization of some pattern of human being, meaning that human being essentially involves a projection beyond, a freedom from, and a surpassing of, my current concrete state. In this way, then, concepts that, from a phenomenological point of view, describe universal structural features of human conscious experience also serve to locate certain fundamental properties of any living system.

MIND IN LIFE & DEEP CONTINUITY

159

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

4. Two Kinds of Deep Continuity Up to now we have been engaged in a critical exposition of the key conceptual elements out of which Thompson constructs his account of deep continuity. Our next task is to build on that exercise to ask precisely what deep continuity might amount to, for Thompson. There are no doubt many ways in which one might pursue these goals. I shall content myself with the following twofold strategy. I shall present two different interpretations of deep continuity that (the preceding analysis suggests) may be extracted from Mind in Life. Of each of these interpretations I shall ask (i) how does it conceive the enrichment process that deep continuity involves?, and (ii) is it consistent with another view that is often mentioned alongside enactivism as part of a generic embodied cognition perspective, namely the extended cognition hypothesis? Let’s pause momentarily to comment on each of these questions. Regarding (i): according to Thompson’s notion of deep continuity, the organizational and existential properties distinctive of mind are an enriched version of those fundamental to life (see above). One might be forgiven for taking this to mean that there are certain non-cognitive properties of all living entities that, when enriched in specific ways, generate the phenomena of mind and cognition, phenomena that are exhibited only by a subset of living things. I think that this must be how Godfrey-Smith (1994) understands his own, related thesis of strong continuity (again, see above). It is certainly how I have always understood the idea (see Wheeler, 1997). But, as we shall see, this seemingly natural interpretation of the enrichment process at the heart of deep continuity cannot be endorsed by Thompson. At root, this parting of the ways is driven by the fact that the alternative visions of life-mind continuity that will be placed on the table embrace more than one conception of the kind of property that gets enriched in the evolutionary and developmental paths that lead from single cells to conscious human life. Put crudely, the question is whether or not what gets enriched is already mental in character pre-enrichment, so a blunt way to raise the right issues here would be to wonder whether it is possible for life and cognition to exist separately from each other. In other words, by the lights of some particular interpretation of deep continuity, could there be non-living cognitive entities or, conversely, noncognitive living entities? That is how I shall approach question (i). Regarding (ii): according to the extended cognition hypothesis (henceforth ExC), there are actual (in this world) cases of intelligent action in which thinking and thoughts — more precisely, the material

160

M. WHEELER

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

vehicles that realize thinking and thoughts — are spatially distributed over brain, body and world, in such a way that the external (beyondthe-skin) factors concerned are rightly accorded cognitive status. In other words, the physical mechanisms of mind sometimes extend beyond the traditional boundaries of skull and skin.6 Although nothing in what I’ve just said strictly requires that the external elements in an extended cognitive system be technological artefacts (e.g. notebooks that provide external information-storage for extended memory systems, handheld computers that execute steps in extended reasoning processes), that is the standard way of filling in the picture. Notice that in asking whether or not a particular interpretation of deep continuity is consistent with ExC, I am not assuming that ExC is correct — and thus that any interpretation of deep continuity that is inconsistent with that hypothesis must be wrong — only that it is instructive, as a way of understanding the character, implications and relative merits of different interpretations of deep continuity, to determine which theoretical options those interpretations leave open and which they close off.7 The two enactive interpretations of deep continuity that I shall consider here endorse the following cluster of claims and inferences.

· Sense-making requires not only autonomy but also adaptivity;

cognition is sense-making; so cognition requires not only autonomy but also adaptivity. · Autopoiesis is sufficient for autonomy. · Immanent teleology is necessary for life; autopoiesis is necessary for immanent teleology; so autopoiesis is necessary for life. Assuming this shared theoretical backdrop, let’s first revisit the position that, I have suggested, is Thompson’s final view. This view is Clark and Chalmers (1998) provide the canonical statement and defence of ExC. For various recent developments and discussions of the view, see Menary (2010). For the dispute so far over whether enactivism is consistent with ExC, see, e.g. Clark (2008), Wheeler (2010), the contributions to Kiverstein and Clark (2009) — especially Di Paolo (2009) and Thompson and Stapleton (2009) — Rowlands (forthcoming), and Cappuccio and Wheeler (forthcoming). [7] It is sometimes suggested that since, on the enactive view, cognition is relationally defined, it makes no sense to speak of cognition being spatially located at all (see, e.g. Di Paolo, 2009; Thompson and Stapleton, 2009). To put the point bluntly, cognition isn’t extended because it isn’t anywhere! This seems to me to be an almighty red herring. There is simply no problem about relationally defined phenomena exhibiting spatiality. The thing on which my laptop is sitting right now counts as a table only because of relational factors to do with the way it enters into a context of human activity, but I have no doubt at all that that table is located in space. Or, if the mention of an object makes you suspicious, the bodily movements I’m currently making count as typing only because of relational factors to do with the way they enter into a context of human activity, but I have no doubt at all that that typing is taking place in space. [6]

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE & DEEP CONTINUITY

161

distinguished by the conjunction of two further claims — first, that life is autopoiesis plus cognition, and second, that autopoiesis is not necessary for autonomy. So, what is the nature of the enrichment process that defines the relevant sense of life-mind continuity? Put crudely, on the present view, life is what happens when autopoiesis is added to cognition, that is, when adaptive autonomy becomes autopoietic autonomy. And what that means is that, far from cognition being an enrichment of life, life is an enrichment of cognition. Assuming (as is surely natural) that cognition signals a phenomenon that we ought properly to call mental, the title of Thompson’s book now takes on its full significance. All living systems are (to use an ugly term that Thompson doesn’t) minded. So when Thompson deploys notions such as needful freedom and self-transcendence, and immanent purposiveness to characterize life, and describes those existential structures as analogous to their psychological counterparts, the analogous character of the relation does not signal any loss of (to coin an even uglier term) mindedness. The enrichment that takes place, such that ‘certain existential structures of human life are an enriched version of those constitutive of all life’ (p. 157), is an enrichment within the realm of the mental, and not an enrichment that, at some (perhaps indeterminate) point, ushers in the mental. That surely runs counter to expectations. Is this view consistent with ExC? The strict answer, I think, is yes. Cognition is sense-making, which is adaptive autonomy, but since autopoiesis is not necessary for autonomy, and adaptivity is a modulation of autonomy, there could conceivably exist adaptively autonomous, that is cognitive, systems that are not themselves alive. Since, for Thompson, life is autopoiesis plus cognition, every living system is necessarily itself a cognitive system; but not every cognitive system is necessarily itself a living system. Thus there might exist an extended adaptively autonomous system — that is, an extended cognitive system — that although not itself alive, nevertheless had living elements, such as biological cells, among its components. So far so good. Unfortunately, however, if we reflect on the teleological structures that would characterize such a system, what we uncover is an unappealing — one might even say intolerable — conceptual mess. To warm up our intuitions, let’s begin by imagining an adaptively autonomous (i.e. a sense-making) robot with no autopoietic elements. By the lights of Thompson’s account, this robot would qualify as an artificial cognitive system. However, if autopoiesis is necessary for immanent teleology, then the teleological structures applicable to this system must be extrinsic rather than immanent. Certainly nothing in

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

162

M. WHEELER

Thompson’s account prevents there from being a cognitive system characterized by extrinsic teleology. But now let’s consider our extended cognitive system. For Thompson, it seems that this system, as a result of its hybrid organic-technological nature, must realize a kind of hybrid immanent-extrinsic teleology. But it’s genuinely difficult to see how to make sense of this idea. One avenue of thought might perhaps be that the extrinsic teleology enjoyed by the external technological elements is derived from the immanent teleology realized by the organic part of the system. Interestingly, this suggestion recalls a point sometimes made in relation to representational content by opponents of ExC. The critic argues that the contents carried by putatively cognitive extracranial elements are merely derived in character, and are thus parasitic on the non-derived content carried by neural states and processes. The implication (so the argument goes) is that the real cognition remains resolutely internal. Adams and Aizawa, for example, argue that cognition always ‘involves non-derived representations, representations that mean what they do independently of other representational or intentional capacities’ (Adams and Aizawa, 2008, p. 31). If the existence of hybrid representational content can be worked up into a point against ExC, then the same might be true of the kind of hybrid teleology with which we are currently concerned. And that seems to set up a potentially disagreeable clash with the thought that cognition as enactive sense-making (adaptive autonomy) is consistent with ExC. It might seem that all is not lost. For at this point in the established content-centred debate, the advocate of ExC might well argue that, pace Adams and Aizawa, a genuinely cognitive trait may feature elements that carry only derived content, just so long as those elements are part of an integrated process in which there is a contribution from other elements (e.g. neural ones) that carry non-derived content. But even if this kind of ExC-friendly manoeuvre is plausible in relation to functionalist-based content realization in extended cognitive systems, the parallel move in relation to enactive teleology seems positively misconceived. After all, if the relevant teleological structures here are brought forth by the adaptive behaviour of the autonomous system as a whole — which is surely what the enactivist is disposed to say — the proposed strategy of partitioning, in which different kinds of teleology attach to different parts of the system, looks to be predicated on something akin to a category mistake. Of course, one might tidy up the whole mess here by adopting the view (scouted above) that autonomy is sufficient for immanent teleology, since then the extended adaptively autonomous cognitive system would realize immanent teleology and

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE & DEEP CONTINUITY

163

the problems I have highlighted would go away. But that ‘solution’ would come at a high price for Thompson, in that it would place even more pressure on the thought that autopoiesis plays an essential role in our understanding of life and mind. In light of the difficulties faced by (what I take to be) Thompson’s own considered view of deep continuity, let’s recall the anomalous position on autopoiesis and autonomy that, I argued earlier, may be found in the pages of Mind in Life, and see if an account of deep continuity based on that position fares any better. The anomalous claim was that autopoiesis is necessary, as well as sufficient, for autonomy. If this is right, then since cognition is adaptive autonomy, autopoiesis is also necessary for cognition. In effect, this new dependency would mean that cognition is adaptive autopoiesis, thus rendering it rather peculiar to claim, as Thompson does, that life is autopoiesis plus cognition, since all bona fide cognitive systems would already be autopoietic. Indeed, since it now seems that both life and cognition are a matter of adaptive autopoiesis, one might as well say that life is identical with cognition. This would mean (i) that there are no non-cognitive living entities, and thus that any process of enrichment must once again take place within the realm of the mental, and (ii) that there are no cognitive non-living entities, and thus, assuming that autopoiesis remains a property only of organic cells, that there are no hybrid organic-technological extended cognitive systems. The first conclusion flies in the face of the most natural understanding of life-mind continuity, while the second displays an inconsistency between enactivism and ExC. I have claimed that the natural understanding of deep continuity is that there are certain non-cognitive properties of living entities that, when enriched in specific ways, generate the phenomena of mind and cognition, phenomena that are exhibited only by a subset of living things. This is, as we have seen, an interpretation of the idea that is closed off to Thompson, who holds that life is, in a tangible sense, minded all the way down. One might wonder whether this tension ought to matter to Thompson. After all, our most natural ways of thinking are sometimes wrong. But it is at least arguable that the more natural understanding of deep continuity is also more plausible than Thompson’s picture. Indeed, one cannot help thinking that, however much we are told that Thompson’s deflationary notion of minimal cognition should not be thought to encompass higher-order intentional states of belief or consciousness, and however much we are warned to think of needful freedom and self-transcendence as ‘no more than’ analogues of their existential counterparts in conscious

164

M. WHEELER

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

human experience, the fact remains that if the language of cognition and its existential structures is to have any real theoretical bite in the case of the basic phenomena of life, the result is a rather unhelpful dose of something akin to panpsychism that doesn’t so much solve the problem of the genesis of mind as throw a cloak over the thought that there is a genuine problem to be solved. If this is right, then we would be well advised to look for an alternative theoretical framework for deep continuity that supports the natural reading. I shall end this paper with a brief sketch of just such an alternative.

5. A Broadly Representational Approach to Deep Continuity First let’s consider one way in which genes might be said to code for (i.e. represent in an outcome-directed manner) phenotypic traits during biological development (Wheeler and Clark, 1999; 2008). The point of departure for this account of genetic coding is that the developmental construction of phenotypic form is often to be explained by the operation of massively distributed causal systems encompassing genes, non-genetic organismic elements and/or wider environmental factors. A much-cited example is sex determination in the Mississippi alligator. These creatures lay their eggs in nests of rotting vegetation which generate heat in varying quantities. Eggs that develop at lower temperatures (within some overall range) end up producing females, whilst those that develop at higher temperatures end up producing males. Environmental temperature is thus a non-genetic factor, the inclusion of which in our explanation is necessary, if we are to account for the details of phenotypic form. Now, given this sort of distributed developmental solution, if one thought that what it means for a gene (or a complex of genes) to code for a phenotypic trait is for that gene (or complex of genes) to fully specify the form of that trait, where ‘full specification’ requires that the form of a trait may be predicted purely on the basis of what may be known about the genes in question, then the claim that genes code for traits during development would be false. In the sort of distributed developmental system with which we are concerned, knowing the entire sequence of an organism’s DNA will not be sufficient to predict phenotypic form. Fortunately, this understanding of genetic coding is far from mandatory. After all, in familiar cases of algorithms, programs, instruction-sets, and other outcome-coding elements, those states and processes are able to perform their outcome-generating functions only given some assumed backdrop of other causally active states and processes (e.g. working

MIND IN LIFE & DEEP CONTINUITY

165

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

operating systems) that themselves bear some of the responsibility for the exact form of the eventual product. Genes may play their coding role in a similar way, that is, by functioning within an assumed ecological backdrop of other developmentally crucial states and processes extended over genes, non-genetic organismic elements and/or wider environmental factors. Of course, in view of this spreading of the explanatory load, one needs to say precisely what it is that makes the genetic contribution a distinctively representational one. One proposal here (e.g. Wheeler and Clark, 1999) is that there are architectural features exhibited by the mechanisms of protein synthesis that, when taken together, mandate a broadly representational picture of the genetic contribution to development. For example, the mappings from particular nucleotide triplets of mRNA (as derived from DNA by transcription) to particular amino acids during protein synthesis are famously thought to be arbitrary, in the sense (tantamount to a kind of information-driven multiple realizability) that the class of equivalent states and processes that could play the same functional role in the developmental process in question is fixed by informational rather than first-order-physical factors. In addition, the mechanisms of protein synthesis involve a functionally distinct (although physically distributed) subsystemic module that implements the process of translation by consuming information from the upstream transcription module. The idea, then, is that where one finds an architecture with these sorts of features, a representational understanding of that architecture will be both appropriate and explanatorily powerful. There is obviously much to be said about the justification for, and the details of, this picture.8 For the present, however, what matters is that if something like it is right, then we can glimpse deep continuity in its more plausible form. Consider the way in which neural states might be said to code for (i.e. represent in an outcome-directed manner) behaviour during intelligent perception and action (see e.g. Wheeler and Clark, 1999; Wheeler, 2005). In harmony with the hypothesis of extended cognition (see above), the systems that explain perceptually guided action-generation will often be massively distributed over brain, body and world. For example, as John Haugeland [8]

Indeed, I myself have argued that once the details of the proposed account are filled in, it turns out that, strictly speaking, it is not the molecules of DNA that code in development, but rather the downstream nucleotide triplets out of which molecules of mRNA are constructed (Wheeler, 2006). Moreover, I have conceded that Godfrey-Smith (2000) may be right to think that there are problems in extending coding talk beyond proteins to phenotypic traits (Wheeler, 2006).

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

166

M. WHEELER

(1995/1998) once noted, one way to succeed in driving to San Jose would be to consult a cognitive map of the route, that is, to access a stored neural representation that specifies how to get there. An alternative method would be to select the correct road, and then follow the signs until you arrive. In the second case, the driver’s innards and the road collaborate as partners in the successful completion of the activity. Indeed, it is not merely that the environment is a cheap and up-to-date source of information (although it is that), but that any adequate characterization of the action-generation mechanism at work here would plausibly need to count the contribution of the road as being similar to that of the neural representation cited in the first solution. Thus, as Haugeland (1995/1998, p. 235) puts it, ‘much as an internal map or program, learned and stored in memory, would… have to be deemed part of an intelligent system that used it to get to San Jose, so… the road should be considered integral to [the] ability’. Where perceptually guided intelligent action displays the sort of distributed solution nicely depicted in Haugeland’s example, the neural states and processes involved will not fully specify the behavioural outcome. Nevertheless, the representational credentials of their contribution (where such credentials are warranted) may well be secured by the same kinds of features as were operative in the case of biological development (again, see, e.g. Wheeler and Clark, 1999; Wheeler, 2005). Thus neural factors may perform an outcome-directed (actionoriented) representational contribution within an assumed ecological backdrop of other behaviourally relevant states and processes extended over the brain, the body and/or the world. And that representational contribution may be illuminated precisely as such by architectural properties as arbitrariness and information-consuming modularity. The former is in truth little more than the kind of information-driven multiple realizability that underwrites broadly functionalist theorizing in psychology, while the latter, stripped of too much Fodorian baggage (Fodor, 1983), figures in a wide range of evolutionarily and developmentally sensitive regions of contemporary cognitive science, from mainstream evolutionary psychology to evolutionary robotics to developmental cognitive neuroscience and beyond (for discussion, see Wheeler and Clark, 2008). Both are wholly consistent with ExC (again, see Wheeler and Clark, 2008). The picture I have just drawn is incomplete in all sorts of ways. It is, as I have said, no more than a sketch. Perhaps the most salient worry in the current context is that it exploits a set of interconnected concepts (representation, information, function, modularity, etc.) of which Thompson is hugely suspicious and largely critical (see e.g. Chapter 7

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MIND IN LIFE & DEEP CONTINUITY

167

of Mind in Life). It is beyond the scope of the present treatment to develop any sort of response to Thompson’s concerns. So my twofold point here remains merely conditional. First, if the broadly representationalist picture I have sketched is right, then the basic structure of ecologically embedded representation that characterizes the genetic contribution to development is replicated at the cognitive level of the neural contribution to online action-generation. In my book, this is an example of deep continuity. To borrow Thompson’s words, it is a case in which life and mind share a set of basic organizational principles, and the organizational properties distinctive of mind are an enriched version of those fundamental to life. Secondly, if my broadly representationalist picture is right, then in line with the more plausible reading of deep continuity identified earlier, there is no reason to think of genes as in any way minded. The phenomenon of enrichment that the view encompasses involves a bottom-up process in which certain fundamental structures from non-minded life are evolutionarily and developmentally incorporated into the basic structures of minded life. In other words, according to the account of deep continuity I favour, enrichment takes us from life to mind, it does not operate within the domain of mind. This leads me to conclude that, at least in terms of such diachronic incorporation, life is in mind, but mind is not in life.

References Adams, F. & Aizawa, K. (2008) The Bounds of Cognition, Malden, MA & Oxford: Blackwell. Cappuccio, M. & Wheeler, M. (forthcoming) The sign of the hand: Symbolic practices and the extended mind, Versus. Clark, A. (2008) Pressing the flesh: A tension in the study of the embodied, embedded mind?, Philosophy and Phenomenological Research, 76 (1), pp. 37–59. Clark, A. & Chalmers, D. (1998) The extended mind, Analysis, 58, pp. 7–19. Di Paolo, E.A. (2005) Autopoiesis, adaptivity, teleology, agency, Phenomenology and the Cognitive Sciences, 4 (4), pp. 429–452. Di Paolo, E.A. (2009) Extended life, Topoi, 28, pp. 9–21. Fodor, J.A. (1983) The Modularity of Mind, Cambridge, MA: MIT Press. Godfrey-Smith, P. (1994) Spencer and Dewey on life and mind, in Brooks, R. & Maes, P. (eds.) Artificial Life IV: Proceedings of the Fourth International Workshop on the Synthesis and Simulation of Living Systems, Cambridge, MA: MIT Press. Godfrey-Smith, P. (2000) On the theoretical role of genetic coding, Philosophy of Science, 67, pp. 26–44. Haugeland, J. (1995/1998) Mind embodied and embedded, in Having Thought: Essays in the Metaphysics of Mind, Cambridge, MA: Harvard University Press. Heidegger, M. (1927/1962) Being and Time, Macquarrie, J. & Robinson, E. (trans.), Oxford: Blackwell. Jonas, H. (1966) The Phenomenon of Life: Toward a Philosophical Biology, Chicago, IL: University of Chicago Press.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

168

M. WHEELER

Kant, I. (1790/1952) The Critique of Judgement, Meredith, J.C. (trans.), Oxford: Oxford University Press. Kiverstein, J. & Clark, A. (eds.) (2009) Special issue on the enacted mind and the extended mind, Topoi, 28. Maturana, H.R. & Varela, F.J. (1980) Autopoiesis and Cognition: The Realization of the Living, Boston Studies in the Philosophy of Science, Vol. 43, D, Dordrecht: Reidel. Menary, R. (ed.) (2010) The Extended Mind, Cambridge, MA: MIT Press. Rowlands, M. (forthcoming) The New Science of the Mind: From Extended Mind to Embodied Phenomenology, Cambridge, MA: MIT Press. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press. Thompson, E. & Stapleton, M. (2009) Making sense of sense-making: Reflections on enactive and extended mind theories, Topoi, 28, pp. 23–30. Varela, F.J. (1979) Principles of Biological Autonomy, New York: Elsevier North Holland. Varela, F.J. (1991) Organism: A meshwork of selfless selves, in Tauber, A. (ed.) Organism and the Origin of Self, Dordrecht: Kluwer Academic Publishers. Varela, F.J., Thompson, E. & Rosch, E. (1991) The Embodied Mind: Cognitive Science and Human Experience, Cambridge, MA: MIT Press. Wheeler, M. (1997) Cognition’s coming home: The reunion of life and mind, in Husbands, P. & Harvey, I. (eds.) Proceedings of the 4th European Conference on Artifical Life, Cambridge, MA: MIT Press. Wheeler, M. (2005) Reconstructing the Cognitive World: The Next Step, Cambridge, MA: MIT Press. Wheeler, M. (2010) Minds, things, and materiality, in Renfrew, C. & Malafouris, L. (eds.) The Cognitive Life of Things: Recasting the Boundaries of the Mind, Cambridge: McDonald Institute for Archaeological Research Publications. Wheeler, M. & Clark, A. (1999) Genic representation: Reconciling content and causal complexity, British Journal for the Philosophy of Science, 50 (1), pp. 103–135. Wheeler, M. & Clark, A. (2008) Culture, embodiment and genes: Unravelling the triple helix, Philosophical Transactions of the Royal Society series B, 363, pp. 3563–3575.

Dan Zahavi

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Mutual Enlightenment and Transcendental Thought There is much of merit in Evan Thompson’s new book, Mind in Life: Biology, Phenomenology, and the Sciences of Mind. The book can be seen as a follow-up to his 1991 book The Embodied Mind: Cognitive Science and Human Experience, a book he co-authored with Francisco Varela and Eleanor Rosch. In both books, Thompson criticizes the disembodied view of cognition and consciousness favoured by computationalist and cognitivist approaches in cognitive science, and argues that the scientific study of the mind shouldn’t continue to ignore the experiential and embodied dimensions of human cognition. In Mind in Life, Thompson more specifically defends the view that the mind must be seen as an embodied dynamic system in the world, and one of his goals is to show that the so-called enactive approach can make real progress when it comes to bringing together a first-person, second-person and third-person approach to consciousness. In making this claim, Thompson is explicitly drawing on and engaging with figures from the phenomenological tradition, mainly Husserl and Merleau-Ponty. Thompson’s many references to phenomenology make it natural to compare and assess his proposal in light of the recent debate concerning the possibility of naturalizing phenomenology (cf. Varela, 1996; 1997; Gallagher, 1997; 2003; Gallagher and Zahavi, 2008; Petitot et al., 1999; Lutz and Thompson, 2003; Zahavi, 2004; 2010). In the long introduction to the milestone publication Naturalizing Phenomenology from 1999 the four co-editors, Jean Petitot, Francisco Varela, Bernard Pachoud, and Jean-Michel Roy set out to delineate what might be seen as a kind of manifesto. It is from the very start Correspondence: Email: [email protected]

Journal of Consciousness Studies, 18, No. 5–6, 2011, pp. 169–75

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

170

D. ZAHAVI

made quite clear that a successful naturalization of phenomenology will entail an integration of phenomenology into an explanatory framework where every acceptable property is made continuous with the properties admitted by natural science (cf. Roy et al., 1999, pp. 1–2). On such a reading, a naturalization of phenomenology is one that eventually makes phenomenology part of, or at least an extension of, natural science. It is consequently natural to read the proposal as basically suggesting that the transcendental philosophical clarification offered by classical phenomenology ought to be replaced with an explanatory scientific account. It is not difficult to imagine how Husserl would have reacted. In the long essay Philosophie als strenge Wissenschaft from 1911, Husserl calls naturalism a fundamentally flawed philosophy (Husserl, 1987, p. 41) and argues that it has typically had two different aims: the naturalization of ideality and normativity, and the naturalization of consciousness (ibid., p. 9). In his view, however, both attempts fail and both are misguided. As he famously put it in Ideas I, ‘When it is actually natural science that speaks, we listen gladly and as disciples. But it is not always natural science that speaks when natural scientists are speaking; and it assuredly is not when they are talking about “philosophy of Nature” and “epistemology as a natural science”’ (Husserl, 1982, p. 39). On another (more cautious) interpretation, phenomenology studies phenomena (including body-awareness, attention, intentionality, social cognition, perception and recollection) that are also open to empirical investigation, and, as has been claimed, in so far as phenomenology concerns itself with such phenomena it should also be informed by the best available scientific knowledge. On this reading, a naturalized phenomenology is simply a phenomenology that is informed by and engages in a fruitful exchange and collaboration with empirical science. The phenomenological credo ‘To the things themselves’ calls for us to let our experience guide our theories. We should pay attention to the way in which we experience reality. Empirical scientists might not pay much attention to the formal structure of phenomenality, but as empirical researchers they do in fact pay quite a lot of attention to concrete phenomena and might consequently be less apt to underestimate the richness, complexity, and variety of phenomena than the standard arm-chair philosopher. One way to appraise Thompson’s proposal is to see it as being situated somewhere in between these two options. On his view, it is not only possible but also necessary to pursue phenomenology and experimental science as mutually constraining and enlightening projects.

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MUTUAL ENLIGHTENMENT

171

For Thompson, phenomenology shouldn’t just provide a careful description and analysis of experience; it should also understand and interpret its own investigations in the light of the empirical exploration of the life of the mind. But according to Thompson, phenomenology is certainly also in a position to teach something to the sciences of mind. If our aim is to have a comprehensive understanding of the mind, then focusing narrowly on the nature of the subpersonal events that underlie experience without considering the qualities of the experience itself will just not take us very far (p. 273). In that sense, a careful description of the explanandum is an obvious requisite. More radically, however, Thompson also claims that a naturalization of phenomenology will lead to a renewed understanding of the nature of both life and mind (p. 14). Indeed, on his view, phenomenology provides a way of observing and describing natural phenomena that brings out features which would otherwise remain invisible to science; features such as selfhood, normativity, subjectivity, intentionality, and temporality. Thus, one of the decisive ambitions of Mind in Life is precisely to show how phenomenology might enable us to appreciate the inner life of biological systems (p. 358). In making claims like these, Thompson is evidently continuing and developing a view already found in Merleau-Ponty, though Thompson’s account — and that is obviously one of the fascinating characteristics of Mind in Life — is informed by the most recent advances in science. At the same time, however, Thompson also shares Husserl’s opposition to objectivism and sees his own project as one that fully respects the transcendental status of consciousness and conceives of it as a condition of possibility for the disclosure of any object (p. 86). For Husserl, objectivity is not something pre-existing, but something constituted, and Thompson explicitly argues that this fundamental insight is one that the enactive approach shares. According to the latter theory, cognition is not a faithful representation occurring between two separate and independent entities, mind and world. On the contrary, rather than being something inside the brain or inside the body, cognition is viewed as something that cuts across the divide between body, brain, and environment. Rather than seeing the world of the cognizer as an external realm that is represented (mirrored) internally in the brain, the enactive approach views it as a cognitive domain that is enacted, i.e. brought about or constituted by the coupling of the living organism with its environment (p. 154). I am quite sympathetic towards large parts of Thompson’s enterprise. However, given the ambitious scope of the book and the many

172

D. ZAHAVI

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

topics it covers, there are obviously many questions one could raise and many issues that call for further exploration and analysis. Let me just mention the two main questions I was left with after having read the book. 1. Thompson insists that phenomenology and cognitive (neuro)science stand in a relation of mutual constraint and mutual enlightenment. Phenomenological analysis can help describe the explanandum as well as contribute to a clarification of the conceptual relation between accounts of experience at the personal level and accounts of cognitive and neural processes at the subpersonal level (p. 269). Presumably, most phenomenologists would heartily embrace the view that phenomenology can be of value to cognitive science, neuroscience, and biology and, more generally, to any empirical investigation of the mind. However, Thompson evidently wants the influence to go both ways, i.e. it is also a question of letting phenomenology profit from — and be challenged by — empirical findings. Indeed, Thompson explicitly argues that neuroscientific analyses can provoke revisions and refinements of the phenomenological accounts (p. 340). That is, the idea is not merely that they can complement, but rather that they can also help us improve and refine the classical phenomenological analyses. Were this not to happen, the often repeated claim regarding mutual enlightenment would start to sound somewhat vacuous. But how can analyses pertaining to various subpersonal processes and mechanisms possibly influence and enrich phenomenological accounts that attempt to do justice to the first-person perspective and seek to understand the experience in terms of the meaning it has for the subject? Here is one simple suggestion: Let us assume that our initial phenomenological description presents us with what appears to be a simple and unified phenomenon. When studying the neural correlates of this phenomenon, we discover that two quite distinct mechanisms are involved; mechanisms that are normally correlated with distinctive experiential phenomena, say, perception and memory. This discovery might motivate us to return to our initial phenomenological description in order to see whether the phenomenon in question is indeed as simple as we thought. Perhaps a more careful analysis will reveal that it harbours a concealed complexity. (Obviously, one might also consider the reverse case, where the phenomenological analysis presents us with what appears to be two distinct phenomena and where subsequent neuroscientific findings suggest a striking overlap, unity, or even identity.) However, it is very important to emphasize that the

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

MUTUAL ENLIGHTENMENT

173

discovery of a significant complexity on the subpersonal level — to stick to this simple example — cannot by itself force us to refine or revise our phenomenological description. It can only serve as motivation for further enquiry. There is no straightforward isomorphism between the subpersonal and personal level, and ultimately the only way to justify a claim concerning a complexity on the phenomenological level is by cashing it out in experiential terms. In any case, given that the slogan regarding reciprocal constraints and enlightenment is of central concern to Thompson, one would expect him to provide a careful demonstration of how neuroscientific findings can lead to a refinement and revision of standard phenomenological analyses. But in so far as Mind in Life explicitly deals with the relation between phenomenology and (neuro)science, the focus is by and large on how phenomenological distinctions and considerations can be of use to various empirical disciplines. Very little detail concerning the reverse influence is provided. Indeed, on the occasion where Thompson undertakes to spell out what is meant by reciprocal constraints and specifically mentions the requirement that neuroscientific findings can ‘provoke revisions and refinements of the phenomenological accounts’ (p. 340), he merely adds a note stating that preliminary examples of this can be found in neurophenomenological studies of epilepsy and pain (p. 474). These results are not discussed in any further detail in the book, and I find it somewhat surprising and also slightly disappointing that Thompson doesn’t provide any examples of his own and fails to engage in a more extensive discussion of this important issue. 2. As already mentioned, Thompson wishes to preserve the transcendental status of consciousness. Consciousness is, as he writes, a condition of possibility for there being any appearance at all (p. 240). He obviously finds this stance compatible with the kind of naturalism that he endorses. What remains unclear to me, however, is how deeply committed to transcendental thought Thompson is. How far is he prepared to go? Does he for instance endorse some kind of compatibility between empirical realism and transcendental idealism? It is clear that he rejects objectivism and regards objectivity as something constituted. But when, for instance, he stresses the co-emergence of self and world, how radically is this to be understood? Thompson evidently wishes to take natural history seriously. But on his view what status should we ascribe to our theories regarding, say, geochronology and the early history of our planet prior to the emergence of biological life? There is, of course, a long debate in phenomenology — particularly in

174

D. ZAHAVI

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

the secondary literature — regarding how to appraise Husserl’s, Heidegger’s, and Merleau-Ponty’s positions on the realism-idealism question. To better understand Thompson’s view on the compatibility of naturalism and transcendental phenomenology it would have been helpful had he addressed some of these issues more explicitly. What remains clear, however, is that Thompson is committed to the idea that a naturalization of phenomenology entails a re-examination of the usual concept of naturalization and a revision of the classical dichotomy between the empirical and the transcendental. Thus, on his view, a naturalization of phenomenology not only entails a modification (rather than an abandonment) of transcendental philosophy, but also a rethinking of the concept of nature — a rethinking that might ultimately lead to a transformation of natural science itself. Regardless of how theoretically fascinating such a proposal might seem, it should be obvious that the task is daunting and that there is still a long way to go. But Thompson’s book is an important step in the right direction. It should be mandatory reading for anybody interested in the topic of naturalized phenomenology or in attaining a better understanding of how recent developments in cognitive science and philosophy of mind converge with views found in phenomenology.

References Gallagher, S. (1997) Mutual enlightenment: Recent phenomenology in cognitive science, Journal of Consciousness Studies, 4 (3), pp. 195–214. Gallagher, S. (2003) Phenomenology and experimental design, Journal of Consciousness Studies, 10 (9–10), pp. 85–99. Gallagher, S. & Zahavi, D. (2008) The Phenomenological Mind, London: Routledge. Husserl, E. (1982) Ideas Pertaining to a Pure Phenomenology and to a Phenomenological Philosophy: First Book — General Introduction to a Pure Phenomenology, Kersten, F. (trans.), The Hague: Martinus Nijhoff. Husserl, E. (1987) Aufsätze und Vorträge 1911–1921: Husserliana XXV, Nenon, T. & Sepp, H.R. (eds.), Dordrecht: Martinus Nijhoff. Lutz, A. & Thompson, E. (2003) Neurophenomenology: Integrating subjective experience and brain dynamics in the neuroscience of consciousness, Journal of Consciousness Studies, 10 (9–10), pp. 31–52. Petitot, J., Varlea, F.J., Pachoud, B. & Roy, J.-M. (eds.) (1999) Naturalizing Phenomenology, Stanford, CA: Stanford University Press. Roy, J.-M., Petitot, J., Pachoud, B. & Varela, F.J. (1999) Beyond the gap: An introduction to naturalizing phenomenology, in Petitot, J., Varela, F.J., Pachoud, B. & Roy, J.-M. (eds.) Naturalizing Phenomenology, pp. 1–83, Stanford, CA: Stanford University Press. Thompson, E. (2007) Mind in Life: Biology, Phenomenology, and the Sciences of Mind, Cambridge, MA: Harvard University Press.

MUTUAL ENLIGHTENMENT

175

Copyright (c) Imprint Academic 2013 For personal use only -- not for reproduction

Varela, F.J. (1996) Neurophenomenology: A methodological remedy to the hard problem, Journal of Consciousness Studies, 3 (4), pp. 330–350. Varela, F.J. (1997) The naturalization of phenomenology as the transcendence of nature: Searching for generative mutual constraints, Alter, 5, pp. 355–381. Varela, F.J., Thompson, E. & Rosch, E. (1991) The Embodied Mind: Cognitive Science and Human Experience, Cambridge, MA: MIT Press. Zahavi, D. (2004) Phenomenology and the project of naturalization, Phenomenology and the Cognitive Sciences, 3 (4), pp. 331–347. Zahavi, D. (2010) Naturalized phenomenology, in Gallagher, S. & Schmicking, D. (eds.) Handbook of Phenomenology and Cognitive Science, pp. 2–19, Dordrecht: Springer.