In Being No One, Thomas Metzinger attempts to develop a thorough and systematic approach to consciousness and subjectivity – one based fundamentally on a teleofunctionalist and naturalist view of consciousness. In other words, an approach to consciousness that presents it as necessarily supervening on a neural basis and as subject-to and a product-of evolutionary pressures. Perhaps surprisingly, in many ways Metzinger’s project comes across as a neurological updating of Kant’s project. Instead of the ‘conditions’ of possible experience, Metzinger deduces the ‘constraints’ constitutive of phenomenal consciousness and self-consciousness. Unlike Kant, however, Metzinger refuses to describe consciousness in a single, unitary way. With the advances in psychology, neuroscience and phenomenology, we can now achieve a much more detailed and nuanced view of consciousness. One of the most significant aspects of Metzinger’s work is therefore the integration of deficient forms of phenomenal consciousness – ones where a particular constraint may be absent as the result of neurological damage. The end result is a set of necessary and contingent constraints for consciousness to occur. The addition or subtraction of a particular constraint leads to entirely different forms of consciousness.
To give an example of this, here is the list of 10 constraints for full-fledged everyday consciousness to appear (note, this is not yet self-consciousness – a phenomenon which requires similar constraints, but that I hope to cover later):
(1) Global availability
(2) Activation within a window of presence
(3) Integration into a coherent global state
(4) Convolved holism
(5) Dynamicity
(6) Perspectivalness
(7) Transparency
(8) Offline activation
(9) Representation of intensities
(10) The homogeneity of simple content
Using global availability as an example, we can begin to also draw some other important analytic distinctions. This first constraint – global availability – refers to information being available to the neurophenomenal system on 3 separate levels: attentional, behavioural, and cognitive. Straightforwardly enough, on a phenomenological level this means, respectively, that we can focus our attention on an object, that we can generate motor responses to an object, and that we can conceive of an object. Just as each constraint may be missing, so too can each of these 3 levels be missing. The most obvious example that Metzinger gives is of the extremely subtle shades of color – we can attend to these, and we can behaviourally use this information, but we cannot conceive of these and put them into concepts. There is an intrinsic ineffability to the subtle nuances of colors that restricts our ability to identify them. For example, we can attentionally distinguish between Red-41 and Red-42 when they are both placed before us, but we cannot identify Red-41 as Red-41 on its own. We have no concept capable of identifying this patch of color as Red-41 and not Red-42 before us. In this case then, the cognitive aspect of global availability is absent.
A more extreme case, and a philosophically even more interesting case, occurs in blindsight. Individuals with this condition are unable to consciously perceive a specific portion of their world. They are missing all three aspects of global availability. Surprisingly, however, in experiments, these individuals reveal that they still have some relation to this information – they can even successfully ‘guess’ whether a wavelength presented to them is red or blue. In other words, there is a subconscious level of neural processing that is still capable of using this information from the senses, but this information is never made globally available. This, therefore, gives an example of how a missing constraint affects the neurophenomenal system as a whole. The phenomenal aspects entirely disappear, but the information is still functioning at a different level.
Already, we can begin to sense some of the important ways in which Metzinger differs from Kant. We’ve already mentioned that unlike Kant’s single characterization of experience (the categories and forms of intuition, plus sensible content), Metzinger provides a much more nuanced view of phenomenal experience. Secondly, unlike Kant, Metzinger refuses to make the transcendental subject the ultimate foundation of these phenomenal conditions. Instead, following the success of neuroscience, it is the brain (and evolution) that must ultimately produce the conditions underlying our experience. This entails that at each step, Metzinger provide some functional explanation of the constraint – what it adds to the organism that helps in its survival.
For example, we can look at the third constraint – integration into a coherent global state. This constraint entails that our phenomenal experience is not only a combination of various events, but is also holistic. Unlike simple unity, this experiental holism means that it is not a mere set of independent events, but rather a set of parts integrated into a whole. They are all coherently related together. What this enables the organism to do is to establish what Metzinger will call a ‘world-zero’. This refers to the untranscendable experience of being situated into a coherent and stable world. In other words, it is the source of naive realism – the simple belief that there is a stable world (although not necessarily that ‘we’ are in a world, yet). Functionally, the evolutionary value stems from the fact that this gives the organism a reference point for any hypotheses, planning or simulation of possible actions. With a reference point in world-zero, the organism can now begin to compare models of actions and try to discern the best plan. It is, clearly, a key evolutionary step.
Like Kant, however, Metzinger falls into a significant problem – one that has gone unanswered so far (2/3 of the way through the book). As Kant was unable to provide knowledge of the noumenal world, so too Metzinger ultimately seems unable to ground his neurological knowledge in the nature of reality. As Metzinger points out numerous times, there is no guarantee that our conscious experiences and the cognitive knowledge built upon them are accurate representations of reality-in-itself. He frequently points out that phenomenal immediacy and phenomenal certainty are not equivalent to epistemic immediacy and certainty. What may be given to us a immediately and intuitively present has no necessary relation to what is actually causing our experience. If our neurological structures produce conscious experience in such a way, the question becomes: what is the status of knowledge derived from these constructed experiences?
We can presume that this knowledge is at least functionally adequate, insofar as our evolutionary success would seem impossible otherwise. But this in and of itself does not constitute knowledge in a rigorous sense. In a brief section on the formation of the temporal present, Metzinger goes so far as to point out that the simultaneity of our experience – the fact of our phenomenal experience occurring within a bounded temporal moment – need not be replicated in the world. In other words, despite our intuitions about there being a single, temporally linear world (an intuition premised upon the dynamism of experience), there is no necessary correlate of this representation outside of consciousness. If this is the case, however, then it’s quite possible to question the epistemic status of neuroscience, since it is a body of knowledge subject to the same constraints outlined by Metzinger. He does argue that consciousness must necessarily be embodied – and so we have reason for some form of realism – but it is also possible that a brain in a vat would give the same phenomenal experience as a fully embodied and active organism. We have knowledge of the necessity of a physical embodiment, but we have no knowledge of what this realm actually consists of.
There seems to be to be a few possible answers to this problem. Perhaps the most common one is to see science as a form of pragmatism that simply makes no claims to absolute truth and instead develops open-ended theories that are then tested against empirical reality. A sort of reciprocal conditioning occurs between praxis and theory. The problem is that this seems to make scientific truth merely a rigorous form of intersubjective agreement, thereby sidestepping science’s realist implications. Another response is to argue that evolution has given us the means to establish accurate knowledge – a position suggested by the apparent evolutionary success of our knowledge. However, Ray Brassier has convincingly argued against this in Nihil Unbound, showing that this relies on untenable metaphysical presuppositions. Perhaps, though, this problem is what leads Brassier to Laruelle insofar as Laruelle’s non-philosophy offers a form of philosophy (or knowledge) that refuses to take itself as self-sufficient. Instead, radical immanence is determining-in-the-last-instance of all attempts to conceptualize the emergence of cognition. Knowledge, even neuroscientific knowledge, cannot logically present itself as being metaphysically self-sufficient – it requires this unobjectifiable immanence as its presupposition. Perhaps…
Speculative Heresy 
Pingback: i promise… « Daily Humiliation
Thanks for posting this, very interesting stuff. I’ve yet to get my hands on ‘Being No One’, so am really only familiar with Metzinger’s project as presented in his shorter pieces and in the work of others. I wonder though, what distinction does he draw between his notion that no such thing as a ‘self’ as ever existed, and the possibility of the subject? Clearly his work has huge implications for thinking subjectivity, but in the work of his I’ve encounter so far he never seems to deal with it explicitly, does he do this in Being No One?
Look forward to hearing more of your thoughts on this work.
To repeat what Ezra Pound scribbled on T.S. Eliot’s draft of “The Love Song of J. Alfred Prufrock:” ‘Perhaps be damned!’
But seriously, this is a great account of a work I am unfamiliar with, and I truly appreciate how you present it. It makes me want to read it now! Great stuff.
I also think that this resonance with Kant makes Laruelle especially relevant…I was thinking that another text, maybe even more relevant than the one on Heidegger, would be Laruelle’s “La methode transcendentale.” He gives a very clear and nuanced account of the transcendental and the a priori–in a sense, would that flow with the theme of anti-correlationism that you referenced?
Thanks Taylor, I struggled over how to present it, but this seemed like a pretty good way to introduce people to the work. It’s a great book, and absolutely packed full of interesting insights. I haven’t even scratched the surface of the ideas here, but I wanted to at least give an introduction. I think, as far as the systematic structure of the work goes, the Kantian parallel holds up nicely (with the caveats I noted). But really there’s a lot of great stuff in the details.
Michael, he does tackle that issue head on in this work. It’s the most complex part of the work, so I don’t want to claim to be infallible on these points. But to try and put it simply, since there’s no such thing as a self or subject, what there is, is the phenomenal appearance of a self. At its most basic level, this occurs when the system (a representational system supervening on neural correlates and integrated with the entire organism) uses its own phenomena producing capacities on itself, to represent a part of itself (but never its entirety). The result is that the relation between the system and its self-consciousness is not reflexive, but is instead mereological. Self-representation is a part of the system, and not a reflexive relation of the whole system to itself. Another important consequence is that as a phenomenon, subjectivity is subject to all the same constraints as any other phenomenal appearance. One of the key constraints will be the transparency constraint. What this entails is that the processes producing a particular representation become transparent – we see through them like a window and only see the outcome. Moreover, it also entails that we aren’t even aware of having this gap in our phenomenal experience. In other words, not only do we have no sense of the neurological or representational processes that produce a particular representation, we don’t even know that we are missing these. In the case of self-representation, this is what leads to the illusion of an independent, immediately given self.
That’s a very schematic overview, but I think it gets across some of the key points. I haven’t even touched on the ‘phenomenal self-model’ or the ‘phenomenal model of the intentionality relation’, but those also contribute a lot to answering your question. I’m still hoping to write on his concept of agency, so I hope to get into these issues in some more depth soon.
I agree with the other commenters that your summary is excellent, and with you that the necessary brevity makes it impossible really to delve into the substance. The ten constraints, for example, conjure associations for me that may bear little resemblance to what Metzinger has in mind. Your example of global availability is very helpful, and to me it illustrates the difficulty of distinguishing consciousness from other forms of adaptability. Many kinds of creatures can attend and respond to objects in the environment, but presumably only humans can form a conception of that object. So now it seems Metzinger would have to contend that the ability to form concepts is that which distinguishes conscious from non-conscious global availability. Or does he position the three aspects of global availability on a continuum: some creatures have more nuanced attentional abilities, others form more readily communicable concepts, etc.?
As you point out, moving from phenomenology to epistemology remains a problem even for neurological descriptions of cognition, but it seems that this problem arises in particular when thinking about conceptualization. Attention and response can be inferred from observable effects, which is why the behaviorists focused on them while intentionally ignoring the inner world of concept manipulation. Even without functional MRI scans, empirical associations can be identified between attention, response and conceptualization, such that in conjunction these three elements verify each other in a theoretical model of cognition. It’s good enough for cognitive scientists, which is the trajectory from which I look at this sort of work. But it doesn’t escape the Correlation; it doesn’t offer a glimpse of conceptual realism, of the reality of thought without thinkers and language-users on which continental philosophers insist. What would the arche-fossil of a concept look like I wonder?
Yeah, the 10 constraints are really suggestive as they’re presented here, but I’d caution against drawing any firm conclusions. Metzinger uses them in a perfectly understandable way, but they’re also highly specific.
Metzinger does draw a ‘minimal’ concept of consciousness; i.e. the minimal constraints necessary for it to arise. For him, it involves presentationality (the temporal immediacy of an appearance), globality, and transparency. The combination of these 3 constraints is sufficient to produce an appearance – the “presence of a world” as he says (204).
There is a continuum on different aspects of the attentional/behavioral/cognitive distinction – being the products of the senses and brain structure, they’re subject to all their variations. As far as drawing a distinction between man and animal, Metzinger brings it up at a few points, but never draws a sharp line. He notes that many of the constraints are apparently shared with some animals. The unique factor of humans may partly be the fact that we (typically) contain all the constraints. Personally, I found his analysis of animals to be really interesting, as a way to conceive of how their “mind” operates, and all their limitations. But it’s unfortunately relatively sparse.
The realism of psychology is interesting too. It seems to very easily fall into paradox if it’s not defined precisely. The brain is the standard answer as to what the “thought-independent basis of thought” could be, but it seems to beg its own questions. The typical correlationist response that I raise above, is that our knowledge of the brain is dependent on our experience of it, which only gives us phenomenal knowledge and not noumenal knowledge. Metzinger’s mantra throughout the book is ‘phenomenal certainty is not epistemic certainty’, which casts doubt on the absolute validity of any sort of psychological knowledge (including introspection!).
But when we look at Meillassoux’s work, it’s not really the current scientific theories that he’s trying to uphold; rather, it’s science as an enterprise – science as a knowledge of the absolute – that he tries to defend. So in that sense, we could say that neuroscience points to the arche-fossil of thought (i.e. the evolutionarily produced brain), without simultaneously declaring that it’s absolute knowledge. The big gap in this idea, and in Meillassoux’s work, is the connection between a world outside the correlationist circle and specific scientific statements. It’s not clear yet what their relation is or could be.
The alternative is that we take other speculative realists’ positions, where it’s already possible to think the world of objects. I think this is what Brassier does when he turns to Laruelle, but Graham Harman also provides an analysis of the reality of objects-in-themselves. Harman’s work entails that any particular appearance of an object (e.g. the mind, whether it appears through MRIs or introspection or behavioral observation) always includes an aspect that withdraws – placing limits on science’s expansion – but it does get past some of the conceptual problems of a realist psychology.
Hi Nick,
Just happened on this post (three years down the line). I’m only halfwway through BNO, but I agree with the other posts that it’s an excellent and very helpful precis.
I’m not so sure, though, that Metzinger should be required to include an account of the nature of reality within his model of consciousness. As you point out, the world-zero postulate gives us a functional account of the property of being the world ‘for-me’. This obviously doesn’t provide a metaphysical account reality as such. As you say, it presupposes it since it is true only if the theory which entails it is true. But this is to say that the truth of Metzinger’s position presupposes that models in cognitive science it employs – like state space semantics – are true, or least approximately true.
This just suggests that Metzinger’s only metaphysical aberration is to assume Scientific Realism. One could urge that Scientific Realism needs to be backed up by some purely metaphysical account – transcendental realism, say – but Metzinger’s theory of consciousness shouldn’t be required to provide that and it is not clear how it could in any case. I’m pretty certain that Harman’s darkling metaphysics doesn’t get us there either. I’m not sure about Brassier’s commentary on Laruelle – that thrust of that chapter completely eluded me on first reading – maybe I should give it another go.
Some theories of consciousness or subjectivity do attempt to modalize transcendent reality in terms of its relation or lack of it to something immanent. For example, transcendental phenomenology gives an account of reality in terms of the transcendence of objects but this theory only works if conscious lived experience) is ‘immanent’.
But auto-epistemic closure entails that conscious experience has no epistemic privileges – epistemically it’s on a par with the core of the sun. Lived temporality, for example, is subjectivly no more and no less accessible than the structure of the physical world if Metzinger is right -and I’ve argued similarly.
So Metzinger’s position implies that phenomenology is a legitimate undertaking but not one with a special pertinence to metaphysics. If that’s right the, for sure, we need a metaphysics of realism but it’s far from clear why we should look to Metzinger’s naturalistic theory of consciousness to provide it.
Best,
David
Pingback: Metzinger and Realism