CEPA eprint 2682

Questions concerning the ontology of autopoiesis and the limits of its utility

Fleischaker G. R. (1992) Questions concerning the ontology of autopoiesis and the limits of its utility. International Journal of General Systems, 21(2): 131–141. Available at http://cepa.info/2682
Table of Contents
The larger questions
Autopoiesis, a recent proposal
The forum discussion
The rejoining authors’ initial concerns
Issues arising in the forum discussion
Regarding ‘domains’
A personal note and acknowledgments
The larger questions
There are several deep and age-old questions which underly science, philosophy, psychology, and social theory: What is the nature of the world? What is the nature of life? and, What is the nature of human beings in the world? These are fundamentally questions of ontology,[Note 1] that is, What sorts of things ‘are’ in the world? Equally ancient are the philosophical questions of ontogenesis, of how things ‘come to be’: What is the origin of the world? What is the origin of the living as one particular sort of thing in that world? and, What is the origin of society or social ordering among the living? And arising from these ontological questions is the fundamental question of epistemology: How do we, as human social beings, come to know the world in which we find ourselves? Both kinds of questions are evolutionary, that is, they are concerned not only with the nature of things in the world but how they come into being and how they change over time as well.
These foundational problems – of what things are in the natural world, how they come to be, how they evolve, and how we can know them – are all major issues in contemporary science. They are brought to the fore in the focal paper of this special forum issue, and an international group of experts from several different fields has been chosen to shed light on them in the ensuing forum discussion.
Autopoiesis, a recent proposal
Autopoiesis is a concept meant to define the organization of the living and thus to distinguish between living and non-living things in the world. Its creators meant to characterize the dynamic whole-system behavior and seeming autonomy of living systems and wanted at the same time to escape a reductionist description of the whole that was drawn from the behavior of its parts. Contrary to an eisegetic narrative elsewhere ascribing the concept to 19th-century writers,[Note 2] the term ‘autopoiesis’ has a short and only-recent history. That the concept of autopoiesis is both conceptually and historically tethered is a point made by several authors here [Fleischaker, Kenny, Swenson, and Mingers]: ‘autopoiesis’ was coined in 1972 by the Chilean biologist Humberto Maturana with Francisco Varela specifically for that peculiar organization claimed to be both necessary and sufficient to characterize “living systems as au-topoietic systems in the physical space.”[Note 3] The specific criteria of autopoiesis and its mechanism of production were published in 1974 as a protocol for identification – that is, in binary-decision form (like a botanical key) to determine categorically if a system under observation is or is not autopoietic.[Note 4]
While a definition of living systems as “autopoietic in the physical space” may seem straightforward enough, Maturana’s and Varela’s initial pronouncements (and Maturana’s later ones) leave ambiguous any definition of ‘autopoietic.’ There is, for example, their 1974 remark that “living systems belong to the class of autopoietic systems”[Note 5] and Maturana’s later statements that “There is no restriction on the space in which an autopoietic system may exist. The physical space in which living systems exist is only one of many… Otherwise, the properties of autopoietic systems as autopoietic systems must be isomorphic in every space.”[Note 6]
Two consequences of this ambiguity in the original definition of autopoiesis arise in the present debate. First, statements such as Maturana’s and Varela’s above serve to reify and proliferate ‘space’(s) proper. What are the ‘many’ spaces to which they refer? What can they mean by ‘space’ if not physical? And if autopoietic systems can exist in non-physical space(s), what epistemological value can be claimed for autopoiesis as distinctive of the organization of the living? If autopoiesis were realizeable in any or all of these non-specified but non-physical spaces, the living would be distinct not by virtue of its autopoietic organization but because of its physical/molecular substrate in which that organization is enacted [Fleischaker, Geyer, Swenson. and Mingers].
Second, in categorizing autopoietic systems as a ‘class’ and in denying “restriction on the space in which an autopoietic system may exist,” Maturana’s statements advance the claim not only that ‘autopoiesis’ transcends ‘the physical space in which living systems exist’ but that ‘autopoiesis’ transcends any embodying substrate whatsoever (or “is miraculously decoupled from the physical world,” as Swenson puts it. A linguistic analogy[Note 7] is the claim not only that poetry transcends its realization in a particular language (in Spanish, Persian, or Greek, for example) but that poetry transcends language. These two claims are analogous in having an ontology that is both dualist and idealist: the organizational schema, either poetic or autopoietic, has logical and temporal priority over the component substrates in which the schema may be realizeable. Such an ontology presents a serious logical difficulty for the realization of autopoiesis as Maturana and Varela originally defined it: if autopoietic organization transcends both physical and non-physical substrates while its criteria are explicitly materialist and mechanist, autopoiesis is impossible to realize in a nonphysical substrate [Fleischaker]. If the criteria of autopoiesis have no force beyond a physical substrate, extending autopoiesis to non-physical ‘space’ is metaphoric only and has no literal meaning [Kenny, Mingers].
The forum discussion
In discussing such central scientific and philosophic questions, this International Journal of General Systems forum joins a discourse that is both wider and older than the systems sciences. The volume’s title specifies the forum’s contents without subtlety [AUTOPOIESIS – A DEBATE: CONTROVERSY OVER PHYSICAL, BIO-LOGICAL, AND SOCIAL SYSTEMS]. Logistically, the debate was coordinated via the postal service in two full rounds of discussion between Zelený and Hufford as focal authors and seven rejoining authors. Each manuscript was edited and revised before entering its round – thus the completed focal paper by Zelený and Hufford was mailed to each rejoining author, all completed rejoinders were mailed to the focal authors for their response, and the completed response was mailed to each rejoining author for counterresponse. Since rejoining authors were sent only the papers by Zelený and Hufford and did not see any of the other papers (this includes the editor who wrote and edited her own before reading other manuscripts of rejoinder), there is no discussion among rejoining authors.
The volume’s format follows this double round of forum discussion: the focal paper by Zelený and Hufford, seven papers of rejoinder, the response by Zelený and Hufford to rejoinders, and short counterresponses from the seven rejoining authors. The sequence of individual papers in rejoinder and counterresponse [and of reference to their authors throughout this introduction] follows the order in which each is discussed in Zelený and Hufford’s response paper.
All forum participants are active in the discourse of system theory and/or cybernetics – that is to say, they are concerned more with system organization than with the stuff so organized and thus work with the expectation, as Geyer notes, “that completely different phenomena in completely different disciplines are subject to the same laws, can be explained by the same or similar theories.”[Note 8]
While those “different disciplines” involve several different system substrates, all forum participants have been occupied by the question of ‘what things are’ in his/ her respective field, several participants have pursued ‘how things come to be’, and a few are explicitly concerned with ‘how things evolve’. Indeed, in developing a general theory of evolution, one participant must consider all three questions as central to his work. The participants and their ‘substrates of choice’ are: Milan Zelený, human social systems and computer simulation of ‘self-organizing’ systems; Kevin Hufford, information systems and chemical systems; Gail Fleischaker, biological systems and epistemological systems; Felix Geyer and Gerard de Zeeuw, human social systems; Vincent Kenny, human psychosocial systems; Fenton Robb, human social organization; Rod Swenson, complex dynamic systems and the physics that constrains their ‘being’, their ‘coming into being’, and their evolution in the world; and John Mingers, human social systems and epistemological systems. Of the focal authors, Zelený has written on autopoiesis and spontaneous social order; Hufford has co-authored papers on osmotic chemical systems. Among the rejoining authors, Fleischaker has analyzed autopoiesis and examined its application to origins-of-life research; Geyer has developed alienation theory in the context of self-referencing and self-steering systems; de Zeeuw has initiated research and innovations in system/ user methodology; Kenny has compared the conceptual frameworks of autopoiesis and radical constructivism; Robb and Mingers have debated autopoiesis in the systems literature; Mingers has analyzed the cognitive theories of Maturana and Varela; and Swenson has formulated the law of maximum entropy production and framed evolution as a process of spontaneous ordering.
Closely contended here is the definition of autopoiesis and its possible substrates of enactment: physical? biological? social? Indeed, the stated object of the forum was to lay clear the ground for different interpretations of autopoiesis and, consequently, for the limits of its possible application.
In their focal and response papers, Zelený and Hufford describe three different kinds of system examples as autopoietic – biological (“cell”), chemical (“osmotic system”), and social (“the human family”). They set forth and then deregulate the autopoietic criteria (viz., the “topological boundary that has been necessary to describe an autopoietic system, … may not necessarily exist in a physical form in other types of systems, e.g., in social systems”[Note 9] ) so as to generalize ‘autopoietic’ to ‘spontaneous’ social systems: thus “individual components, especially humans, are fully realized only in spontaneous (i.e., autopoietic) social systems.”[Note 10]
The wide range of issues raised in Zelený and Hufford’s focal discussion required that debate participants come from a range of discourse equally as wide. Rejoining authors were chosen for their expertise in these several focal areas and as a group they present what Hufford appreciates as “a very interesting cross-section of disciplines”[Note 11] : philosophy of biology, sociology, social science methodology, psychosocial analysis, sociocybernetics, evolutionary theory, and practical systems philosophy.
The rejoining authors’ initial concerns
The concerns which the rejoining authors bring to the discussion fall roughly (and overlappingly) into three areas as follow.
1) The logic of discourse and its defining terms
While Zelený and Hufford wanted comments to be elicited by the “clarity, simplicity, and importance of the premises and conclusions” of their focal paper,[Note 12] the rejoining authors are nearly unanimous [Fleischaker, Geyer, de Zeeuw, Kenny, Swenson, Mingers] in bemoaning that paper’s lack of clarity and consistent definition of terms upon which any coherent argument must depend. All rejoining authors offer further or alternative distinctions and clarifications, definitions and re-definitions of contentious terms. Several [Fleischaker, de Zeeuw, Swenson, Mingers] point to propositional and relational terms as crucial for determining what constitutes the systems under debate.
In the addendum to Geyer’s paper, de Zeeuw raises ontological questions concerning the different kinds of systems to which Zelený and Hufford refer – asking whether or not they are entities (“have ‘thinghood’”) and so can be distinguished in the real world apart from the system observer’s act of distinguishing them – since only systems-as-entities can be attributed system properties. He points out that there can be no easy resolution of those questions as they concern social systems because the social entity being distinguished and the observer who distinguishes it are so intricately intermeshed [a point Kenny addresses in more detail as it concerns the family-as-system in the therapeutic context].
Swenson analyzes ‘autopoiesis’ – as originally defined by its coining authors and as increasingly conceptually burdened in being equated with ‘self-production’ and ‘self-organization’ – and rejects the concept as inappropriate to generalize to spontaneous-order-production across system substrates. For an alternative concept that is appropriate to such generalization, Swenson offers an unburdened term (“autocata-kinetics”), discusses its principles, argues why it is generalizeable, and provides its historical documentation.
Fleischaker, Geyer, and Mingers all analyze Varela-et al’s six-point key (the formal criteria of ‘autopoiesis’) and its application to the example systems of the focal paper, and all three attempt to clarify crucial terms of the discussion, e.g., “social,” “social system,” “human family.” Fleischaker argues for limitations on the coining authors’ definition of ‘autopoiesis’ so as to render it a categorical identification of living systems that is both universal and minimal. She then turns to the criteria as they are applied keypoint-by-keypoint to both the osmotic and the human-family systems, finding that neither system meets the autopoietic test in practice, and concluding that non-physical systems must fail to do so in principle. Mingers focuses on the human-family system and on the general problem of social autopoiesis, referring to the different approaches taken to that problem by Maturana, Varela, and Luhmann. He acknowledges the difficulty of a coherent description of social autopoiesis (but “does not rule out” its possibility), points briefly to the ontological and epistemological implications of Maturana’s theories, and voices concern that those theories are severely limited in social analysis since they can apply only to small and self-validating groups.
2) Social systems, particularly human family systems
Geyer traces out Luhmann’s arguments to widen autopoiesis to social and psychic systems as meaning-using systems (distinct from living systems as component-producing systems) and provides several recent research examples of self-observing and self-referring systems in the social sciences, e.g., health care systems, studies in research methodology. [Compare Geyer/Luhmann’s use of ‘communication’ as the basis of social-system production to Kenny’s regarding family systems and Robb’s concerning supra-human systems. ]
Kenny argues against extending autopoiesis to “third-order systems (families, clubs, businesses, nations)” on two grounds. His first argument, drawn from psychother-apeutic praxis, is rooted in the problematic issue of ‘boundary’ thus: in the criteria of autopoiesis an observable system boundary is required as an ‘objective’ part of the system entity, yet in family-therapy practice both ‘boundary’ and ‘family’ are observer-dependent and not ‘objective’ entities. His second argument, drawn from ethical praxis, is rooted in the value of human individuality, autonomy, integrity: because the components of an autopoietic system are of necessity non-autonomous entities, a third-order autopoietic system must, of the same necessity, be either a ‘parasocial system’ (which denies the autonomy of its individual members as irrelevant to the system) or a dictatorship or totalitarian regime (for which individuals-with-autonomy are a decided threat) [Fleischaker also notes that mechanistic (autopoietic) social systems would depend entirely upon the individual’s “mindless entrainment”; Robb warns that autopoiesis in social systems would mean “human lifetime … entirely committed to compliance”; Swenson scores autopoiesis as having “abhorrent social consequences” on its epistemological ground; and Mingers questions whether Maturana’s theories are competent to deal with “wider social phenomena such as power, and inequalities based on sex, race, and class”].
3) Spontaneous ordering
Both Swenson and Robb begin with a sympathy for what they take as Zelenji and Hufford’s intent to generalize the spontaneous ordering of real-world entities across different system substrates. Swenson holds that spontaneous order production is an inherent property of the energetic/physical world. In spelling out the laws and constraints by which spontaneous order arises in the world, Swenson propounds a principle of natural law (showing “why the world is in the order-production business”) that is generalizeable to spontaneous order-production as a behavior in all substrate systems – including non-living systems, living systems, cultural systems, and the single global system (a population-of-one), Robb pursues Swenson’s law of maximum entropy production to explain the emergence of ‘supra- human systems’ as autopoietic systems, “self-defining and real-world entities” subject to natural law. ‘Supra-human systems’ are exemplified by shared (but non-unitary) overriding systems of values, beliefs, sentiments and behavior, social currents, public opinions, the Durkheirnian ‘conscience collective’, cultural preoccupations or Weltanschauungen – and, presumably, human-action systems such as spontaneous mob action. While constituted by the energy of human activity in communications and conversation, ‘supra-human systems’ are of a different natural order from the human interactions in which they emerge. [Compare Robb’s treatment of shared language, conversations, and compliances about medical matters (a supra-human “system of belief which categorizes the properties of ‘ill people’”) to Kenny’s discussion of the observer-community’s role in defining ‘patients’ and to Mingers’ reference to Foucault’s ideas concerning the social construction of ‘madness’.} The boundaries of supra-human systems lie thus beyond human observation, and their behavior beyond human control. Robb urges general-systems study of such entities so that, heeding their potential threat to mankind and their necessarily accelerating emergence, we may learn how to predict and anticipate some of their effects on human life-time and life-energy.
Issues arising in the forum discussion
While the significant concerns brought to the forum by each author are quite specific, there are several general and far-reaching issues which emerge from the discussionat-large – among them the problem of system ‘wholes’ and ‘parts’ and the related problem of system constraints.
System ‘wholes’ and ‘parts’
The issue of ‘wholes’ and ‘parts’ arises as an ontological problem at the. outset of any system analysis, that is, in the determination of what ‘the thing’ is that’s being analyzed: What constitutes the system as a whole? and, What are the system’s component parts? And with the ontological problem arises a whole set of epistemological questions as well: Where are distinctions drawn between ‘parts’ and ‘wholes’? By what criteria are any such distinctions drawn? What is the purpose (intention) of drawing those distinctions? and, What is the context in which they are drawn?
Both sorts of questions, the ontological and the epistemological, must be addressed critically in describing any system, regardless of its substrate. While answers to these questions are far from clearcut under any circumstance, the issues are especially nettlesome as they concern social systems. Defining ‘social systems’ is fraught with special problems, as its numerous and contradictory attempts in the literature of the social sciences can quickly attest.
Zelený and Hufford press a non-conventional definition of ‘social systems’ to include “all autopoietic” systems (this volume, p. 147, p. 156). Because the focal authors include both living and non-living among autopoietic systems, they mean ‘social systems’ to include not only living systems but certain non-living systems as well – referred to in their focal paper as ‘spontaneous social’ orders or systems, and in their response paper as ‘cooperative systems’ or ‘coordinated structures’.
Rejoining authors raise the concern that thus defining ‘social systems’ places an emphasis on what is common across system levels at the expense of what is specific and peculiar to each level. Such a definition then means a loss or denial of what distinguishes between systems, especially of what distinguishes the living from the non-living [Geyer, Swenson], or it gives the term ‘social’ such a wide meaning that it loses any value of distinction whatsoever [Fleischaker, Mingers]. That is, it is not at,all clear what is characterized as ‘social’ thereby, as several of the rejoining authors inquire: all collective phenomena? [Fleischaker, Mingers], spontaneous ordering? [Swenson], spontaneous systems as distinct from engineered systems? [Kenny, Mingers], cooperative behaviors toward a common end? [Geyer, Kenny], voluntary choice of behaviors? [Fleischaker, Geyer, Mingers].
Yet even where all discussants speak of the more conventional notion of ‘social’ (i.e., that members of social systems are biological entities), defining ‘social systems’ per se is still not without problems. To what does ‘social system’ refer, after all: To a specific cast of biological actors (members of a swarm, family, troop, or tribe)? To a particular set of relationships among those biological actors (roles played out within swarm, family, troop, or tribe)? To the second-order narrative within which those roles are made normative (the human observer’s unifying system description of any perceived roles within swarm, family, troop, or tribe)? In practice, of course, the phrase ‘social system’ refers to all three, without distinguishing among the three and without articulating which one is at the moment intended [that the three are system levels is a point made by Fleischaker, Geyer, Kenny, and Mingers].
Among the forum’s rejoining authors, the phrase ‘social system’ refers usually (but not exhaustively) to human social systems. In the literature of cybernetics and systems theory, the definition of ‘human social system’ in general and ‘the human family’ in particular is a matter of considerable debate, as the ensuing papers make clear. While at first glance it may seem strange that human families are considered at all under the rubric of ‘cybernetics’[Note 13] (since that term is often used nearly exclusively to describe artefactual or technological devices), the ‘systems’ and ‘cybernetic’ approaches have provided new perspectives for the social sciences where, in emphasizing system levels and interdependence of the whole and its parts, they allow different definitions of social systems and draw different kinds of distinctions within certain social systems.
The forum’s foundational questions are specifically addressed in the field of family therapy, for instance, where various models from general system theory and cybernetics have held sway, each in its turn defining an entire school. A recent history of that field[Note 14] outlines a typology of models developed from general system theory (distinguishing individual parts from the family-entity as a whole), from cybernetics (distinguishing a stable pattern across the family-entity as a whole), from ‘autonomous systems theory’ (Blount’s term; distinguishing the family-whole as a system that defines and maintains its parts), and, from second-order-cybernetics and second-order-autonomous-systems, models centering on system punctuation and the role of the observer and her/his position relative to the family system under observation. At issue in the systemic therapeutic view in general is what ‘family systems’ are [a context and issue raised both in Geyer’s and Kenny’s papers of rejoinder that follow]. Additional issues in the second-order models concern ‘family-plus-therapist systems’, i.e., in what ways therapists act as system participants or act in a position that is ‘meta’ to the family system and, hence, to its pattern and natural order.
System constraints
The issue of constraints makes itself felt in identifying what things ‘are’ and how they ‘come to be’ – that is, by asking, What is the substrate of ‘the system’ under analysis? and, What are the operating principles that result in the system’s formation? Answers to those questions turn on the distinctions between ‘laws’ and ‘rules’.
In the course of discussing conventional human social systems, the question arises whether ‘social systems’ are themselves natural (real-world) entities or symbolic (constructed) systems. That dichotomous question is variously answered as follows: Robb and sometimes Zelený and Hufford hold that social systems are natural entities; Fleischaker, Geyer, de Zeeuw, Kenny, Mingers, and sometimes Zelený and Hufford hold that they are symbolic entities; retaining the distinction but rejecting the di-chotomy, Swenson holds that cultural systems are law-governed systems – in his words, autocatakinetic systems that use non-deterministic symbolic (rate-independent) constraints in their spontaneous ordering.
In holding that human social systems are not of a natural kind, one rejoiner points to ‘the family’ as observer-dependent [Kenny] and others point to its relative cultural definition dependent upon social convention [Fleischaker, Geyer, de Zeeuw, Kenny, and Mingers]. Several authors distinguish social systems from mechanistic systems on the basis of their different component interactions and kinds of constraints. Geyer, for example, remarks that interactions within social systems “are definitely not determined, although admittedly some interactions are more likely than others.”[Note 15] Kenny focuses on this non-determined behavior as characteristic of “genuine social systems,” contrasting it both to institutionalized manipulation of behavior in “parasocial systems” (e.g., concentration camps, asylums), and to mechanistic behavior in “first- order autopoietic systems” where components are “irrelevant beyond their capacity to realize the autopoietic organization.”[Note 16]
Mingers applies Wittgenstein’s philosophic/linguistic discussion of ‘rules’[Note 17] to social systems, contrasting rule-governed behavior to law-governed behavior: “The essence of rule-governed as opposed to cause-and-effect behavior is that a rule can be broken, or performed in right or wrong ways.” “[T]he interaction of the components of social systems is fundamentally different from that of physico-chemical systems. The. former are governed by rules which may be disobeyed, the latter by causal laws.”[Note 18] Thus, whether social rules are adopted by popular assent or imposed by dictatorial decree, society’s members can in principle disobey them – although under tyrannical rule, disobedience can result in punishment, exile, or death [Fleischaker, Kenny].[Note 19] Fleischaker cites Pattee’s distinction between ‘laws’ and ‘rules’[Note 20] to emphasize, “While it is clear that human beings are capable of following rules, human family members are not inexorably lawful entities like the components of mechanistic systems.”[Note 21]
Swenson points to two different kinds of constraints involving processes that are either ‘rate-dependent’ or ‘rate-independent’.’ He distinguishes among system levels according to their constraints and, in his theory of general evolution, formulates ‘spontaneous ordering’ as the emergence of those new levels – the coming into being of a new environment, of a new set of constraints. Thus it is the different set of constraints at each emergent level that defines and distinguishes non-living, living, and cultural (symbol-using) systems.
Because emergence marks a discontinuity of constraints between levels, no new level is reducible to the level from which it emerges. A considerable systems literature[Note 23] ,[Note 24] , of course, lays out the historical groundwork for arguing the ontological irreducibility of system levels, the decomposability of system wholes to their parts based on the “all-or-none nature of the circular relations that define their being: either they exist or they do not. Nonlinear behavior is primitive to its kind.”[Note 25] And living systems are no exception to this irreducibility [Fleischaker, Swenson].
Formal simulations and ‘artificial life’
Related to the ontological ambiguity that still attends autopoiesis from its first appearance in the literature[Note 26] is a logical contradiction framed by its earliest English publication: [Note 27] while on the one hand autopoiesis is defined as an irreducible circularity of component operations, a formal algorithmic model of those very operations is given on the other. Historically that opened up a reductionist route to computer simulation of autopoiesis, a route pursued by Zelený in his previous work and restated by Zelený and Hufford in this forum’s focal paper: “… autopoiesis (autopoietic organization) can be studied by postulating each component as a separate entity and tracing its behavior through cellular automata types of computer simulation” (p. 157). Yet whereas component behaviors within living systems are constrained by universal natural laws [Fleischaker, Robb, Swenson, Mingers], component behaviors traced through automata are determined by the programmer in formal program rules. For this fact alone, if autopoietic organization could be produced by computer simulation, autopoiesis would be inadequate as a definition of the living. Because they are computer simulations, cellular automata and ‘artificial life’ require an artefactor (designer, programmer). They are necessarily artefacts – ‘Newtonian mechanisms’, totally determined in the specifications devised by their artefactor – and not the real thing [see Swenson’s discussion of the Newtonian worldview as artefactual and its “smuggling in” of an external maker].
Regarding ‘domains’
Domains are discussed either directly or indirectly by all participants throughout the debate [although Swenson does not refer to domains as such]. In some discussion, however, the way in which the term ‘domain’ is used elides its two very different meanings: (1) as it refers to system levels, that is, to scalar distinctions within the natural world, e.g., to ‘the molecular domain’ as distinct from ‘the planetary domain’[Note 28] ; and (2) as it refers to areas of discourse or inquiry, that is, to arbitrary categories of human discussion or jurisdiction, e.g., to ‘the domain of literature’ as distinct from ‘the domain of fine arts’. As a question of system levels, the issue of domains lies at the heart of the present debate, namely, at the ontological basis of autopoiesis and the substrates of its possible enactment.
The focal authors’ treatment of ‘domain’ is taken up by several of the rejoining authors [Fleischaker, Kenny, Robb, Mingers] who charge Zelený and Hufford with confounding natural domains. In their response paper, the focal authors argue against those charges and assert the following:
There are no ‘domains’ in Nature, only in the minds of men and women” (p. 240);“…, there is only the physical domain” (p. 250); and“The only new and distinct domain is that of human creation (or construction), i.e., symbolic, conceptual, and labeling systems … such as mathematics or engineering” (p. 251).
Each assertion is the proclamation of an ontological position: the first is nominalist (or idealist), the second is physicalist (or materialist), and the third is artefactualist. While each of these positions has been evinced in the traditional philosophic literature, Zelený and Hufford here proclaim simultaneously for all three.
From the debate’s outset nearly every rejoining author [Fleischaker, Geyer, de Zeeuw, Kenny, Swenson, Mingers] complains that the focal authors do not define or distinguish clearly the kinds of systems being analyzed – that in their comparative analysis of living systems, social systems, and autopoietic systems, Zelený and Hufford do not define living systems.(although ‘living’ is parenthetically equated to biological’); do not define social systems (although ‘social’ is used throughout the focal paper as variously inferring ‘human’ or ‘spontaneous’ or as having ‘communicating members’ or ‘components with individuality’, and defined in the response paper as “coordination (or harmony) of individual action achieved through communication among individuals” (p. 244)); and initially define autopoietic systems per the criteria of Varela et al. only to deny the criteria! requirement of ‘boundary’ soon thereafter.
Because our choice of terms reflects our parsing of the world – an ontological venture requiring the utmost care and clarity – it is imperative that terms and concepts be precisely defined in any work aimed at resolving the deep issues under discussion here.
It is my sincerest hope that this forum contributes in some measure to that end.
A personal note and acknowledgments
I personally know only one of the seven original participants of the forum discussion (I am the eighth) – although, as it turns out, I also know the ninth participant, sub-sequently brought into the forum through just that kind of spontaneous behavior not infrequently exhibited by such dynamic systems. Yet in the inevitably back-and-forth course of correspondence over manuscripts and editorial detail, I feel I have come to know my rejoining fellows to some degree. I thank them wholeheartedly for their thoughtful participation, good-humored responsiveness, and attentiveness to schedules. I can only hope we get to meet someday in person.
I want to proffer a special word of thanks to Richard Lewontin, my host at Harvard University’s Museum of Comparative Zoology Laboratories for the past year and a half, without whose generosity and gracious hospitality this forum project could never have seen the light of day.
My contributions here (guest editor’s introduction, rejoinder, and counterresponse) are based on work supported by the National Science Foundation under Grant No. DIR-8903206 awarded in 1989.
I use the term ‘ontology’ and ‘ontological’ throughout in a Quinean sense, that is, to refer to the
M. Zelený, “Autopoiesis: A paradigm lost?” In Autopoiesis, Dissipative Structures and Spontaneous Social Orders, edited by M. Zelený, Wesiview Press, Boulder, CO, 1980. pp. 3-43.
H. R. Maturana, Introduction to ‘Biology of Cognition’ 1970. In Autopoiesis and Cognition – The Realization of the Living, edited by H. R. Maturana and F. 1. Varela, D. Reidel Publishing Co., Dordrecht. Holland, 1980, p. xvii – xix.
Varela et al., 1974, pp. 192-193.
Varela el al., 1974, p. 189.
H. R. Maturana. “Autopoiesis.” In Autopoiesis: A Theory of Living Organization, edited by M. Zelený, North-Holland, New York, 1981, pp. 22-23.
I am most grateful to Francisco Varela for this fruitful analogy.
F. Geyer, “Autopoiesis and social systems-1” this volume, p. 175].
M. Zelený & K. D. Hufford. “The ordering of the unknown by causing it to order itself” [this volume, p. 248].
K. D. Hufford. personal communication. 28 November 1990.
M. Zelený & K. D. Hufford, The ordering of the unknown by causing it to order itself” [this volume, p. 239).
Systems are defined as ‘cybernetic’ in having inherent circular feedback processes, regardless of particular system substrate. Thus, ‘cybernetic’ defines a system type across substrates or levels.
A. Blount, “Introduction to Special Issue on Organizations.” Journal of Strategic and Systemic Therapies, 4, No. 4, 1986, pp. 1-7.
F. Geyer. “Autopoiesis and social systems-1” [this volume, p. 177].
V. Kenny, “On the subject of autopoiesis and its boundaries: Does the subject matter?” [this volume, p. 192].
L. Wittgenstein. Philosophical Investigations. Macmillan Publishing Co., Inc.. New York. 1953.
J. Mingers, “The problems of social autopoiesis” [this volume, p. 231]; “Reply to Zelený and Hufford” [this volume, p. 271].
Compare this discussion of ‘rules’ to the focal authors’ idiosyncratic use of the same term: “Rules can arise spontaneously, they are purposeless, and they interact for no goals: they coordinate action. ‘Rules of the game’ coordinate the conduct of the game but do not bring the game forth; they do not engender it.” M. Zelený & K. D. Hufford, “The ordering of the unknown by causing it to order itself” [this volume, pp. 242-243].
H. H. Pattee, “The complementarity principle in biological and social structures.” Journal of Social and Biological Structures. 1, 1978, pp. 191-200. It must be noted that Pattee’s whole intent in distinguishing between ‘laws of nature’ (as inexorable) and ‘rules of constraints’ (as arbitrary) is to demonstrate their complementarity as primitive of living systems and as required for explanatory descriptions at the biological and social levels. While Fleischaker takes Pattee’s distinction between ‘laws’ and ‘rules’, she does not take the consequent complementarity; indeed, she uses the distinction toward an opposite end, namely, to distinguish between the physical/biological and social levels.
G. R. Fleischaker. “‘Are osmotic or social systems autopoietic?’: A reply in the negative” [this volume, p. 170].
Swenson eschews the term ‘rule’ as the descriptor for the general case as Pattee distinguishes it, referring more crisply to “arbitrary rate-independent” constraints of which intentional human rules would be a special case.
P. Weiss, “The living system: Determinism stratified.” In Beyond Reductionism-New Perspectives in the Life Sciences. The Alpbach Symposium 1968, edited by A. Koestler and J. R. Smythies, Hutchinson of London, 1969, pp. 3-55.
L. von Bertalanffy, General System Theory-Foundations. Development. Applications. George Braziller, New York, 1968; “Chance or law.” In Beyond Reductionism-New Perspectives in the Life Sciences, The Alpbach Symposium 1968, edited by A. Koestler and J. R. Smythies, Hutchinson of London, 1969, pp. 56-84.
R. Swenson, “Emergent attractors and the law of maximum entropy production: Foundations to a theory of general evolution.” Systems Research, 6, No. 3, 1989, pp. 187-197.
H. R. Maturana & F. G. Varela, DE MAQUINAS Y SERES VIVOS - Una teoría sabre la organización biológica. Editorial Universitaria, S. A., Santiago, Chile, 1973 1972.
F. G. Varela, H. R. Maturana, & R. Uribe, “Autopoiesis: The organization of living systems, its characterization and a model.” BioSystems, 5, 1974, pp. 187-196.
Where one wants to speak very cleanly of system levels, and where the double-entendre of ‘domain’ makes that term infelicitous, I commend the use of ‘substrate’ in its place. This more precise language is patently adopted from Swenson’s rejoining argument that emergent order “depends upon the particular level-dependent substrate upon which the level-independent laws operate.” I have used this language in writing the present introduction, but must report, alas, 1 did not yet have it in hand in writing my own rejoinder to the debate’s focal paper.
Found a mistake? Contact corrections/at/cepa.infoDownloaded from http://cepa.info/2682 on 2016-05-13 · Publication curated by Alexander Riegler