Aristotelian Posthumans

I’ve argued e that posthumans would have to be, in some sense of the term, “autonomous entities” capable of operating outside the scope of the socio-technical network I refer to as the Wide Human (Roden 2013). A being is autonomous if it is self-governing. According to the modern practical philosophy that follows Rousseau and Kant, autonomous beings (paradigmatically human beings) are those that can freely determine the principles by which they act and live.

A nonhuman animal may have desires and drives, but not being rational it cannot represent those desires to itself or attempt to be motivated by different ones (Frankfurt 1971).

It appears that a posthuman entity would need to be autonomous in something like this sense since it will have to become both functionally and existentially independent of the Wide Human (WH).

Within WH functional and existential independence are related. Entities belonging to WH accrue functions that have come to be required by biological (narrow) humans. They also tend to exist in a form that is perpetuated over time. Further, their existence in historically stable forms is explained by this contribution to narrow human ends.

We can express this relationship as a “consequence law”: where E is the dated fact that some entity ?  exists in a historically stable form in a context and (E—>H) is some humanly valuable causal function F of ?.

Thus: (E—>H) –> E Expresses the fact that (E—>H)  accounts for or causes E.

A consequence law supports counterfactuals. If  (E—>H) ceased to obtain,  ? would cease to occur in in a historically stable form in a context  (E would become false).

Technological change is capable generating new functions and it is quite normal for realizers of those functions to be superseded by new realizers. For example, automobiles have superseded horses in the developed world because the functions associated with horses can be better served by cars. The result is that horses have ceased to occur in the contexts in which they were formerly used (public transport and the provision of motive power). However, objects can acquire new functions. So we need to complicate this formula by stating that ?  exists only so long as there is some human-related function that it continues to serve in some context (EF) .

EF expresses the functional dependence of those Wide Humans that are not Narrow Humans upon WH. They exist in a given context only so long as there is some narrow human purpose that they fulfil. It also expresses the fact that WH is predicated on the existence of narrowly human individuals. Were human individuals to disappear so would the functions and thus the historically stable forms of these Wide Humans would disappear also.

Certain species of domesticated animal like pigs and cows are thus Wide Humans (as I have advertised) since they would cease to exist in most of the current contexts in which they occur were it not for the fact that they serve human needs for milk and meat. For example, if all humans converted to veganism and vegetarianism cows and pigs would likely survive only as domestic pets or as feral pigs or feral cattle. Those belonging to the former group would remain parts of the wide human while those in the latter group would be effectively outside it since their continued existence in their new context would not depend on their fulfilment of human related functions.

Important Note: the term “human related function” does not refer to something like core human needs or authentically human needs, or invariant human needs. No such essentialist commitment is implied here. A human related function is simply a function that would not exist but for the existence of narrow humans. These can be as artificial and as historically contingent as you like: computer gaming, cash dispensing and pornography are human-related functions in this sense.

Nor does EF commit us to some voluntarist conception of functions. It does not require that functions come to exist through the intentions of individuals or groups. It is quite compatible with the view the some human-relative functions come into being by incremental processes in which no individual intended that a particular function or activity come into being.   We can call the independence from the human-related functions of beings outside WH  negative autonomy (analogously perhaps to Berlin’s notion of negative liberty).

Arguably, however, we need a positive conception of posthuman autonomy. Here’s why. There are plenty of objects that are a) technically fashioned and b) can exist for some extended period of time after they have ceased to be technically useful: hulks and ruins, for example. However, unless they are preserved for aesthetic purposes hulks and ruins have no functions at all (having aesthetic functions would, of course, qualify them for membership of WH).

Hulks and ruins are unlike feral animals in that the latter seem to carry out many non-human-related functions: for example, mating, foraging, giving birth, etc. A conception of disconnection that resulted in hulks and ruins having posthuman status would be simply too broad. Hulks and ruins are existentially independent of WH but not functionally independent in the way, say, that feral animals are. But functional independence would require that posthumans would be able to determine non-humans purposes or functions.

Rational subjects can do this within certain parameters. They can choose the projects that will give meaning to their lives or (as Rawls puts it) their conceptions of the good.

However, I have argued elsewhere that the notion of the rational subject is too narrow to comprehend Posthuman Possibility Space (PPS). It is also not well defined since we do not have a self-standing conception of the subject that could not be revised by some future sciences of cognition.

Thus we will need a more general conception of autonomy if we are to get a sense of the kind of being that inhabits PPS. Some of these may have a mode of being that is rather like the Dasein of humans – reflecting upon their plans for life in terms formulated in shared public languages, for example – but others may lack what Thomas Metzinger refers to as a self-model. They may not think of themselves as unique individuals whose lives could go better or worse and they may not experience or participate in intersubjectivity. If so, the category of the subject is not going to furnish the general conception of autonomy we require.

The example of feral animals points us towards a different and arguably more fundamental conception of autonomy. In the philosophical tradition, the bodies of biological organisms have been understood as having functions fixed by their contribution to activity or self-maintenance of the organism.

This conception has its classical formulation in Aristotle’s biology and metaphysics who argued that the presence of certain arrangements of parts and materials in plants or animals can be explained by their specific contribution to the life activity of the creature. Thus in the Physics, Aristotle argues that the presence and specific arrangement of teeth in the mouth of animals is explained by their contribution to animal nutrition (Phys 198b25-35). Similarly, the form of the parts of organs are fixed by their contribution to the functions of those organs, whose function is determined by their contribution to the life activity of the entire organism. In animals without hard external shells, eyelids have the function of protecting the thin, moist eyes, which, in turn, realize the function of seeing in the whole organism (Ariew, Cummins and Perlman 2002, 14).

Now, this explanation conforms to the consequence law schema set out above. It is teleological (explanation in terms of purposes) because the propensity of these structures (teeth, eyelids) to contribute to the functions of the organism explains their form and thus their being. Moreover, these functions depend on the continued life of the whole organism.  The contribution of these parts to life-activity of the organism is thus the ontological ground of their functions: “Thus, organisms and their parts are what they are only when living. As for the products of techne … they are defined by the function they must perform.” (Moya 2000, 321).

An animal is similar to a technological artefact like a couch in that both it and its organization can only be understood in terms of the function is performs. According to Aristotle the function (and thus the form or being) of an artefact like a hammer or couch depends upon something external to the matter of the artefact.

However, the form and function of a natural thing like an animal or plant depends only on the thing itself ((Met 1070a7-8). Whereas an instrument serves a function set by its users, the body of the animal is an “instrument (organon) performing or making manifest its own act of living (its entelecheia)” (Moya 2000, 326).

The autonomy of the organism as understood by Aristotle, and by long tradition of biological thought that follows, consists in its capacity to determine its existence: where “existence” here needs to be understood as its mode of life or function – it’s being the kind of thing it is. This clearly a different conception of self-determination from the Kant-Rousseau conception of rational-autonomy since the organism does not need to consciously choose its function or being in order to have it.

So does this provide the concept of positive autonomy that we are seeking?   Well, if we understood posthumans as organisms in the Aristotelian sense we would be able to explain their existential independence in terms of their functional independence. They would not need humans to set their ends because (qua organisms) they would be teleologically self-fixing.

However, the Aristotelian reading of posthuman autonomy will not do. There are at least six objections that can be levelled at it and they all carry some weight.

1)      The Aristotelian account of organism is committed to objective teleology. However, post-Darwinian biology provides a far more satisfactory explanation of biological order than Aristotle and other organicists. Since it rejects objective teleology, we should too (Darwinian Objection)

2)      The Disconnection Thesis implies that some posthumans could be WHD’s of artefacts, not organisms. But artefacts have only derived functions and organisms have original functions. There is thus no technological process whereby the WHD’s of artefacts could acquire original functions and thus jump ontological category (The Category Objection).

3)      Contrary to Aristotle, the functions of the parts of creatures are not exclusively determined by their contribution to the whole organism. Mechanisms at a lower level than the organism have functions independently of the systems to which they belong. Treating the organisms as a self-sufficient biological totality or whole is thus a mistake (Anti-Holism Objection).

4)      Organisms are not self-determining, in any case. They and their components can acquire new functions and thus new forms of existence by being “iterated” into new contexts (Functional Indeterminacy Objection).

5)      The organismic perspective is a kind of vitalism; but the wrong kind. It envisages an ordered nature with functions analogous to those of the human world. But “nature is not natural” in this sense. ( Neo-Vitalist Objection).

6)      The organismic perspective is, in any case, parochial. The justification for specifying the posthuman in terms of WHD is surely that posthumans are liable to be postbiological. Thus the organic/inorganic distinction would not apply to them (The Post-Biology Objection).


Ariew, Andre ; Cummins, Robert C. & Perlman, Mark (eds.) (2002). Functions: New Essays in the Philosophy of Psychology and Biology. Oxford University Press.

Collier, John (2002). “What is autonomy?”,, accessed 04/03/2011.

Colebrook, Claire (2010). Deleuze and the Meaning of Life. Continuum.

Frankfurt, Harry G. (1971). “Freedom of the will and the concept of a person”. Journal of Philosophy 68 (1):5-20.

Moya, Fernando (2000), “The Epistemology of Living Organisms in Aristotle’s Philosophy”, Theory In Biosciences 119(3-4): 318-333.

Roden, David (2012). “The Disconnection Thesis”, in The Singularity Hypothesis: A Scientific and Philosophical Assessment, edited by Amnon Eden, Johnny Søraker, Jim Moor, and Eric Steinhart. Springer Frontiers Collection.

6 thoughts on “Aristotelian Posthumans

  1. Excellent piece. Have you ever thought of tackling this issue from a ‘problem of induction’ angle, David? On my view, ‘autonomous behaviour’ only seems ‘special,’ a break in the response-continuity of mechanism because of various perceptual and cognitive shortcomings. (Dennett says as much as well, only with a different axe to grind). The idea, in other words, is that the only way to crack the autonomy nut is to see it as the artifact of TWO mechanistic systems, observed and observational.

    This is the thing I continually play with in the fantasy novels, btw, how autonomy arises and vanishes depending on the mere presence of Kellhus! And this is what makes things interesting from a posthuman standpoint, because you could then suggest that OUR ‘autonomy’ depends on our computational constraints, and that the development of some supercomputional posthuman would mark the end of human autonomy, and thus, you could say, the category of ‘human’ itself.

    This in turn would allow you to diagnose attempts to ‘ground autonomy’ in some set of natural facts as noocentric errors, a kind of categorical (as opposed to merely psychological) anthropomorphism.

    1. It’s a bit of a bump for a mug who barely knows their entelechiea from their psuche, but I’m having to tackle the issue of autonomy head on and this means tackling the body with organs.

      There’s some great stuff on biological autonomy by people like Mark Bedau and Kepa Ruiz-Mirazo in Synthese 185. The only way I can approach this is by thinking of self-maintaining entities, entities that, as Ruiz-Mirazo and Moreno say, are responsible for maintaining the boundary conditions in which their characteristic dynamic behaviour can persist. But I also like what John Protevi and Claire Colebrook have been saying from a Deleuzean perspective and Martin Haggelund, via deconstruction. I guess this commits me to a kind of transcendental enquiry, but it’s not humanist and it’s fallabilist in spirit (I hope).

      I suppose, I could approach it from your angle. Maybe I’ll do that when I miserably retract everything I will say in Posthuman Life. But right now my life is complicated enough and I need that neo-vitalist spark!

      So, I should be shit-scared of Kellhus. He’s such a magnificent conceit though!

  2. Kellhus IS the eternal return… So I guess it depends on your powers of affirmation 😉

    I’m not familiar with the literature, but I just worry about the ease with which ‘autonomy’ or ‘agency’ when used in a life sciences context can be equivocated, say the way Bechtel tries to pull the intentional rabbit out of the hat with his ‘active’ (as opposed to ‘responsive’) characterization of mechanism. So you take those systems that we happen to interpret intentionally, describe the properties (autopoetic boundedness, etc.) that distinguish them from other natural systems (‘fish’ versus ‘river,’ say), then claim to have explained ‘autonomy’ or ‘agency.’ (Not to say that this isn’t what *everybody* does. Just think of how even Churchland simply assumes he’s accounting for what we think we do, ‘know,’ when he accounts for what the brain does: process environmental information. All you have to ask is, ‘What makes this processing ‘about’ the environment, such that it can be ‘right or wrong’? to see the missing imperial clothes.)

    Otherwise, I always thought psukhe was just entelechia with a wife and mortgage…

  3. 🙂 Salutary words. That said, even a description of a mechanism qua mechanism requires a functional characterizations of components like neurons or organelles. Now, we can say something like “Functional ascription reflect our homespun normative judgements about what is for the sake of what” implying that almost any account of the world – short of a full specification of its microdynamics – is in the eye of the beholder: it’s qua’s all the way down, or nearly so. Or we commit ourselves to things like mechanisms and real functions (thus real screw ups). One way of naturalizing function is via selection-history (the etiological account: a component is an F’er even if it does G, H… if it has been slecting for F’ing). But this is problematically historicist because it makes what a thing currently does in actual systems epiphenomenal with regard to fixing function. So another approach (championed by John Collier, for example) is to develop an ontology of autonomous systems where a system is autonomous if it contains lots of relatively closed and self-reinforcing processes. As far as I understand it, process is more closed the more it takes it own outputs as inputs, and processes are more cohesive the more that they interdepend on one another for their maintenance. So a causal role can be understood as functional if it contributes to the continued cohesion of a system. For example, cellular processes require the transport and storage of energy. So the function of components like mitochondria is energy storage because this the causal power that contributes to the cohesion of cellular processes. This seems to qualify as objective and mind-independent function. It resembles a kind of immanent teleology, but it is subtly different. For one thing it allows us to identify functions without engaging in functional explanation. So there may be an evolutionary story whereby mitochondria exist because they function as energy stores (consequence law (M — >E)— > M)) but this conception of function even works for Swampman since it does make the current role of a component epiphenomenal in relation to its history.

  4. Sounds very interesting, and it’s definitely something I’ll check out. It’s the ‘ontology’ and ‘mind independence’ I find sticky. If it turns out our brains are Bayesian prediction machines, ‘comprehension’ at the most fundamental level will consist in prediction, and the Aristotelian ontologization of ends seems inevitable given the kind of system we are – an inevitable cognitive ‘illusion,’ as profound as they come (things like the phi phenomenon, for instance, certainly suggest the brain plays with time to organize stimuli into ‘onto-narratives’). My guess is that the ‘teleological impulse,’ the need to insert ends into the order of things to feel we’ve ‘explained’ that order, will be diagnosed as a consequence of the problem of induction faced by the brain.

    The fact is, we simply did not evolve to theorize these things. We should expect to be constantly bumping up against heuristic limits – and our intuitions as a result.

Leave a Reply

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

You are commenting using your 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