Reconnecting biology, social relations and epistemology: A systemic appreciation of autopoietic theory
Brocklesby J. (2004) Reconnecting biology, social relations and epistemology: A systemic appreciation of autopoietic theory. International Journal of General Systems 33(6): 655–671. Available at http://cepa.info/3759
Table of Contents
2. Autopoietic theory: the general contours of a systemic description
3. Biology: the first phenomenal domain
4. Relations and interactions: the second phenomenal domain
5. The relationship between the two domains
6. From biology and social relations to phenomenology and the derivative epistemology
This paper seeks to tease out the systemic character of a body of work that elsewhere in both the primary and secondary literature tends to be described, discussed and applied in fragmented and reductionist terms. The origins of “autopoietic theory” may be traced back to experimental work in cellular biology and neuro-physiology and to the concept of “autopoiesis” (a theory of living systems) itself. From there, it has extended its coverage into a wide range of diverse areas including cognition, perception, emotion, evolution, language, culture, epistemology, the philosophy of science and ethics. Against this background, the paper seeks to outline a high-level systemic interpretation of autopoietic theory; specifically one that integrates its various biological, social and epistemological components and which shows that it is best evaluated and understood as an explanatory whole and not in a reductionist manner.
Key words: Autopoiesis, autopoietic theory, systems, living systems, biology, epistemology
This paper has an explicitly systemic purpose. It seeks to articulate the essential coherence and systemicity of a body of work that is almost always described, discussed and applied in fragmented and reductionist terms. “Autopoietic theory” – the body of work in question – is mainly associated with the Chilean biologists Humberto Maturana and Francisco Varela. Its origins may be traced back to experimental work in cellular biology and neuro-physiology and to the concept of “autopoiesis” itself, a term that describes the self-producing, self-generating dynamic that, so the argument goes, allows us to distinguish living from non-living systems. From these biological and experimental beginnings, autopoietic theory has evolved into a complex theoretical edifice spanning domains as diverse as cognition, perception, intelligence, emotion, evolution, language, culture, epistemology, the philosophy of science and ethics.
Given such breadth of perspective, it could be said that autopoietic theory provides a “connection” of sorts across areas of interest and fields of inquiry that elsewhere continue to be mediated by the traditional academic division of labour, and is therefore inherently systemic. But appreciating this “inter-relatedness of things” is by no means easy. It is the case that Maturana (1980, 1988) encourages readers to consider his work as a coherent whole, but disappointingly fails to present it as such. Seminal texts such as “Autopoiesis and Cognition” (1980), and “The Tree of Knowledge” (1987) provide a broad perspective but these works concentrate mainly on the biological aspects of autopoietic theory, and were written before Maturana fully worked through the epistemological and phenomenological derivatives of these. Looking elsewhere, the primary literature itself does not contain an accessible systemic account of the theory as a whole that spells out how the various component parts relate to each other and importantly how the two key domains – the biological and the social – interconnect.
Recently, Maturana (2003) has sought to take stock and “reflect on the history of some biological notions such as autopoiesis, structural coupling and cognition”. Yet his retrospective account is no less reductionist than the work that precedes it. The emphasis is still very much on the parts rather than the whole. The task of “putting it all together” is something of a conjuring trick that the reader is required to fathom out independently.
And what of the secondary literature? Here, one group of writers – usually with application in mind – have tended to focus on a particular aspect of autopoietic theory and have couched it in terms of its ability to shed light on defined areas of interest or professional practice such as the law (Teubner, 1984), family therapy (Dell, 1982; Goolishian and Windermann, 1988; Hoffman, 1988), psychotherapy (Efren and Lukens 1985; Efran et al., 1990), organization theory (Morgan, 1986), viable systems (Espejo et al., 1989; Beer, 1994), information theory (Winograd and Flores, 1987; Mingers, 1996), organizational change (Kay, 1997) and systems education (Kay, 2002). These contributions have generated novel insights in their respective domains of application. At the same time, when used in a specific context, this does carry with it the possibility of concepts becoming modified and even distorted to levels that betray their original intended meaning (see, for example, how the term autopoiesis itself is used in the context of viable systems theory (Brocklesby and Mingers, 2004)). Equally when abstracted from the broader conceptual whole, theoretical concepts and explanations can be difficult to reconcile with practical concerns.
Another group of writers focusing on distinctive aspects of the theory such as philosophy, cognition and social autopoiesis, have sought to engage in theoretical debate, to connect with other literatures, and to delve into and debate various controversies (see, for example, Luhmann, 1986, 1990; Berman, 1989; Mingers, 1990, 1991, 1992a,b, 1996a,b, 1997; Birch, 1991; Robb, 1991; Zeleny and Hufford, 1992; Kay, 2003). From time to time these debates have concluded with a negative assessment being made about the overall value of Maturana and Varela’s work. This, I submit, is not entirely fair. Surely if validity judgements are to be made then it ought to be done at the level of the systemic whole and not at the “component” level.
In what is almost certainly the most comprehensive secondary text produced to date on autopoietic theory, Mingers (1995:2) claims that ”…the work as a whole has a consistency and coherence across a wide range of domains that is rare indeed”. Building upon previous work (Mingers, 1990, 1991) in which he examined different aspects of the work of Maturana and Varela, Mingers’ main project was to demystify autopoietic theory and evaluate applications within various areas of professional practice. When measured in these terms the book is a seminal contribution. However, gaining a sense of the internal consistency and coherence of autopoietic theory from this source is by no means easy. It requires a degree of cross-referencing and mental agility that many readers will find extremely difficult.
Set against this background, the following paper seeks to outline a high-level systemic interpretation of autopoietic theory. Specifically the paper seeks to integrate the various biological, social and epistemological components of autopoietic theory and demonstrate that it is best considered a coherent whole whose contribution should be measured in these and not in reductionist terms.
2. Autopoietic theory: the general contours of a systemic description
If we are to think of autopoietic theory more as a systemic conceptual whole and less as an aggregation of different “aspects”, then the most obvious question to address is what sort of whole is it? My take on this is that fundamentally autopoietic theory is an explanatory system. It is a systematic explanation of the capacity of human beings, as biological and social beings to “observe”, i.e. to make distinctions, to utter cognitive statements about their worlds, to explain experiences and to act on these.
Before we consider this proposition in more detail, it is worth noting what Maturana has to say about explanations in general. First and foremost, explanations are observer-dependent. They arise as abstractions of people’s experiences, not as truths that pertain to an independently existing world. This proposition applies as much to science as it does to any other form of explanation. It also applies to autopoietic theory itself, i.e. it is an observer- dependent theory of observing.
More generally, Maturana claims that if we attend to how people explain in daily life, we see that there are two processes involved. Firstly there is a “generative process” that accounts for the experience to be explained. Secondly there is, or are, other criteria of evaluation that arise as an observer assesses the congruence between his/her experiences and the proposed generative process. Such criteria can be informal and personalized (e.g. in the case of natural phenomena does the observer prefer a scientific or a spiritual explanation?), or they can be highly formalised (e.g. in the case of the so-called “scientific method” (see Maturana, 1990)). On this view, the basic form of an explanation is noticeably similar to the structure of activity models or “holons” that are such a well-known feature of soft systems methodology (Checkland, 1981; Checkland and Scholes, 1990). There is a linked generative system of actions that leads to a defined outcome (in Maturana’s case, the experience to be explained), and there is some underlying “worldview” that “justifies” the existence of the system.
In general terms, it could be said that autopoietic theory demarcates a generative process that gives rise to all the phenomena that is entailed in observing. But what is this process? In answering this question, Maturana suggests that looking to the conditions that interfere with observing provides a broad clue as to what the key components to such an explanation might be. Thus, he claims that since any interference with either biology or language impedes observing (Fig. 1), then – at the highest level of abstraction – biology and language are pivotal explanatory components. Autopoietic theory then arises as a comprehensive and coherent explanation of how these two elements arise and, importantly, of how they interact. More precisely it arises as an explanation of the nature of (biological) living systems, of how – through social processes – basic language capabilities arise, of how these become more fully developed in living systems that possess a nervous system, and (in human beings) of observing as something that happens “as a manner of living in language” (Maturana 1993a).
And the “worldview” that underpins and gives legitimacy to this generative process? To answer this question we need to turn to Maturana’s suggestion (1988, 1990, 1993), that there are two distinctive ways of explaining the capacity of human beings to observe.
Figure 1: Explaining observing – basic elements of an explanation.
One approach – which he rejects – is to regard observing as being the result of innate properties, or as an endowment of the human condition. On this view, language is a characteristic of human beings, reasoning is a primary ability and knowledge is a property of the human mind.
Alternatively one can, as Maturana puts it, “accept the question of the observer”. This rejects the idea that observing is a natural endowment. Instead it asserts that observing is the result or outcome of a generative process involving other phenomena that are linked together in a distinctive way.
Extrapolating from the basic proposition that biology and language generate the phenomenon of observing, Maturana claims that there are two “phenomenal domains” involved (see Fig. 2). Thus, he says:
As living systems, we exist in two non-intersecting phenomenal domains: the domain of our realisation in our bodyhoods (the domain of physiology) and the domain of behaviour (the domain of our interactions as totalities).” (Maturana, 1988: 63).
I shall say much more about the precise nature of the interaction between the two domains later. For now let us outline their key aspects.
3. Biology: the first phenomenal domain
The biological aspects of autopoietic theory originate in experimental work on colour vision carried out in the early 1960s (see Maturana et al. 1960, 1968; Maturana, 1970). Subsequently, this led Maturana and Varela to reject the prevailing idea that the nervous system is open to information from the environment, and that human cognition mimics the symbol-based operation of computers.
In these experiments, it was found to be possible to identify direct correlations between the configuration of external wavelengths of light and activity on the retina of the eye, and between activity in the optic nerve and subjects’ verbal descriptions of colour. It was not possible however, to establish such a link between cellular activity on the retina and activity in the optic nerve. Hence the conclusion that:
.. we must stop thinking that the color of the objects we see is determined by the features of the light that we receive from the objects. Rather, we must concentrate on understanding that the experience of a color corresponds to a specific pattern of states of activity in the nervous system which its structure determines.” (Maturana and Varela, 1987: 22)
Figure 2: The relationship between the elements of an explanation of observing and the two “phenomenal domains”.
Later, Maturana was to summarize these findings in more detail. Firstly he proposes that no external stimulus acting on the nervous system can determine an organism’s experience of it. Experience corresponds to neuronal activity; it does not correspond to external perturbations. Thus, whereas popular opinion takes the view that the nervous system is open to environmental inputs, Maturana claims that it is closed, autonomous and circular. It operates, he says:
.. as a closed network of changing relations of interactions between components in which every change in relations of interactions between its components gives rise to further changes in relations of interactions between its components, and in which all takes place in a system of highly interconnected loops of unending recurrent circular processes of changing relations of interactions of different lengths and time constant.” (Maturana, 1988: 54).
Unpacking this further, it can be said that the sensory and effector surfaces of the nervous system have a dual character. They operate as elements of the organism and as elements of the nervous system. Yet their manner of operation depends upon whether they are operating in one or the other mode. Acting as components of the organism they operate in the interactions of the organism in its medium, but acting as components of the nervous system they operate in its closed dynamics. This means that the organism as a systemic totality interacts with the medium, but the nervous system as a component part does not. Hence the notion that the latter is closed and circular (see Maturana, 1997: 6).
An important consequence of this is the understanding that there is no “outside” to the internal components of the nervous system; there are only internal correlations of neuronal activity. Notions such as “inside” and “outside” require the existence of an observer who can see both and explain changes in one in terms of the other.
The nervous system, says Maturana, is an example of a “complex” or “composite” system, i.e. one that an observer decomposes to highlight not just the system itself, but also its component parts. All such systems, he suggests, are “structure-determined”. That is:
…everything that happens in them happens as a structural change determined… either in the course of their own internal dynamics or triggered but not specified by the circumstances of their interactions.” (Maturana, 1990: 13).
Logically, the circularity and closure of the nervous system means that it cannot work with representations. Cognition, therefore, cannot be taken to involve the brain processing symbols that stand for external elements, manipulating representations and then computing a response that is adequate in the light of prevailing circumstances. Although we humans frequently speak as if we know about the “real world” and can accurately perceive the “things” in it, this is a biological impossibility. Thus, instead of characterizing an object (e.g. “that tree is green”) according to its intrinsic characteristics (the tree is green), we do so on the basis of what happens to us through our experiences. Hence Maturana’ s claim that we exist through the physical space, not in the physical space.
Extending the key notion of circularity, Maturana (1993) sought to relitigate the age-old question of what is meant by the term “living”. Eventually this led him and Varela to claim that all living systems are “autopoietic” or “self-producing”, i.e. they are:
“networks of molecular production such that the molecules produced, through their interaction, generate the network that produced them and specify its extension” (Maturana, 1993).
On this view, life is constituted through a dynamic; through a particular “manner of relating” of molecules. The term autopoiesis describes the nature of this invariant or “organization”,which is common to all living systems. This invariant is conserved as long as the living system maintains its class identity. However, the “structure” of the living system (i.e. the manner in which autopoiesis is realized in practice) is subject to a myriad of different forms and changes constantly.
At first sight, the idea that living systems are structure-determined, that they have an autonomous autopoietic organization, and, where the nervous system – if there is one – is closed to information from the outside world, presents serious difficulties in accounting for the manifest adaptability of living systems. It makes it equally difficult to account for the distinctive evolutionary trajectory that each living system follows over time.
Conventionally, the environment is seen as the primary source of adaptation and change. Thus, in human beings, adaptation stems from the ability of the nervous system to construct and manipulate environmental representations and, on the basis of these, to act with intent. On this so-called “cognitivist” view, adaptation is very easy to explain. In contrast, autopoietic theory claims that the nervous system is not open to external information, indeed it regards this as biologically impossible. So, if adaptation is not about building representations of the “outside world” and then acting on these, what mechanism is involved?
In answering this question, we have to return to the inherent inter-connectedness between biology and human cognition. First, however, it is helpful to look at what autopoietic theory has to say about the main sources of change in living systems (see Fig. 3). Essentially there are three of these. Firstly, there is change that arises conditional upon the flow of molecules through the system. The circularity of autopoiesis pertains to a dynamic, i.e. to a peculiar manner in which component parts of a system relate to each other. It does not refer to the physical pattern of movement of specific molecules or components, as it would, for example, in the case of a tornado where molecules and particles do trace a circular pattern. Living systems are different in that the molecules themselves do not trace a circular pattern. Instead they enter the system, they participate in the process of self production, and they then die and/or leave the system. While this is happening, as molecules flow through the system, its structure changes, as long – of course – as the self-producing dynamic is conserved.
Figure 3: Sources of structural change in living systems.
Secondly, there is a change that results from internal dynamics. These can be incremental such as those that result from metabolic changes or from ageing, or they can be more drastic such as those that occur when the flow of molecules in dissipative structures sets up instabilities in and around systems that otherwise appear to be characterized by order.
The third, and – in the context of this discussion – most important source of change, occurs through a process known as “structural coupling”. Maturana employs this term to refer to situations in which it is possible to observe a congruence or “fit” between the living system and medium. When there are recurrent interactions between an organism and what an observer would regard as its environment, or between one organism and another, structure- determined changes occur in both. In other words, the two structures change congruently each according to its own structure-determinism. Through this process, the structure of the organism, at any point in time, becomes a record of previous interactions. Or, as Maturana (1988) puts it, the structure and operation of a nervous system always embodies ”… the behavioural present of the history of interactions of the system that it integrates…”.
As long as the system survives, i.e. as long as there is conservation of autopoiesis and adaptation, external interactions trigger structural change. Structural coupling then, is a key mechanism for change in a living system during its lifetime.
Whereas cognitivism tends to emphasize the response of the system to its environment, and posits a primarily one-way information flow, the idea of structural coupling implies a much more complex interdependent relationship. It says that when one system perturbs the other, two things can happen. Through spontaneous homeostatic processes the system can revert back to its original state. Alternatively, the structure of the system can change. However, this is not adaptation as cognitivism construes it. Where a change does occur in response to an environmental perturbation, it is because the particular change that occurs is a pre-existing feature, or one possible state, of the system’s structure. The perturbation only triggers it.
Turning now to cognition, structural coupling generated by the demands of autopoiesis plays the role that conventional wisdom attributes to having a representation of the world. Traditionally, as I have said, we have come to regard cognition as a process that has the brain manipulating representations of the external world. For organisms that possess a nervous system, this sounds eminently plausible. But what about organisms that are not endowed with a nervous system? How is it that they are often just as well-adapted to their medium? And does the absence of a nervous system mean that these organisms are not “cognitive” beings?
Of course it is possible to answer the first of these questions using cognitivist logic. Thus an observer might regard physical movement in a single cell unity such as an amoeba – as it surrounds a source of nourishment in its medium – in terms of the amoeba having somehow “perceived” its environment and having computed an appropriate response (Maturana and Varela, 1987: 147). Yet this is unsatisfactory since the absence of a nervous system and the inability to operate in language would seem to preclude the amoeba carrying out the necessary “perceive and compute” functions.
Despite this, under normal circumstances, the amoeba’s behaviour is adequate in its domain of existence. It is adapted to its medium, and one could say that the amoeba is a cognitive being that has the capacity to “know”. Obviously it does not “know” in a conventional sense, rather it “knows” because it is successfully coupled to its medium, something which is necessary for its conservation of autopoiesis.
Cognition then is inextricably linked to structural coupling which in turn is linked to biology. The physical movement of the amoeba does not involve perception. Instead there is a process in which changes in the chemical composition of the medium trigger changes at the sensory surface of the amoeba. This sets up an internal dynamic that results in the amoeba altering its position in relation to the food source in its medium.
It is this logic that leads to the conclusion that, in its most basic form, cognition is not a mentalistic phenomenon. Rather, it is effective action in a defined domain of existence (see Maturana and Varela, 1987: 29). This process involves the whole organism and is not limited to what might happen in the nervous system or brain if these exist. If an observer sees a unity behaving adequately in some defined domain then we can surmise that it “knows” relative to whatever criteria the observer applies in making the necessary judgment. In this sense, cognition equates to the whole process of living. The abstract thinking that goes on in human beings is but a special case of what is a much broader phenomenon (see Varela et al., 1991, for a detailed extension of this argument).
Maturana’s linking of cognition with structural coupling provides a distinctive take on the increasingly popular idea (see, for example, Heidegger, 1927; Schon, 1983; Feyerabend, 1987; Varela et al., 1991; Mingers, 1996; Introna, 1997) that cognition is an integral part of our normal everyday mindful and unmindful activity. According to this line of thought, action that looks like adaptation to an observer, and which the representationalist perspective would explain in mentalistic terms, is merely the system operating in a relationship of structural coupling with a medium.
The distinctive contribution of autopoietic theory then, is to allege that cognition has to do with the process of living which, in turn, is inextricably linked to structural coupling. This turns on its head reductionist conceptions of cognition that equate – for example – the mind with the brain and/or limit it to conscious thought. Instead it asserts that the brain is merely one of many structures through which the mind operates. It demonstrates quite clearly the systemic nature of autopoietic theory and the systemic, holistic nature of human cognition.
Such then are the basic distinctions that pertain to the first of the two domains that are central to Maturana’s systemic account of observing. To summarize, we can say that living systems are self-producing systems of molecular production that operate in a dynamic structural coupling with a medium. In the case of human beings, the basic biological character of the nervous system is critically important since it makes observing possible, and it has implications for the what and how of observing in specific instances However, it is in the second phenomenological domain, that of relations and interactions with a medium, where observing, and where behaviour more generally takes place.
4. Relations and interactions: the second phenomenal domain
Although the basic biological constitution of human beings, including the conception of cognition as being tied to the full process of life, is important when it comes to understanding how we operate as observers, it is in the second phenomenal domain: that of relations and interactions where observing actually takes place.
Maturana’s explanation of the development of observing in human beings extends the idea of a living system structurally coupled to a medium to that of structural coupling between two or more living systems. It is in the relational space between these two structurally coupled systems that the key mechanism that allows observing to take place occurs. This process is referred to as “languaging”.
The idea of languaging connotes a process in which the fundamental aspect is not the speech act itself (see, for example, Searle, 1983), but behaviour. Again this reveals the systemic interconnectedness of the components of autopoietic theory – observing (explaining etc.) depends upon language, language is rooted in behaviours and behaviour is an emergent property that arises out of the interactions between physiology and anatomy and the medium in which the system exists (see Fig. 4). This, of course, is a complex set of propositions, so let us examine in more detail what it involves.
Figure 4: Behaviour and language as relational phenomena.
In order to explain languaging, and hence to explain observing, is necessary to revisit the concept of structural coupling. According to Maturana, when we observe two or more entities “in language” with each other, what we see is a behavioural process. Using his terminology we see an initial “co-ordination of action” between two structurally coupled entities, followed by a further coordination of action. Consider the following: a parent in the company of strangers looks disapprovingly at her misbehaving child; the child immediately ceases to misbehave. Alternatively consider the owner of a dog whose whistle attracts the attention of a dog that brings it “to heel”. In both cases, there is an initial coordination of action: the child’s gaze falls upon the parent and the dog’s gaze falls upon the owner. Next both the child and the dog do something on the consequences of the original interaction. In both cases, there is a further coordination of the already coordinated actions.
The basic contention then, is that a “coordination of a coordination of action” constitutes the minimum operation that is involved in language (see Fig. 5). On this definition, it follows that a rudimentary form of languaging is possible even in organisms that do not possess a nervous system, and/or in those which do have a nervous system, but which are incapable of abstract thought in the manner of human beings.
Much learned animal behaviour such as the courtship, nest building and chick rearing behaviour of birds, or the “dance” of bees, is of this basic type (see Mingers 1995:78). However, to the extent that there is no abstract thought involved in these activities, and in order to differentiate it from languaging proper, Maturana refers to this as “linguistic” consensual behaviour.
However, where there is an advanced nervous system, the languaging potential of the organism is increased dramatically. In simple terms, the intrusion of a nervous system severs the direct link between sensory and motor surfaces of the organism. That having occurred, although the sensory surface can and does perturb the nervous system, the nervous system begins to interact with itself. Activity within the nervous system thus becomes the object of further activity, which becomes – ad infinitum – the object of further activity and so on. This process (see Mingers, 1995:73 –79, for a fuller description) provides the basis for a massive expansion in the cognitive capability of the nervous system. In human beings, this culminates in abstract thought.
Because we human beings have such a capacity, and because we live our lives completely immersed in the full complexity of language, it is hard for us to see that, in its minimum form,
Figure 5: The minimum operation of “languaging” (“that which happens when there are co-ordinations of co-ordinations of behaviour”)
languaging involves a simple coordination of an already coordinated set of actions. Thus we tend to describe language in semantic terms i.e. involving a transmission of information from one organism to another that embodies some meaning. However, when – in a relationship of structural coupling – one organism coordinates its actions with another and there is a further coordination, then we have the basic dynamic that culminates in languaging. Neither action is necessarily determined by meaning; rather they are a consequence of the dynamics of structural coupling.
Further recursions of this basic coordination of actions result in language becoming increasingly complex and sophisticated, especially, as I have just said, in organisms such as human beings that are endowed with a sophisticated nervous system. Here, in networks of structural coupling, human beings use words as linguistic distinctions to coordinate actions. Through this process objects become tokens for highly specific behavioural coordinations.
In his classes and public seminars, Maturana (see, for example, 1993) uses a number of examples to illustrate this process. The designation “taxi”, for instance, connotes the actions that are involved in carrying someone from one place to another, in a motor vehicle, in return for the payment of money. The designation “clock” connotes the various actions that are necessary to read the time. Accordingly, Maturana claims that even highly abstract entities such as “justice” or “democracy”, are anchored in concrete co-ordinations of behaviour.
Over time, having arisen through the aforementioned process, the anchoring of objects in actions tends to be forgotten. Thus: ”…objects take place as distinctions of distinctions that obscure the co-ordination of actions that these co-ordinate.” (Maturana, 1988: 47). At this point in time, taxis, clocks and all other objects – to all intents and purposes – become entities “in themselves”. This, of course, is how we tend to live them.
Once objects have arisen, other developments are possible. We can make distinctions of distinctions and this allows abstract concepts to arise. Meaning then arises as patterns of relationships among descriptions. And once we have objects, we can carry out the operation that allows us to refer to ourselves. Thus we can become self-aware. This whole process is shown in Fig. 6.
Fundamentally, because language is rooted in behaviour, which is a relational phenomenon, and because the agreements to symbolize behavioural coordinations through tokens occur in networks of structural coupling, language is a relational phenomenon. Even our so-called “inner world” of solitary reflection and consciousness arises determined by our existing structure, which is a record of previous interactions and structural couplings. It could even be said that which we describe as the mind is a social, not an anatomical or biological, phenomenon.
From a systemic perspective, the key point is that language cannot be reduced to speech acts and to abstract mentalistic processes. Although abstract processes are important to human beings, fundamentally language is relational and is anchored in behaviours. Although language might appear to us as being about symbols, fundamentally it is about doing. Entities and objects correspond to “doings”, and language is a flow of coordinations of coordinations of behaviours. And this is a process that arises in our network of structural couplings, i.e. in our relationships with other people. As we coordinate our behaviours in different ways, as we make new distinctions, and as we come up with new tokens, we are continually weaving linguistic networks with other people. And languaging, says Maturana, is what humans do. It is the total immersion in language, as a manner of living together, that constitutes the human being as the class of living system that it is. Human beings, he claims, are not basically different from other animals, ”…all that is peculiar to us is that we live in conversations.” (Maturana, 1993: 47).
Figure 6: How objects arise through languaging.
In summary, combining the two key components involved in Maturana’s explanation of observing creates the understanding that when human beings observe phenomena they do so as living systems first and as languaging beings second. As living systems we are two things: firstly we are an aggregation of an infinitely large number of individual autopoietic systems; secondly we are structure-determined entities that exist in a relationship of structural coupling with a medium. As human beings we are living systems that live our lives totally immersed in the social phenomenon which is language. All of this happens whether we like it or not. It is an ever-present backdrop that provides the context for all of what we do as observers.
Having provided a description of the biological and social basis of observing, we can now look in a bit more detail at how these two domains interact.
5. The relationship between the two domains
The precise nature of the “relatedness” between the two key phenomenal domains is another aspect of autopoietic theory’s systemicity that is not often acknowledged, or I submit, fully understood. Key to this is Maturana’s insistence that various phenomena be understood in terms that pertain to it and not in terms that pertain to some other component or domain. This is critically important when it comes to understanding the process of observation and in understanding human behaviour more generally. Observing, as we have seen, happens through the mechanism of language. Logically then, since language is a relational phenomenon it cannot be taken to be a process through which the brain manipulates symbols, nor is it reducible to the operation of the nervous system. The same can be said about observing activities that take place through language, for example thinking and explaining. These need to be explained in relational terms and not biologically.
From this we can further deduce that since “objects” arise in language, notions such as “self”, “the mind”, “personality” etc. are also relational. As such they arise in, and vary according to the relational circumstances of the moment. They are not inscribed in the bodyhood of the individual concerned. Remove the relational context and to all intents and purposes they do not exist. All that is left is a de-humanized biological entity.
On this logic behaviour too is something that arises out of the interaction between the biological entity and the medium. Consider, for example, the nature of the “behaviour” or “act” that results from an anatomically generated movement in a particular direction of someone’s arm. This depends wholly upon the medium with which the arm interacts and the prevailing relational circumstances. Identical anatomical movements can generate very different “behaviours” depending upon the physical and social context of its interaction with the medium. Just as a nervous system is necessary for speech acts to take place but language is not explainable simply through reference to the nervous system, so too are anatomy and physiology necessary for “behaviour” to occur, but it is the interaction with the medium that makes a behavioural act “what it is”.
This illustrates one of the key features of autopoietic theory. Although the biological and relational domains are separate and independent phenomenal domains, they are nonetheless inextricably intertwined as components of a larger systemic whole.
”…bodyhood and manner of operating as a totality are intrinsically dynamically interlaced; so that none is possible without the other, and both modulate each other in the flow of living. The body becomes according to the manner (in which) the living system operates as a whole, and the manner (in which) the organism operates as a whole depends on the way (in which) the bodyhood operates.” (Maturana, 1997: 2).
It follows then that when someone has any sort of relational encounter – when they reflect on an issue (in isolation or with others), when they participate in a conversation, they do so with a built-in predisposition to act in a particular way. The biological structure of the system contains a record of past interactions and the physiological and emotional predispositions of the moment delimit the range of possibilities. However, what happens during that relational encounter depends upon what happens in the relational context. As a conversation proceeds, as its flow of languaging and emotioning unfolds, people see and do things differently; as someone reflects upon an issue their biological processes and mood impact upon what they do.
Biology then impacts upon but does not determine what happens in the relational domain. At the same time relational encounters impact upon but do not directly imprint themselves upon biology. At one extreme, an acrimonious exchange between two people may increase the blood pressure of the people concerned. Longer term, if the exchanges are repeated it could trigger cardio-vascular disease; one of the participants might even suffer a heart attack and die. At the other extreme, using various meditation techniques and through processes of “self talk”, someone might learn how to reduce muscular and nervous tension to good longterm effect. In both, there is a structural transformation of the system in line with changes in relational circumstances and in the medium itself. Indeed this is the basis for evolutionary development and learning (see Winograd and Flores, 1987: 44-47; Maturana, 1988: 74). However, such changes are always subject to the system’s structure determinism.
6. From biology and social relations to phenomenology and the derivative epistemology
The issue here is how the epistemology of autopoietic theory, which again is often handled as a separate discrete topic or aspect, arises out of and relates to the biological and social components of the work as a whole. Fundamentally, for both Maturana and Varela, “knowing” is a biological issue since, as we have seen, cognition is inextricably tied to the process of living and structural coupling between the organism and the medium. Varela et al.’s (1991) concept of enaction or embodied cognition develops this basic idea further. Knowledge, it is argued, is constituted in our actions. As an individual confronts new situations various experiences are gained through thinking, sensing and, importantly, moving. This means that the way we experience (and “bring forth”) the world, is very much an active construction involving the whole body. Effective action depends upon having a body with various sensorimotor and orienting capacities that allow an agent to act, perceive and sense in distinctive ways.
The most fundamental point then is that cognition is an integral part of normal everyday activity. As Mingers (1994:109) rightly puts it, cognition “is not detached cogitation, but situated practical action”.
Epistemology however, as the term is commonly understood, refers to the grounding and underpinnings of knowledge that, in human beings, emanate from more conventional language-based processes of understanding and explaining the world. For Maturana, what human beings “know” about their worlds depends moment by moment upon the precise nature of whatever “domain of explanation” is operational for that individual. A domain of explanation is an explicit or, more commonly perhaps, an implicit frame of reference that is defined at the very moment in which an observer, actively or subconsciously, distinguishes a particular area of experience, and then explains it As I have already said, such domains are constituted through the particular criteria that the observer applies in his/her listening in accepting a generative statement relevant to the area of experience as valid.
Because there are as many domains of explanation as there are generative processes and validity criteria, it follows that there is a “multiverse” of equally legitimate realities. Because of the closed circular nature of the nervous system and because of their structure determinism human beings cannot know in a transcendental sense, and there is therefore no single “universe”. Again it is important to see the inextricable linkage between the various components – in this case between biology and processes of knowing.
Maturana’s notion of “conversation” is central in further understanding the linkage between the biological and the social on the one hand and the process of knowing on the other hand. “Conversation”, when used in this context, refers to a dynamic braiding of “languaging”, which we have already discussed, and “emotioning” which is construed as “bodily dispositions for actions” (see 1988: 42).
Importantly, as conversations flow through an interweaving of distinctions and emotions, experiences and explanations alter (see Fig. 7). While the structure of the observer imposes limits on what is and what is not possible, it is the nature of the conversation (with self or other), and its flow, that determines the outcome.
Thus as people shift from one emotional state to another, changes take place in the kinds of things that they will and will not do. People behave differently, they see differently, and, importantly, they describe and interpret things differently according to the emotion in which they do these things. Moreover, languaging and emotioning are braided, each process affecting the other. This can be seen in everyday conversations where particular distinctions (words, comments, categories, attributions etc.) can invoke an emotional response from those to whom they are directed. Equally distinctions used often reflect the emotion of the moment. Yet again this provides an illustration of the systemicity of autopoietic theory. The distinctions that we use to explain and understand our worlds are never isolated from bodily processes on the one hand and from social processes on the other.
Figure 7: Explanations and notions of ‘what is real’ and what actions are ‘legitimate’ as a function of the interplay between the structure of the observer and changing flow of ‘conversation’.
Maturana uses the term “consensual domain” to describe these networks of structural coupling that are the site for conversations. In these contexts people learn their emotioning and their languaging with people to whom they are structurally coupled, and, through recurrent interactions, structural patterns of these become conserved.
The picture is further complicated on account of the bodyhood of the observer being at the intersection of not one but many different conversations within and across different structural couplings. Each one of these has its own braided flow of distinctions and emotions that have been learned through recurrent interactions over time and which alter subject to the dynamic flow of the conversation. This means that just as thoughts and descriptions within a single conversation are subject to change depending upon the flow of the conversation, they can also alter as the observer shifts from one conversation to another.
In this paper, I have attempted to plug a gap in the existing autopoietic theory literature by articulating and teasing out how its various components and aspects are inextricably interlinked. The overall effect of this is to remind us that fundamentally autopoietic theory is an holistic system of ideas and propositions; it is not a body of work that has aspects or components that can be treated separately from the whole and evaluated independently. Because of its breadth it is inevitable that some will choose to focus on some aspects and jettison others. This is acceptable providing they understand the wider ramifications of this. Over the years, numerous systems thinkers have cogently reminded us that components, when extracted from the system of which they are a part, lose not only their extrinsic value but also their intrinsic value (see, for example, Ackoff, 1994). This is a point that those who take aspects of autopoietic theory into far-flung areas of intellectual debate and practice might want to reflect on.
Ackoff R. L. (1994) Systems thinking and thinking systems, System Dynamics Review 10: 175–188.
Beer S. (1994) Preface to autopoiesis: The organisation of the living by Humberto Maturana and Franscisco Varela, In: Haenden R. & Leonard A., eds, How Many Grapes Went into the Wine: Stafford Beer on the Art and Science of Holistic Management (John Wiley and Sons, Chichester): 318–345.
Berman M. (1989) The roots of reality: Maturana and Varela’s the tree of knowledge, Journal of Humanistic Psychology 1989(2): 277–284.
Birch J. (1991) Re-inventing the already punctured wheel: Reflections on a seminar with Humberto Maturana, Journal of Family Therapy 13: 349–373. http://cepa.info/2798
Brocklesby J. & Mingers J. (2004) Autopoiesis and the theory of viable systems, Systems Research and Behavioral Science, forthcoming.
Checkland P. (1981) Systems Thinking, Systems Practice (John Wiley & Sons, Chichester)
Checkland P. & Scholes J. (1990) Soft Systems Methodology in Action (John Wiley & Sons, Chichester)
Dell P. (1982) Family therapy and the epistemology of Humberto Maturana, Family Therapy Networker 6(4) 39–41.
Efran J. & Lukens M. (1985) The world according to Humberto Maturana, Networker 9(3): 23–28, and 72–75.
Efran J., Lukens M., et al. (1990) Language, Structure, and Change – Frameworks of Meaning in Psychotherapy (W. W. Norton and Co. Inc., New York)
Espejo R. & Hamden R. (1989) The VSM – an on-going conversation, In: Espejo R. & Hamden R., eds, The VSM: Interpretations & applications of S Beer’s VSM (John Wiley & Sons, Chichester): 152–175.
Feyerabend P. (1987) Farewell to Reason (Verso, London)
Goolishian H. A. & Winderman L. (1988) Constructivism, autopoiesis and problem determined systems, Irish Journal of Psychology 9(1): 130–143.
Heidegger M. (1927) Being and Time (SCM, London)
Hoffman L. (1988) A constructivist position for family therapy, Irish Journal of Psychology 9(1): 110–129.
Introna L. D. (1997) Management, Information and Power (Macmillan Press Ltd, Basingstoke)
Kay R. (1997) Applying Autopoiesis to the Facilitation of Organisational Change. Australian and New Zealand Systems Conference, Brisbane.
Kay R. (2002) Autopoiesis and systems education: Implications for practice, International Journal of General Systems 31(5) 515–530.
Kay R. (2003) Organizations as self organising and sustaining systems: A complex and autopoietic systems perspective, International Journal of General Systems 32(5): 459–474.
Luhmann N. (1986) The autopoiesis of social systems, In: Geyer, R and Zoewen J. V. D., eds, Sociocybernetic Paradoxes (Sage, London): 23–51. http://cepa.info/2717
Luhmann N. (1990) The cognitive program of constructivism and a reality that remains unknown, In: Krohn W., Kuppers G. & Nowotny H., eds, Selforganization: Portrait of a Scientific Revolution (Kluwer, Dordrecht): 136–171. http://cepa.info/2712
Maturana H. (1970) Biology of cognition, In: Maturana H. & Varela, R, eds, Autopoiesis and Cognition (Reidel, Dordrecht) http://cepa.info/535
Maturana H. (1988) Reality: The search for objectivity or the quest for a compelling argument, Irish Journal of Psychology 9: 25–82. http://cepa.info/598
Maturana H. (1990) Science and daily life: The ontology of scientific explanations, In: Krohn W., Kuppers G. & Nowotny H., eds, Selforganization: Portrait of a Scientific Revolution (Kluwer, Dordrecht): 12–35. http://cepa.info/607
Maturana H. (1993) Videotaped Seminar on Language and Cognition. St. Kilda, Melbourne.
Maturana H. (1997) Metadesign, http://cepa.info/652
Maturana H. (2003) Autopoiesis, Structural Coupling and Cognition: A History of These and Other Notions in the Biology of Cognition http://www.matriztica.org/1290/article-28335.html
Maturana H., Lettvin J., et al. (1960) Anatomy and physiology of vision in the frog, The Journal of General Physiology 43: 129–175.
Maturana H., Uribe G., et al. (1968) A biological theory of relativistic color coding in the primate retina, Archives of Biological Medicine and Experiment 1: 1–30.
Maturana H. & Varela, R (1980) Autopoiesis and Cognition: The Realisation of the Living (Reidel, Dordrecht)
Maturana H. & Varela, R (1987) The Tree of Knowledge – The Biological Roots of Human Understanding (Shambala, Boston)
Mingers J. (1990) The philosophical implications of Maturana’s cognitive theories, Systems Practice 3(6): 569–584. http://cepa.info/2778
Mingers J. (1991) The cognitive theories of Maturana and Varela, Systems Practice 4(4): 319–338. http://cepa.info/2253
Mingers J. (1992a) Criticizing the phenomenological critique – autopoiesis and critical realism, Systems Practice 5(2): 173–180.
Mingers J. (1992b) The problems of social autopoiesis, International Journal of General Systems 21(2): 229–236. http://cepa.info/2811
Mingers J. (1996a) A comparison of Maturana’s autopoietic social theory and giddens theory of structuration, Systems Research 13(4) 469–482.
Mingers J. (1996b) Embodying information systems, In: Jones M., Orlikowski W., Walsham G. & Gross J. D., eds, Information Technology and Changes in Organizational Work (Chapman Hall, London): 128–135.
Mingers J. (1997) A critical evaluation of Maturana’s constructivist family therapy, Systems Practice 10(2) 137–151.
Morgan G. (1986) Images of Organization (Sage, Newbury Park)
Robb FE (1991) Accounting – a virtual autopoietic system, Systems Practice 3(4): 215–235.
Scholl D. (1983) The Reflective Practitioner: How Professionals Think in Action (Basic Books, New York)
Searle J. (1983) Intentionality: An Essay on the Philosophy of Mind (Cambridge University Press, Cambridge)
Teubner G., ed. (1984) Autopoiesis and the Law (de Gruyter, Berlin)
Varela F., Thompson E., et al. (1991) The Embodied Mind – Cognitive Science and Human Experience (MIT Press, Cambridge Mass)
Zeleny M. & Hufford C. (1992) The application of autopoiesis in systems analysis: Are social systems also autopoietic systems?, International Journal of General Systems 21(2): 145–160. http://cepa.info/1207
Found a mistake? Contact corrections/at/cepa.infoDownloaded from http://cepa.info/3759 on 2016-10-31 · Publication curated by Alexander Riegler