Donald Davidson Among the Aliens

The speculative posthumanist holds that a) radically alien forms of life that could not be interpreted or evaluated in ‘our’ (human) terms are possible and b) humans or their ‘wide’ descendants could evolve into such beings.  So speculative posthumanism (SP) is  committed to the possibility of radical aliens. But how are we to understand the claim that there could be radical aliens, let alone motivate it?

At first sight the possibility claim seems exposed to standard objections to the possibility of radically incommensurable conceptual schemes or languages. Some philosophers advert to the apparent absurdity of claiming that some terms or concept from a historical culture is incommensurable with ours while at the same time spelling it out. Thus Hilary Putnam claims it is incoherent for Thomas Kuhn to say that Galileo had ‘certain incommensurable notions and then go on to describe them at length’ (Putnam 1981).

If we can describe the incommensurable in our terms of our language or conceptual framework, it seems, it can’t be incommensurable with it. However, citing specific incommensurabilia is not an issue when considering hypothetical aliens. SP is consistent with the dated non-existence of posthumans so it cannot require examples of incommensurable posthuman forms of life.

However, the air of paradox remains since it can be objected that any thinker must meet satisfaction conditions for mentality that preclude drastic recalcitrance to interpretation.

The strongest challenge to the possibility of uninterpretable mental lives is  found in philosophies that tie  mentality to the possibility of interpretation. Donald Davidson’s integrated theory of mind, meaning and interpretation is perhaps the most systematic of these, so it is worth considering whether it might still leave space for the radical alien.

Davidson’s theory of mind is primarily concerned with the propositional attitudes – states of mind whose contents are expressible using declarative sentences. It holds that there are no intrinsically hidden or private intentional facts. This is because belief-desire discourse is utterly enmeshed in inter-subjective practices of interpretation. To be a true believer (or true desirer, etc.) is to be the kind of creature that can be interpreted as holding fine-grained, inter-articulated beliefs and desires.

Davidson claims that understanding fine-grained intentional distinctions such as between the belief that a predator is around and that a tiger is around requires a grasp of their different inferential consequences for other beliefs, or for action (The latter belief could warrant the action ‘Run!’, whereas the former might not). Moreover, even the capacity to have beliefs requires a grasp of the role of belief in intersubjective belief attribution since holding a belief requires an understanding of the possibility of being mistaken, an understanding which arises ‘only in the context of interpretation’ (Davidson 1984, p. 169). We thus have no choice but to attribute beliefs using the common coin of public language. The only publicly accessible nuances that could warrant fine-grained distinctions between attitudes are utterances of structured sentences.

The intensionality we make so much of in the attribution of thoughts is very hard to make much of when speech is not present. The dog, we say, knows that its master is home. But does it know that Mr Smith (who is his master), or that the president of the bank (who is that same master), is home? We have no real idea how to settle, or make sense of, these questions (Davidson 1984 Ibid. 163).

So being a true believer is predicated on part of one’s behaviour being interpretable as utterances. Conversely, for a creature’s behaviour to be thus interpretable presupposes it has attitudes with content as fine-grained as the sentences it asserts. It is only if assertion evinces belief that we can reconcile evidence for the truth conditions of sentences with occasional errors which owe not to differences in meaning but differences in belief: “belief is built to take up the slack between sentences held true by individuals and sentences true (or false) by public standards” (Davidson 1984, 153).

A creature whose behaviour is interpretable as sentences in a recursively-structured language would thus be a true believer. From this we can infer that the constitutive principles of rationality for such a creature would be like ours. For example, its beliefs would have to be generally consistent. Incoherence would sunder the links between beliefs that secure their content; making efforts of interpretation fruitless. Likewise massive error would render a cognizer uninterpretable since there would be no basis for inferring truth conditions from states of the speaker’s environment or for testing recursive truth theories that predict these correlations.

So Davidson-style interpretationism presents the speculative alien lover with a dilemma. Posthumans presumably do not merit the name if they are not some kind of sophisticated thinker. Yet if interpretable as sentence users their intentional states would satisfy constitutive conditions for belief. But beings whose mental lives are characterized by belief-desire psychology and holistic rationality and truth could not be all that alien it seems. Indeed, if the possession of beliefs entails the concept of belief, posthumans would need second order mental states (beliefs about beliefs or desires, etc.) and could be expected to evaluate their beliefs and actions in terms humans could understand in principle.

Thus our ‘posthuman’ successors might have beliefs nuanced  in ways that we flat footers can only imagine, but they would still be beliefs and desires and not weird hermeneutically opaque entities like Churchlandish brain-states.

Likewise, posthuman phenomenology might be alien in content. Posthuman pleasures might have phenomenal characters that require some currently unavailable body plan. Nonetheless, if the interpretationist is right, posthuman phenomenology could not be radically ‘other’ in form. Evaluating beliefs in the light of consistency requirements, planning utility-maximizing actions requires the capacity to self-attribute beliefs and desires. This, in turn, arguably supposes the capacity to think of oneself as an enduring ‘I ‘ – a persistent subject of mental states.

Thus, it seems, beings would need to be rational to qualify as posthuman and need phenomenological subjectivity to qualify as rational. Virnor Vinge’s suggestion that post-singularity intelligences might have non-subjective phenomenology would be demonstrably misplaced (Vinge 1993). However they might differ from humans outwardly or inwardly, our technologically uplifted successors would be people capable of evaluating their agency in the light of rationality norms.

So either we reject interpretationism tout court or we motivate the claim that posthumans might be recalcitrant to interpretation of the particular kind that Davidson envisages.

Interpretationism might be worth saving in some form because it secures a conceptual link between what a creature thinks and its activity; thus explaining intimate relationships between public psychological attribution and the folk ontology of propositional attitudes. However, interpretationism appears committed to a kind of regional ‘correlationism’ (Meillassoux 2006). It holds that thinking about the attitudes is at bottom equivalent to thinking about how attitudes are interpreted. By contrast, the speculative posthumanist needs a way of thinking about posthuman mentation without committing to a theory of interpretation for posthumans. If this philosophical account can be rendered coherent, then we will have left space for a hermeneutic obstruction that is not vitiated in the very act of thinking it.

Bibliography

Davidson, D. (1984), Inquiries into Truth and Interpretation, 155-70. Oxford: Clarendon Press.

Meillassoux, Q. (2006) After Finitude: An Essay on the Necessity of Contingency, Ray Brassier (trans.). New York: Continuum.

Putnam, Hilary [1981], Reason, Truth and History, Cambridge: Cambridge University Press.

Vinge, V. (1993), ‘The Coming Technological Singularity: How to Survive in the Post-Human Era’, http://www.rohan.sdsu.edu/faculty/vinge/misc/singularity.html. Accessed 24 April 2008.

3 thoughts on “Donald Davidson Among the Aliens

  1. A good post. I have a couple points to make in response though:-

    1) Interpretationism needn’t be understood as anything like a ‘regional correlationism’, as long as it’s understood to treat the being in certain propositionally contentful states as a matter of occupying a certain normative status, rather than being in any naturalistically or metaphysically describable state (e.g., having certain Fodorian internal symbol states, or bearing a relation to certain Russellian propositions). It’s not a correlationism if it takes such states to be genuinely instituted by wider social practices, rather than simply relative to them in their description. Indeed, we can treat being a rational agent as itself such a normative status (http://speculativeheresy.wordpress.com/2009/11/23/the-rational-animal/).

    2) I don’t see how being able to think of oneself as an ‘I’ requires anything like phenomenological subjectivity. At minimum, it requires that theoretical and practical commitments involving the ‘I’, e.g., ‘I’m in the path of a moving train’, should play a special motivating role in one’s own practical reasoning, e.g., disposing me to move out of the path of the moving train. This needn’t require anything like a special first-personal phenomenological dimension to my experience. If my only perceptual capacities were provided by cameras outside my body, this would open up new ways in which I could be mistaken whether I was genuinely in certain states (e.g., I might be viewing a doppelganger standing in the path of the train), but these mistaken beliefs would still dispose me to act, because they are beliefs about me.

    3) I think it’s worth recognising a principle/practice distinction when it comes to the idea of incommensurability. The Davidsonian argument just demonstrates that there couldn’t be a community of speakers that we could not in principle communicate with through some kind of translational schema, not that the actual construction of such a translational schema would be practically feasible. It’s the difficulty involved in producing and operating such schemas that constitutes real incommensurabilities between linguistic communities. The presence of radically different perceptual capacities and practical abilities, along with different practical ways of living, are the biggest practical obstacles to producing such schemas, and thus constitute real incommensurabilities by degree. Nonetheless, this wouldn’t undermine our ability to take such difficult to translate communities as genuine communities of rational agents able to interpret one another, as this only requires the identification of certain abilities, not a fully operational translational schema.

    4) The real advantage of broadly interpretational approaches to the nature of thought (Brandom’s included) is precisely that they separate it out from the causal mechanisms (be they biological or otherwise) which make it possible, and thus open up the possibility of entities that both genuinely *think* and are radically different from us in the precise ways that they manage to do so. If you want to abandon this kind of approach, you need a different way of characterising what thought is that is at the very least equally general, and you need to do so without stretching the meaning of the term ‘thought’ until it is unrecognisable to our common meaning of the term.

  2. Thanks for this detailed and helpful response, Pete.

    1) I’m not persuaded that describing intentional states within Davidson’s account as ‘normative states’ is an accurate reading of his position. Saying that coherence and consistency are constitutive conditions for mentality doesn’t require attributing prescriptive force to them. Consistency is a formal property that some collections of sentences have and others don’t. Some bodies of belief cohere; others don’t. The fact that a particular agent’s beliefs can’t be massively incoherent tells us something about the kinds of states that they are, just as the fact that the property of ‘….being longer than….’ is transitive tells us about the kind of property it is.

    To be sure, meaning is constitutively accessible from a third person point of view; but that doesn’t make it normative. Suppose Davidson is right and believers need a concept of belief; this tells us about the kinds of capacities that a system needs to count as a believer. It might be psychologically implausible to claim that a being could acquire the notion of belief without schooling in its application to others in a social context, but this schooling may simply involve acquiring a grasp of the truth conditions of belief attributions.

    2) The point about phenomenology needs working through, I’ll admit. Since I’m inclined believe that our phenomenology partly depends on our capacity for self-knowledge (or ‘probing’), I’m inclined to regard the capacity for self-attribution as a high level phenomenological constraint. But there’s a lot that needs to be worked out here. The distinction between phenomenological form and content is something else that I’m uncomfortable with and was merely retained here for expository reasons.

    3) Yes, I accept that even with identical constitution conditions for mentality there could be all sorts of incommensurabilities of the less radical type. The most that Davidson shows is that these could be inaccessible in principle. However, I want to push the idea of the incommensurable as far as it can go and this doesn’t yet go far enough.

    4) Very well put and most helpful. That said, I think it’s obvious that there are general accounts of cognition – those that treat it as computational processes over representations, say – that don’t presuppose propositional thought as Davidson understands it. I think Davidson’s approach – far from being anti-naturalistic – actually helps us build plausible explanatory bridges between computationalist accounts and a theory of full blooded intentionality. Firstly, Davidson tells us that an intentional system would have to behave in ways that make radical interpretation possible and worthwhile (here’s he’s with Dennett – just less permissive about the conditions for intentional systemhood). Second, while intentional attributions are not reducible to descriptions of computational states, it’s plausible that only beings capable of certain kinds of computation could be intentional and that these can be discovered. In short, an intentional system may be just a computational system of a particular kind. There may be others that by virtue of the kinds of data structures they employ wouldn’t yield usable intentional interpretations even in principle. But they could be thinkers nonetheless. I’ve gone someway to arguing for this in my piece on instrumental eliminativism and cognitive augmentation, available in a rough form here http://enemyindustry.net/MM_InstElim7.pdf

  3. In my response to point 3, I meant to write: ‘The most that Davidson shows is that these couldn’t be inaccessible in principle.’

    D

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s