The three concepts of autopoiesis, equilibrium and homeostasis function in all these domains: neuroscience, cognitive science, philosophy of mind, social theory and future studies. These concepts all presuppose a certain understanding of time, and suggest – as I state in the title to this essay – that the organism has no future. In itself, or if it remains in itself, the organism has no future. There can only be a time to come if we recall our embodied, relational, world-attuned being. The world within which we are situated – if we accept that ‘we’ are nothing other than the situated and responsive beings that we are – is always a world encountered in terms of possible responses. We exist in meaningful milieus. Our condition as embodied, as relating to the world as the beings that we are, is that the world is given as this world for us. To a certain extent, then, we are proto-ecological, originally attuned to our milieu. If we have a future, so it is argued, it cannot be one of calculation, instrumental reason and the mere continuance of ourselves in isolation. Our future could occur only if we remind ourselves of embodiment, if we recall what we really are and once again live our attunement to our milieu not as accidental but as intrinsic to our very being.

Claire Colebrook, ‘Time and Autopoiesis: The Organism Has No Future’ in Deleuze and the Body, Laura Guillaume and Joe Hughes (eds.) Edinburgh University Press, 2011, pp 23-4

Long before I began working on autopoiesis, Claire Colebrook had already suggested critical points based on time against the concept. Like mine, her arguments owe something to Deleuze and Guattari, but that lineage underplays the originality of her ideas and her wider sources in feminism, psychoanalysis, history of philosophy, literary criticism and theories of body.

Colebrook’s critique is not based directly on time. She argues from a theory of desire and embodiment towards the conclusion that theories of autopoiesis leave life without a future. This is because an image of the body as protected by a boundary and desiring its own continuance leads to a misrepresentation of life as balance (‘equilibrium’) between completeness and self-sufficiency. In turn, this focus on stability and preservation denies the future as unpredictable change within a milieu.

When Colebrook concludes that autopoietic systems can have no future, she is relying on a narrowed definition of the future. The future isn’t what will happen; it is what happens differently. Repetitive and relatively balanced systems have no future because the future cannot be about ‘mere continuance’. This is a dramatic definition, reminiscent of the phenomenological and existential critiques of life devoid of freedom covered in her introduction. It is justified by a theory of desire. Even if a life appears to be one of simple persistence, this likeness masks deeper desire for – need for – change.

Autopoietic balance is across a boundary defining inside as drive for sufficiency (the autonomy of autopoiesis) and outside as source of a drive for completeness (the need for energy, coupled systems and a homeostatic relation to environmental change). An autopoietic process is attracted to the outside and dependent upon it, but it must weigh this movement beyond its boundary against the need to maintain autonomy and self-containment.

The direction of the critical point is set by scepticism towards the successful maintenance of balance. Instead of perpetuating autonomy, boundaried systems seeking completeness are self-destructive. The image of an attainable harmony between inside and outside conceals a necessary ‘malevolence’ proper to desire for completion allied to goals of internal stability (I will control it/them; it/they will not control me).

The desire for completeness comes into conflict with self-sufficiency or the desire not to be exposed to contingency, risk or an influx of otherness so that it would destroy all border or limit […]

‘Time and Autopoiesis: The Organism Has No Future’, p 17

Colebrook’s argument goes beyond direct criticism of how autopoiesis works in relation to time. As shown in the passage I opened with, her point encompasses many fields and ways of thinking. Autopoiesis is symptomatic of a wider error in ‘neuroscience, cognitive science, philosophy of mind, social theory and future studies.’ There is a mistaken image of life based on autopoiesis. It is corrosive in two complementary ways: as representation and as ethical ground.

First, it is a failure of observation. Living forms and ways of life do not conform to the image. Counterexamples include viruses requiring no boundaries and technical assemblages beyond the boundaries of the human body, like artificial limbs or cognitive aids. As parasites, viruses escape the model of autopoiesis by denying autonomy to host or parasite. As extended mind, technical assemblages blur the boundary between inside and outside.

More broadly, life isn’t boundary-defined, as inside and outside, but rather milieu-defined, as multiple, open-ended relations beyond any defined boundaries. These relations unfold in risky and destructive ways creating an immanent milieu, instead of happening in a pre-defined environment. This milieu is a space made by productive and meaningful relations, rather than constituted by a series of stable divisions and distinctions.

Second, the image leads to misleading ethical principles. Life isn’t based on harmony designed to ensure self-preservation. It is based on creative and risky drives. Following Freud in Beyond the Pleasure Principle, Colebrook defines other-oriented drives as malevolent: unpredictably destructive for self and other, even as they constitute productive and rewarding dependencies and alliances. The problem of ethics isn’t about how to live in harmony, but how to live with the necessity of malevolent but creative destruction.

The contrast between ethics of harmonious preservation and creative malevolence leads to Colebrook’s repudiation of sacred images of autopoietic harmony: the animal in a fitting environment; the tended to infant; the Buddhist sage; practices of mindfulness. Those images are paradigms for ethics that have missed the inherent destructiveness necessary for any living interaction.

Colebrook’s critique of images of harmonious balance across a well-tended boundary shows the usefulness of a semiology of autopoiesis. The concept works not only as description of life but also within a structure of signs. It reinforces images of successful self-containment and self-sufficiency in contact with a wider environment.

These settled images contrast with those defended by Colebrook, such as more antagonistic and risky existence within a meaningful but also bewildering milieu calling for new interpretations. Her point about lack of future can also be seen as semiological. Autopoiesis bolsters images of continuity and permanence in the future, against more open images of difference and novelty.

In her conclusion, she draws attention to the high political stakes of this clash of signs for different ways of thinking about life and the future of the world, at a time of natural catastrophes caused by global warming.

Perhaps it is only in our abandonment of ownness, meaning, mindfulness and the world of the body that life, for whatever it is worth, has a chance.

‘Time and Autopoiesis: The Organism Has No Future’, p. 26

The difficulty with these semiological analyses is that while making points applicable to the adoption and dissemination of autopoiesis, they are also vulnerable to counterarguments pointing out that the concept is different from the images generated from it. Autopoiesis might encourage signs of boundaries, controlled futures and harmonies, but they aren’t entailed by it, nor do they reflect its precise sense.

These ripostes are all the more important given the fraught political context. If it is only the image at fault for mistaken political acts, then to discard the concept of autopoiesis could compound an error by removing a valid explanation on false grounds.

I have focused my studies on Maturana and Varela because their rigorous definitions give the most robust version of autopoiesis. As such, it is also the most distant from a vaguer and more pliable image, despite their influence on subsequent works and popular understanding. The challenge is then to decide whether criticisms of the image still apply to the more strict concept.

Semiology and psychoanalysis are prone to distort concepts. Signs, symbols, images and their structures tend towards objects, bodies and pictures. This prevalence of representation is not necessary – indeed, my definition of signs as selected sets goes against it – but it is a dominant feature of signs and communication. They privilege the sensual: the seen above all, but the felt too, and least but still significantly other senses, such as the heard, intuited, absorbed, forewarned, desired and eroticised.

This bias towards sensation is problematic for any interpretation of Maturana and Varela’s logical, rationalist and dynamically mechanical philosophy of life. Their concept of autopoiesis is neither sensual, nor subjective, nor objective. Its processes are not based in sensation. They do not have a subject. They do not aim at objects, or seek to produce them, or operate thanks to them. The only requirement is autonomous, perfectly contained and creative self-production of an abstract logical process.

The pure definition of autopoiesis leads to the following answers to Colebrook’s criticisms:

• The boundary in autopoiesis is logical. It is ‘perfect continence’ in the sense of autonomy and separation from anything else as a process. Physical boundaries are only approximative expressions of this prior logical distinction, not between inside and outside, but between the distinct and what it has been distinguished from;
• The distinct is a logical form: an abstract plan for autonomous self-production. It is therefore not to be confused with an ostensible subject, object or body;
• The abstract plan constituting the inside of autopoiesis is about production and change, not stasis or preservation. The only thing that must be preserved is the logical process for production;
• The abstract plan can involve different levels of abstraction, where lower levels of a plan can change radically while leaving the core logical process autonomous and perfectly contained;
• Autopoiesis does not correspond necessarily to the preservation of any given physical form or any given being, type of species, or observed goals, behaviours, inputs or outputs.

Colebrook claims autopoiesis has no different future. This does not apply to Maturana and Varela’s concept, since an autopoietic system can have any number of different futures in physical appearance, ways of living and ways of operating, so long as a high level abstract plan remains the same.

Ideas such as bootstrapping ‘load n programs’ and scaffolding ‘learn from trusted peers’ capture this multi-tiered form of systems and processes. A processing core can initiate a great variety of other systems. A common ability to use environmental scaffolds can give rise to many different life forms.

If the algorithm for an autopoietic system has the principle ‘survive by shapeshifting’, or ‘thrive by experimenting with modes of living’ or ‘choose the most successful operating system to adapt to new conditions’, or all three, then it will be highly differentiated over time to an external observer, to other life forms, and – in most aspects – to itself.

The rejection of autopoiesis due to the images it generates does not apply to Maturana and Varela’s strict definitions, since it contradicts those images. To avoid confusing implied image and correct definition, my criticism of autopoiesis is through a timed logic that precedes the concept, in the sense of being a necessary condition for it.

Autopoiesis has a problem with the future, not because of ‘mere continuance,’ or because it fosters signs and images of secure boundaries and stasis, but rather because its efforts to change and adapt are governed by an unchanging core set of principles and methods.

This core cannot ensure its own perfect containment over cycles of change. The timed logic that holds for any passage from the past through the present and towards the future involves processes that must necessarily cross boundaries and transform autopoiesis right down to its supposedly unchanging central command. This is what Deleuze’s philosophy of time teaches us. Time is a series of processes that respect no limits and no claim to autonomy.

Returning to the answers to criticisms against autopoiesis given above, my thesis about a prior timed logic leads to the following critical refinements:

• A logical boundary between the distinct and that which it is distinguished from is constituted and maintained over time. Constituting processes cross the boundary they are designed to maintain;
• Abstraction is not outside time. It takes place in relation to the past, present and future such that nothing is timelessly abstract;
• The unchanging logical core of any autopoiesis is illusory, if taken independently of how it is put into action in specific cases. Each expression of the plan inserts it into a timed logic that it cannot be separated from;
• The distinction between higher unchanging levels of an abstract plan and its alterable lower levels does not hold over time. There is always interference between levels. The logical core is porous and corruptible;
• Since any process in time is in principle connected to the whole of time, to all of the past and all of the future, the abstraction designed to preserve an autopoietic core is doomed to failure. The limitless histories and futures of any autopoietic process contradict its claims to autonomy.