I’m ending an all too brief sojourn in Western Crete, just as Greece seems set to become Europe’s new experiment in post-democratic capitalism – its very own Interzone. Many, if not most, economists claim that the conditions cannot be met and that attempting to do so will shred Greece’s economic, social, educational and cultural life as much as the initial round of austerity.
Nonetheless, a bubble of ease is maintained here for those with euros. We who bask in the light and heat of the Aegean summer can condemn the deprivations heaped upon the Greek state and its citizens without having to experience them.
However factitious, this moment has allowed me to pause and think about some generous philosophical discussion of Posthuman Life on a number of excellent websites. These have forced me to think harder about the basic assumptions in of the book. So here begins a series of reflective responses to my commentators under the rubric of “Dark Posthumanism” – though, as shall become clear, my use of the d-word is seriously tendentious.
I should begin by citing Debbie Goldgaber’s excellent post on Speculative Posthumanism and dark phenomenology. This catalyzed an exchange between deflationary naturalists like Scott Bakker and those like Jon Cogburn or Goldgaber, who favour a deconstructive or “weird realist” construal of dark phenomena. This debate resurfaced during a lively discussion at the New Centre for Research and Practice‘s Posthuman Life 1 seminar, in which Debbie also participated. Its trenchancy was a surprise, although a welcome and productive one, which I’ll try to address in this post.
Meanwhile, the Philosophical Percolations Summer Reading group on PHL rolls on to Chapter 2 and 3 and the Ultima Thule of Unbounded Posthumanism! I should also bow to John Danaher’s fine clarificatory effort over at Philosophical Disquisitions. He has not yet addressed the role of dark phenomenology, but it will be interesting to see what he makes of it.
Scott’s interview with me over at Figure/Bound communications recapitulates similar tensions while holding me to account for the ethical commitments of the book. I think there’s a connection between the epistemological issues arising from the dark phenomenology hypothesis and the ethics and politics of becoming posthuman. These are taken up in B P Morton’s terrific piece on trans/posthumanism and transgender (also at philpercs) which I to return to in the sequel to this post.
So what’s the deal with Dark Phenomena?
On a first (and extremely shaky) approximation, there is a tension between a thin epistemological interpretation of Dark Phenomena – experiences that furnish no tacit yardstick for their description – and a weird reading that I hesitate to term “ontological”, since its presuppositions seem more difficult to articulate than the naturalist side.
On the epistemological reading, the dark side is a placeholder for structures of experience that phenomenology cannot elucidate without the help of science – in particular, psychology, neuroscience or cognitive science. Dark phenomena reveal the point at which the putative domain of phenomenology eludes the scrutiny of philosophical method. It does not imply any obscurity in principle, since what may elude phenomenology may be explicated in other terms.
On the weird (horror?) reading, the dark side must be understood via its disintegration or truncation of the subject: experiences of horror, alienation, humour or compulsion such as the spectral thing that, for Levinas, depersonalises the consciousness of the insomniac. As Cogburn points out, these incursions and eruptions in experience can be related to the late Idealist view that our experience of embodiment provides privileged insight into a pre-subjective Nature (Schelling) or a noumenal body that eludes representation. I think Eugene Thacker’s discussion of Schopenhauer in his book Starry Speculative Corpse captures the latter idea particularly well:
The Will is, in Schopenhauer’s hands, that which is common to subject and object, but not reducible to either. This will is never present in itself, either as subjective experience or as objective knowledge; it necessarily remains a negative manifestation. Indeed, Schopenhauer will press this further, suggesting that “the whole body is nothing but objectified will, i.e. will that has become representation” (122-3)
So darkness on the naturalist reading is a local problem for phenomenological method, whereas on the weird reading it is an obscure disclosure (“negative manifestation”) of something (some thing) that resists any form of representation or theory. It must also be contentless if it is to do the work of undercutting the claims of transcendental conceptions of the subject, whether phenomenological, existential or pragmatist.
So far this seems as if it might be almost consonant with Bakker’s take on dark phenomenology. As he writes in his commentary on Goldgaber, phenomenology qua method:
assumes we have a reflectively accessible experiential plenum to begin with, that we actually possess a ‘phenomenology’ worth the name. The problem, in other words, is that we have no way of knowing just how impoverished our ‘phenomenology’ is in the first place.
If phenomenology is dark then phenomenological method is at best incomplete and at worst benighted. For example, experienced temporality is as transcendent and inaccessible to us as the structure of matter. Phenomenology can never be more than a descriptive science of nature according to this account and should not aspire to a priori status since there is no good reason to think that its descriptions are authoritative. There are good empirical reasons for thinking that we take our judgements about the contents of our minds or experiences to be based on an unmediated givenness only because we are not mindful of the heavy lifting required to produce them. If phenomenology is dark we are, as Bakker implies, in the dark about the dark.
The weird reading might now seem a little shady. Even the metaphor of darkness is misleading if it implies a phenomenology of the “gaps in presence”. This would be feasible only if we already knew the structure of the plenum and (or so the argument goes) there is no good reason to think that we do.
This seems to warrant a cautious analogy between the thesis that there is a dark side to phenomenology and Derridean deconstruction, which, though drawing on the language of phenomenology, cuts it free of any secure domain by generalizing subjective temporality well beyond anything conceivable as a subject, to the iterable mark, to generalized writing etc. (PHL: 94).
Goldgaber imputes to me the claim that this structure, at least, is generalizable beyond the human:
were it possible to show that there are dark elements in our own phenomenology, experienceable but not amenable to description or interpretation, we would have grounds, Roden thinks, for understanding human subjectivity in terms of both its unity and radical difference or rupture from world–as dependent on structures that are shared by nonhumans.
I’m not sure that I go this far. I suspect a purer Derridean like Martin Haggelund might. But, like Bakker, I don’t see any reason to see why such claims are on securer ground. Their virtue is salutary rather than informative; exposing the indeterminacy of claims about structure of worldly agency and time.
On the other hand, once we take dark phenomenology (or Bakker’s blind brain theory) as serious epistemological proposals we seem confronted with a darkness without negation, not one contrary to the light side (which, by hypothesis, is already striated with it). And here one is almost tempted to say that harder-than-hard naturalism bites the tail of mysticism. In Speculative Corpse, Thacker distinguishes a metaphysical correlation (between thought and object) presupposed by philosophy from a mystical correlation that can only verify itself by breaking against an impersonal “divine” darkness (84-5) that can never be recuperated by thought. A similar failure of correlation seems to obtain here. Even the tools (concepts like plenum) with which we are attempting to think the absence of a proper topic for phenomenology have to fail us. A thought that reiterates its failure in this way obeys the logic of the mystical as Thacker describes it.
So while we may not have any knowledge of what we could share with unboundedly weird posthumans, or nonhumans of other stripes, we led into a defile that is boundless on either reading. Perhaps the deflationary reading is as weird as it gets. Perhaps as Bakker puts in Neuropath, we are all already “vast and terrible with complexity” . As the tagline to the novel states: you do not know what you are. You do not know what it is that does not know this. We do not know where the darkness ends, how far it extends. And perhaps it is this pervasive boundlessness that can provide a tentative opening beyond the human, freeing us, as Morton might say, to explore the near inhuman, the trans of alterable bodies and desires.
Or maybe this is too quick! It’s easy to make imaginary progress in a frictionless milieu. I’ll return to Morton in Dark Posthumanism II.
Anti-reductionist physicalists or materialists deny that psychology can be theoretically reduced to physics but allow physics sovereignty concerning what exists. Anti-reductionist arguments vary but a common line of attack against reductionism is that psychology expresses rational or normative relationships between mental states; not causal or functional relationships of the kind expressed in theories of natural science. Thus in Sellars “Two Images” account physics and natural science tells us what exists but humans still encounter themselves in a normatively structured “space of reasons”. Donald Davidson refers to his own version of this position as “anomalous monism” (AM):
“Anomalous monism resembles materialism in its claim that all events are physical, but rejects the thesis, usually considered essential to materialism, that mental phenomena can be given purely physical explanations. Anomalous monism shows an ontological bias only in that it allows the possibility that not all events are mental, while insisting that all events are physical” (Davidson 2001: 214)
Davidson’s account seeks to reconcile three claims that appear to be in tension: 1) that mental events causally interact with physical events; 2) that causal relations occur only where the events in question are covered by strict deterministic laws; 3) “that there are no strict deterministic laws on the basis of which mental events can be predicted and explained (the Anomalism of the Mental).”
Davidson aims to do this by arguing from the claim that the existence of causal relationships between events only implies that there is some true description of the relationship expressing a strict nomic relationship. The reconciliation is possible because causal relations obtain between token singular events while laws are linguistically expressed generalisations. Mental events can be causally related to one other or to non-mental events.
But, according to Davidson, causality is nomological only in that where two events are causally related, they have linguistic descriptions that express a law. It does not follow that “that every true singular statement of causality instantiates a law” (215). Thus a statement like “Helen’s belief that Justin was murdered was caused by her seeing blood in the kitchen” adverts to a law like relationship between a token of blood in the kitchen and a token belief about murder but does not state it. The law-like relationship, for Davidson, would have to be expressed in terms of the states and dynamics of a physical system which allowed a deterministic inference about a future state – her belief token – again rendered in some physicalistic idiom.
Claim 3) Follows, Davidson thinks, if mental states are those addressed in propositional attitude ascriptions and that such ascriptions depend holistically on overall assessments of the rationality and cognizance of agents in their world. In the space of reasons, where propositional attitudes are ascribed to persons, it is always possible to revise attributions in the interests of overall cogency. There can be no single translation scheme that pre-empts all the evidence that could be relevant to such ascriptions (222-223). Thus whereas the theories in which physical regularities are stated must be closed to allow the formulation of exceptionless laws (homonomic) the language of propositional attitude ascription is necessarily open to multiple idioms or “heteronomic” (219):
“The heteronomic character of general statements linking the mental and the physical traces back to this central role of translation in the description of all prepositional attitudes, and to the indeterminacy of translation. There are no strict psychophysical laws because of the disparate commitments of the mental and physical schemes. It is a feature of physical reality that physical change can be explained by laws that connect it with other changes and conditions physically described. It is a feature of the mental that the attribution of mental phenomena must be responsible to the background of reasons, beliefs, and intentions of the individual.”(222)
In Nagelian terms, it would be impossible to formulate true bridge laws between a reducing theory in some physical idiom and a reduced psychological theory because the intentional side the biconditional could always be revised in the light of holistic considerations irrelevant to the “physical side”. Thus type-type psychophysical reduction appears impossible. Note that an analogous result is obtainable if we view the space of reasons as structured by implicit norms irreducible to behavioral regularities.
Of course, not all accounts of reduction require bridge laws between reduced and reducing theories, or treat theories as interpreted sets of sentences. It is still open to the reductionist to argue for a different form of reduction (Bickle 1993: 222-4). It is also open to the reductionist to argue that psychology is not peculiar in being inexpressible “as sets of generalizations” – this being true of all scientific theories (226) – or in being open to extra-theoretical idioms in which to describe their contexts of application to real systems. Maybe no theory (physical or otherwise) is truly heteronomic.
However, in the argument that follows I will suppose that Davidson’s anomalism is right, or, at least, that his account can be rectified in a form that is proof against neoreductionist assaults.
So let us assume that the psychological perspective in which agents have beliefs and desires and utter meaningful statements is conceptually irreducible (as Sellarsians say) to the scientific image of the world as a causal-physical system.
If so, then the possibility of a certain form of technological descendant of current humans (posthumans) implies that intentional psychology will be instrumentally if not theoretically eliminated.
That is, whatever its current value for humans, it could not play a similar role for the relevant class of posthuman. And this not because of any logical or ontological vices but because of it would be incapable of functioning as an idiom for interpretation and understanding among these hypothetical successors. So the anti-reductionist argument against theoretical reduction/elimination supports a metaphysical case for instrumental elimination.
The hypothetical entities in question are what I refer to in Posthuman Life and elsewhere as “hyperplastic agents”. An agent is hyperplastic if it can make arbitrarily fine changes to any part of its functional or physical structure without compromising either its agency or its capacity for hyperplasticity. For example, suppose a hyperplastic agent dislikes some unpleasant memories associated with the taste of milk. Whereas a merely plastic agent like ourselves might need hours of cognitive behavioral therapy to excise these, the hyperplastic simply needs to locate the neuronal ensembles and pathways associated with these memories and ensure that they are no longer linked in such a way that the memory of milk causes them to activate in turn.
Likewise, a hyperplastic would be in a position to alter any other informational or value-relevant state by physically altering the relevant brain states. Obviously, use the term “brain” broadly here to refer to those systems within the hyperplastic that are associated with “cognition”, “perception” or the “control of behaviour” in some intuitive sense of these terms. The original inspiration for the idea of the hyperplastic came from Steve Omohundro’s speculations about the goal structures of generally intelligent robots in his essay “The Basic AI Drives” (2008). We need not assume that the “brain” in question is a known biological system.
Davidson’s anti-reductionism implies token physicalism (each event that can be brought under a psychological description is identical to some physical event, since ontological physicalism is taken as a given).
So for any state in an agent with a psychological description there will be physical description of that state. For any such state there will interventions that the agent can make into the state which will produce a physically distinct successor state such that the former psychological description will no longer be true of it.
Now we can suppose that any hyperplastic agent will have an Agenda at a particular time. That is, it will not tinker with its internal states arbitrarily but wish to do so in ways that don’t kill it, do not undermine its capacity for hyperplasticity and that fulfill whatever desiderata are listed on the Agenda.
The interesting question (assuming Davidsonian anti-reductionism) is how the Agenda can be formulated. Can it be expressed in psychological terms (roughly, in terms of propositional attitudes or values)? If it is expressed in psychological terms, then anti-reductionism implies that for any Agent intervention at the physical level, it will not be possible to reliably infer the psychological outcome of the alteration.
This follows simply because there are no psychophysical laws. Moreover even rough generalisations over past interventions would not be much help. These might be reliable for merely plastic creatures whose basic design and structure remain fairly constant over time. But a hyperplastic agent is protean. Thus it cannot assume that the rough and ready psychophysical generalisations that have held over one phase of its existence will extend into another phase.
It follows that however a hyperplastic agent frames the Agenda it cannot be psychologically expressible because no reliable inferences can be drawn from future physical form to future psychology.
So if hyperplastics have Agenda’s, they would have to represent states that could be reliably inferred from facts about their physical constitution at a given time. But given Davidson’s anti-reductionism, they would have little use for psychological self-description for making generalisations about their current or future actions. Suppose a hyperplastic Agent self-attributes a belief b. A merely plastic agent like you or me might assume generalisations along the lines of “I will continue to hold b unless I find evidence from which some contrary of b can be inferred”. But a hyperplastic agent would not be able to assume such generalisations because there could be no evidence that an auto-intervention would not cause it to lose b regardless of the evidence in its favour.
So a hyperplastic agent could not use propositional attitude psychology to predict its own behaviour. Folk psychology would be equally impotent for predicting the behaviour of its fellow hyperplastics for the same reason.
If hyperplastic agents could exist and plan their self-interventions, they would have to employ an entirely different idiom to understand themselves or one another. A posthuman-making disconnection that resulted in the emergence of hyperplastics would inevitably to result in the instrumental elimination of folk psychological capacities among the population of hyperplastics, at least; since neither the capacity nor the linguistic idiom for attributing propositional attitudes would have predictive or hermeneutic utility.
This means that were humans to encounter hyperplastics, they would not be radically interpretable (in Davidson’s sense) because radical interpretation depends on the principle of charity and this, again, is framed in folk psychological terms.
I conclude that if hyperplastic agents are possible, we could not understand them without abandoning the conceptual framework we currently use to understand ourselves and our conspecifics. They would be radically uninterpretable.
Bickle, John (1992). Mental anomaly and the new mind-brain reductionism. Philosophy of Science 59 (2):217-30.
Davidson, D. 1984. Inquiries into Truth and Interpretation. Oxford: Clarendon Press.
Davidson, Donald (2001). Essays on Actions and Events, Vol. 1. Oxford: Oxford University Press.
Omohundro, S. M. (2008). “The Basic AI Drives”. Frontiers in Artificial Intelligence and Applications 171: 483
In this excellent presentation Saxe claims that Transcranial Magnetic Simulation applied to the temporo-parietal junction (TPJ) – a region specialized for mentalizing in human adults – can improve the effectiveness of moral reasoning by improving our capacity to understand other human minds.
This suggests an interesting conundrum for moral philosophers working in the Kantian tradition, where recognizing the rationality and personhood of offenders is held to be a sine qua non for justifications of punishment. We can imagine a Philip K Dick style world in which miscreants are equipped with surgically implanted TMS devices which zap them where an automated surveillance system judges them to be in a morally tricky situation calling for rapid and reliable judgements about others’ mental states. Assuming that such devices would be effective, would this still constitute a violation of the offender’s personhood – treating the offender as a refractory animal who must be conditioned to behave in conformity with societal norms, like Alex in a Clockwork Orange ? Or would the enhancement give that status its due by helping the offender become a better deliberator ?
Assuming the TMS devices could achieve their aim of improving moral cognition, it seems odd to say that this would be a case of “tiger training” which bypasses the offender’s capacity for moral reasoning since it would presumably increase that very capacity. It is even conceivable that an effective moral enhancement could be co-opted by savvy Lex Luthor types to enhance the criminal capacities of their roughnecks, making them more effective at manipulating others and sizing up complex situations. At the same time, it would be quite different from punishment practices that appeal to the rational capacities of the offender. Having one’s TPJ zapped is not the same as being asked to understand the POV of your victim – though it might enhance your ability to do so.
So an effective moral enhancement that increases the capacity for moral reasoning in the cognitively challenged would neither be a violation of nor an appeal to to their reason. It would not be like education or a talking therapy, but neither would be like the cruder forms of chemical or psychological manipulation. It could enhance the moral capacities of people but it would do so by tying them into technical networks that, as we know, can be co-opted for ends that their creators never anticipated. It might enhance the capacity for moral agency while also increasing its dependence on the vagaries of wider technical systems. Some would no doubt see such a development as posthuman biopower at its most insidious. They would be right, I think, but technology is insidious precisely because our florid agency depends on a passivity before cultural and technical networks that extend it without expressing a self-present and original human subjectivity.
There’s an epic flame war over at Three Pound Brain in response to Scott Bakker’s discussion of Levi Bryant’s Object Oriented Ontology. I’m sitting this one out like my hero Custard the Cat. In part because, I’m just too busy and in part cos’ I don’t want to distract Scott from the trudge to Golgotterath and the moral necessity of euthanizing our immortal souls.
Brother Cavil’s speech from BSG’s episode ‘No Exit‘ is the plaint of a being whose morphological freedom has been arbitrarily denied. Cavil’s romantic transhumanism is far more cogent and appealing, here, than Ellen Tigh’s feeble humanism.