Two Conflicting Theories: Turner and Harré

Brian S. Turner's sociological theory of human sociocultural life and Rom Harré's social psychological theory of the human ways of social, personal, and physical being are two prominent, congruent theories sharing the intent to recover a robust conception of human agency. Their congruency involves the conviction that the primary focus of investigation is 'the social' as patterned action rather than 'the social system' as social structures that determine the patterning of action. They agree that the concept of 'the social' must entail concepts of culture, person, and body; however, the two theories conflict in their endeavors to accomplish the recovery of human agency through the embodying of persons. Each theory is characterized by a different ordering of the concepts of 'person' and 'body.' This paper is about the ontologically conflicting ways in which the two theories embody persons.

     Turner's thesis is that in the beginning is the body, therefore, the person; thus, we arrive at his recovery of human agency: the effectiveness of persons in sociocultural life resides in the discourse-independent agency of the bodies of persons.

     Harré's thesis is that in the beginning is the person, therefore the body; thus, we arrive at his recovery of human agency: the effectiveness of persons in sociocultural life resides in the discursive agency of persons embodied.

     Given their commitment to the recovery of human agency against the challenge of classical mechanistic determinism, it is no surprise to learn that both Harré and Turner believe that the study of human life (as a life of human action) should be scientific. The problem of determinism and the recovery of human agency (in terms of a theory of embodied persons) is a problem of ontology, and we are to understand scientific theory as disguised ontology (Aronson, Harré and Way 1995).

     What is surprising is that Turner's recovery of human agency leads him to Merleau-Pontian phenomenology and the concept of the 'lived-body,' rather than leading him to a philosophy of science that provides a conception of agentive causation situating human agency in the natural world as one among other natural kinds of powerful particulars—what in Turner's own words would be called "effective agency."

     A helpful way to address the issue of differing conceptual solutions to the problem of determinism and embodying human agency is to place it within the framework of the general question of agency and structure. In this context, it can be seen as a specific problem of the proper location of agency with reference to the body, the person, and society. Our task is to assign causal powers to the right powerful particulars without committing the fallacy of 'agentifying' psychobiological or sociocultural structures. We need a conception of the right kind of determining factors that will provide a conception of the right kind of causation. We can then make principled judgments about assignments of agency, knowing when they are legitimate and when they are not.

     Given that it is clear that ubiquity determinism and agentive causation (rather than regularity determinism and correlated passive particulars) is the right kind of determinism (Bhaskar 1978 and 1994: 139; Harré 1986), it must be equally clear that it is only under the auspices of a referential policy realist theory of science (rather than a transcendentalist realist theory of science) that we possess a systematic development of causal powers theory, hence, the critical weapon necessary to make principled judgments for the proper location of agency with regard to the body, the person, or society (Varela and Harré 1996).

     Within the provisions of Turner's 'recovery of agency thesis' (that is, effective agency resides in the discourse-independent agency of embodied persons), I will argue that we are reasonably advised to favor Harré's 'recovery of agency thesis' (that is, effective agency resides in the discursive agency of embodied persons). Only then can we successfully meet the challenge of what Turner believes to be Foucault's concept of discourse determinism. To do so, I will accept Turner's suggestion that Foucault is a structuralist and a discourse determinist, although I suspect he is neither. It is sometimes easier to know what a scholar is not than what he or she is.

Turner's Vision: Beyond Nostalgic Social Theory

An appreciative examination of Turner's intellectual efforts to work out a program for a theory of embodied sociocultural action must be rooted in Turner's relationship to The Sociological Tradition (Nisbet 1966). Turner defines his program around the twin poles of sociology of knowledge and a profoundationalist social construction of reality. These poles are centered in humanistic Marxism within an integrated and updated Weberian and Parsonian theoretical orientation, allowing him to utilize both Foucault and Merleau-Ponty judiciously to introduce the topic of the body as a target and as a resource. An array of bodily activities—techniques, movements, and gestures—are thereby incorporated into the structure of a social theory of human action. Under the vigilant eye of a sociologically informed conception of action, Turner intends simultaneously to control for Foucault's discourse determinism and Merleau-Ponty's individualistic phenomenology.

     As Turner sees it, to realize his program for a theory of embodied sociocultural action, he must free himself from the serious limitations of the classical sociological tradition. He accomplishes this in a series of discussions in which he develops a very important critique of the endemic character of the classical tradition. His paper titled "The Rationalization of the Body: Reflections on Modernity and Discipline" (1987) provides a clear transition between his Marxist-centered book The Body and Society (1984: 38–59 and 227–51) and his later position in "Recent Developments in the Theory of the Body" (1991: 6–24) and his Weberian-centered book Regulating Bodies (1992: 1–28, 31–121, and 229–62).

     Since the classical sociological tradition inspires and informs nearly all twentieth-century sociological theory, Turner expands the boundaries of the tradition to include crucial sympathetic variations on Weberian sociological theory, namely, the critical theories of Adorno and Horkheimer and Foucaultian structuralism. He refers to the character of his suitably expanded classical tradition as "nostalgic social theory." Turner's stance is clear: he is against nostalgic social theory and in favor of going beyond it. He does this from a Weberian framework, moving through Foucaultian embodiment theory and the Merleau-Pontian phenomenology of the body, to the optimistic but updated theory of Parsonian sociology. The theme of his critique is that nostalgic social theory fails to account for the decisive and impressive emancipatory element of modernity (Holton and Turner 1986: 211–16; Turner 1987: 236–41).

     The substance of Turner's critique is that nostalgic social theory fails to account for the emancipation in virtue of its commitment to a biblically inspired Nietzschean fatalism that is in turn wedded to a positivistically informed scientific determinism. The upshot is that the emancipatory element is not accounted for because there is no acknowledgment of its empirical reality and historical importance nor any recognition that such matters are theoretical problems (Turner 1987: 237–38). Turner asks if theorists can continue to accept an account of the historical process of modernity based on the assumption that fatalism and its determinism-driven motor are the sole constituents of the process. After all, if emancipatory success is acknowledged for what it is (that is, bringing into being possibilities which would never have occurred, without active human intervention), then the explanatory force of fatalism and its determinisms is contradicted and severely undermined.

     Turner's critique of nostalgic social theory turns to philosophy to deal with the critical factor of effective human agency that he believes is signified by the emancipatory element of modernity. He is aware that the effective agency of persons in modern social life raises the obvious question of the ontological status of 'agency' with reference to classical modes of determinism which challenge the reality of the status. For some reason, however, Turner does not turn to the philosophy of science (thus to the problem of determinism and effective human agency) but to phenomenology.

     As Turner enriches Merleau-Ponty's notion of the 'lived-body' with additional dimensions of 'having,' 'being,' and 'doing,' he solidifies a persistent conviction that the solution to the problem of determinism will emerge from the resources of an enriched phenomenology of the body (Turner 1984: 245–48; 1991: 31–44; 1992: 38–44).¹ Simply stated, if persons are effective, they are (or will be) effective by virtue of the agency of their active bodies. However committed Turner is to achieving a science of embodied sociocultural action, it appears that science is either being implicitly rejected or presumed irrelevant in his turn to phenomenology.

Turner's Program

The thesis of Turner's argument for phenomenology is that Cartesianism and determinism together constitute the problem of disembodied nostalgic social theory. He believes that the ideal solution is to kill both birds with one philosophical stone. Implicitly, the essence of his solution is to invert the Cartesian strategy of saving the freedom of the mind by the trick of restricting mechanistic determinism to the bodies of the material universe. For Turner, the 'agentic body' means that it is 'active' (1984: 51–54 and 243–51; 1987: 240–41; 1992: 31–33 and 67–68). Because it is alive, the human body is now (magically!) agentic.

     All that has to happen is for nostalgic social theorists to acknowledge the demand for a notion of effective agency, then to recognize that some theory of embodiment must be a precondition for its realization. The crucial point at this stage of Turner's argument is his assumption that Merleau-Pontian philosophy is itself unproblematic for his theoretical purposes (Turner 1984: 51–54 and 243–51; 1987: 240–41; 1992: 16 and 42–43). Where Merleau-Ponty used the Marxist strategy of standing an idealist theory on its head, Turner does exactly that with Cartesianism. He stands the agentic mind on its agentic body. Apparently, he believes this will guarantee nostalgic social theorists what they most need: a notion of effective agency.

     The full burden of the structure of Turner's argument then falls exclusively on the task of presenting historical facts and declaring that their meaning represents an emancipatory victory over the cooperating social, psychological, and discursive determinisms. For example, nostalgic social theory tends to be unidimensional and incorporationist in its view of modern culture, thus failing to grasp the paradoxical and dynamic processes of modern civilization where consumerism may have (at least in principle) an emancipatory impact. The negative and nostalgic perspectives fail to provide a positive evaluation of such elementary developments as sanitation, improvements in communication, the development of modern medical therapeutics, and the availability of pain-killing drugs. He is now ready to announce that "[t]hese negative theories see the body as merely . . . an object or environment that is manipulated by consumerism or regulated by discourses" (Turner 1987: 240). Marx's idea of false-consciousness and Freud's notion of the unconscious are smuggled in as implicit cooperating mechanisms that nostalgic theory claims determine

[t]he whole consequence of modern consumerism [that it as a] political subordination bringing about the stability of capitalism either as a dominant ideology or a dominant form of life-style (1987: 238; emphasis provided).

In the teeth of the emancipatory victory, Turner can then assert,

Sociological theory will have to incorporate an entirely new perspective in the nature of human embodiment in order to achieve a more dialectical grasp on the character of modernity. (1987: 241; emphasis provided)

     Since the body is seen as a deterministic effect, there is a failure on the part of nostalgic social theory to "recognize that a theory of embodiment is a necessary precondition for . . . a notion of effective agency" (Turner 1987: 241). Traditionally, sociology presumes a Cartesian dualism "where mind is seen to be the causal knowledgeable agent and the body is relegated to an object or an environment" (ibid.). The "new perspective" is the idea that the body too is a causal agent. Somehow, the causal agency of the mind is now given to the body, so that ipso facto the body is active: "Neither Weber nor Foucault provide a phenomenology of the active body as an essential component of human knowledgeable agency" (ibid.).

Phenomenology: Residual Positivism and the New Ontology

There are two questions concerning phenomenology that need clarification: 1. Does the residual positivism of Husserlian and Merleau-Pontian phenomenology render both irrelevant to the problem of determinism and human agency? 2. Does Merleau-Ponty's new ontology for the interpretation of the actual body as 'flesh' (not simply as 'lived') threaten to end in mystification rather than being the promise of a solution to the problem of determinism and agency? In this regard, I suggest that Merleau-Ponty reached for—but never quite knew how to realize—a conception of causal powers.

     Indeed, if Harré is correct in thinking that Merleau-Ponty probably never intended what in effect he did—namely, assign agency to the lived body in order to legitimate the belief that persons are effective—this may well be one of many reasons why he searched for a new somatic ontology in The Visible and the Invisible (1968). Since this issue has been discussed elsewhere in full, I will merely use the fruits of that explication here (Varela 1994b: 168–69 and 176–83). With this caveat in mind, I want to say that Merleau-Ponty's search led him to slide away from the idea of the 'lived' actual body toward an idea of the actual body as 'flesh' because, in that slide, he tries to capture his vision of its dynamic constitution by referring to it as a "[p]regnancy . . . a power [of] fecundity" and, thus, "a power [that] is not a factual power" but instead "an internal arrangement . . . extant by its own efficacy" (Merleau-Ponty 1968: 160–61, 208).

     In this new primacy of what Merleau-Ponty actually calls "primordial pregnancy," I suggest that he is on the threshold of a scientific ontology of the generativity of powerful particulars (1968: 208). This suggestion allows us to believe that his post-Cartesian recharacterization of the actual body in terms of "I can" (not the Cartesian "I think") carries that very theoretical force, for the "I can" is to be seen as referencing the powers of particular persons. And Merleau-Ponty also discovers the complementary primacy of language as part of the new ontology of phenomenology. Consider the following remarks:

[I]f we were to make completely explicit the architectonics of the human body, its ontological framework, and how it sees itself and hears itself, we would see that the structure of its mute world is such that all the possibilities of language are already given in it (1968: 155; emphasis provided).

[T]he actual body I call mine [is] this sentinel standing quietly at the command of my words and acts (1968: 254; emphasis provided).

We are now up against the mystery of understanding how the two hand-in-glove primacies of flesh and language are to be connected.

     My hypothesis is two-fold: 1. the "I can" of the actual body is in fact the discursive agency of embodied persons, and 2. the actual body as 'flesh' simply represents the principle of causal powers. In view of this reinterpretation of Merleau-Ponty's new ontology, it would be possible to construe it as consistent with Harré's version of recovering human effectiveness as the discursive agency of embodied persons.

     Turner's preference for phenomenology over a philosophy of science suggests that he believes the latter is irrelevant simply because it does not have the required resources to handle the problem of determinism and the recovery of human agency because it is constitutively committed to empirical realism. Probably no one would disagree. As Harré and Bhaskar have made definitively clear, Humean empiricism banished agentive causation from the natural world. As a result, the only available theory of relational activity among things is regularity determinism and its conception of the correlation of passive particulars (Bhaskar 1978: 56–62). There is no doubt that Turner—or anyone—would and should prefer phenomenology to positivism wherein natural kinds of particulars are categorically defined only in terms of liabilities, never in terms of powers, forces and capacities (language being of primary importance). These are what make the notion of causal effectiveness possible, but more of that later.

     Turner's problem is that, without Merleau-Ponty's slide toward a new ontology, phenomenology cannot legitimately be considered to have any philosophical relevance with regard to the issues at stake, especially when the active body of any effective person is the sole reason for the belief in its relevance. Apart from the fact that phenomenology is not a philosophy of science, it presumes an ontology of empirical realism (the heart of positivisim) and so, strictly speaking, can provide no conception of agentic causation because correlation is the substitute for causation. That is, the 'relationality' of the corelation is assimilated to empirical concretizations, such as, for example, 'experience,' 'perception,' 'sensation,' 'feeling,' and 'function' in the Jamesian sense (Varela 1995b: 263–74). Clearly, Turner (or anyone) cannot possibly derive any notion of agency, effective or otherwise, from a phenomenology of the 'lived-body.'

     As I understand the matter, the sensitizing point of the concepts of the 'lived-body' and (the body as) 'flesh' is meant to highlight the importance of the notion of human agency—a fact that still has to be explained. Before going on, however, it is telling that Turner's program for an embodied theory of social action appears coherent on the surface only because he already brings the idea of an active body along with him to phenomenology.

     We must now ask: where does Turner get his assumption of causality, agency, and (put together) the idea of a 'causal agent'? In the context Turner chooses, the idea is ontologically groundless. His assumption is revealed by his apparent denial of the relevance of a philosophy of science on the one hand and his error of believing that phenomenology can provide what it cannot.

     Turner is not alone in making such fundamental errors. Almost all of the prominent sociocultural theorists of embodiment have made their commitment to Merleau-Pontian existential phenomenology against positivism in the hope, perhaps, of preserving the agentic status of human being (namely, Csordas 1989, Frank 1991, Jackson 1989, O'Neill 1989, and Shilling 1993. There are exceptions, of course, for example, Farnell 1995, Harré 1991, Varela 1994a, and Williams 1982).

Determinism and the Ubiquity of Causal Powers

Having cleared the underbrush, we can now confront the problem of determinism and embodied human agency as a problem in the philosophy of science, which is where the problem belongs. We are now free to pursue the need for an appropriate conception of determinism so that we may arrive at a viable conception of causality. We can make principled determinations in the ascription of agency to powerful particulars: to genuine (in contrast to freakish) candidates for the status of causal agents. Realist theories of science reveal that a suitable kind of determinism for a conception of agentive causality (and for a fruitful theory of causal powers) is ubiquity determinism rather than regularity determinism because agentive causality is lost to correlation (Bhaskar 1978, 1979; Harré 1986, 1993).

     Bhaskar's central point concerning regularity determinism is that it mistakenly assumes that laws are invariant conjunctions of events. That is, given a preceding event, another event that follows must follow since nothing else could have. That an event is caused does not entail that it is bound to happen, when in fact caused. After all, Bhaskar declares, determinism is not predestination (1994: 139). The crucial insight is that the empirical conditions of classical determinism are closed conditions, that is, the special conditions of the experimental setup and the rarity of such closed conditions in the open natural world of competing and intervening events. The hidden premise in the idea that what is caused is bound to happen is the experimental provision of ceteris paribus (Bhaskar 1978: 63–79).

     We can fully appreciate Bhaskar's suggestion that the reason for regularity determinism is actualism (1979: 25–26), but his suggestion wants spelling out. Behind actualism—a seriously misleading positivist conception of determinism—is an empirical realist ontology, which is what makes the conception work. That is, if what is real is what is perceived (directly observed) and, therefore, what is experienced, then deterministic laws are seen to reside exclusively in the interrelating events that are manifested to perception, observation, and experience. Laws are thus 'there' (so to speak) since they are not to be realized. One suspects that Plato's metaphysic of plenitude is the origin of actualism and empirical realism because what nature is (its Being) is not in its becoming, since it already is what it all 'is.' This means there is no 'can,' as if there were no possibility of surprises. There is only what is and the necessity of what is.

     A new realist conception of causal powers reveals that actualism is a fallacy since it is false to identify a causal power with any of its occurrences. It is equally false to exhaust a causal force in any of its instantiations. It follows that, as a fallacy, regularity determinism is an artifact, and ultimately an illusion. This form of determinism is in error because 1. it is a limiting case and not a central case of law-like activity in the natural world and 2. (more fundamentally) even in this form, it is a causal production exercised by forceful particulars at work. In other words, the classical determinist presumes that particulars mutually involved in law-like relations are inert. There is a hidden bias to define such particulars only by their liabilities—never by their powers.

     Harré captures this point precisely in the following characterization of one of Newton's famous three laws:

Vis insita is the power [of an entity] to persevere in a present slate, whether at rest or in motion, and it is exerted only when another body tries to change its direction. So in the absence of that other body, that power simply drives the body forward or maintains it at rest (Harré 1993: 23).

If this is the case, then this question arises: Why did the traditional conception of regularity determinism constitute 'causal' and 'caused' particulars as inherently passive, inert, and characterizable only in terms of liabilities?

     Harré answers this question thus: "[T]his was the legacy of theologians who wished to reserve powers to non-material beings. Matter, the stuff of the human body, was passive and could not perform any agentive tasks whatever" (2001: 91–92). In the same context, he takes readers to the next step, which is to understand that the theological view also gave rise to a Humean metaphysics of the passive conjunction of events.

     Resulting from the categorical dismissal of a mechanistic ontology is Harré's replacement of traditional ontological concepts with a new realist ontology. Harré's comments in this regard turn around a central realist insight: "to be is to have a place among the beings of a world, not to be the value of a variable" (Harré 1986: 322). Thus, scientific theory is a disguised metaphysic (a 'type-hierarchy') of the beings of a world for a given domain of interest (see also Aronson, Harré, and Way 1995). We can paraphrase Harré's relevant points in this regard as follows (2001: 91–100):

1. The whole of the sciences are or should be ordered by the same highest ontological supertype, the powerful particular, the general idea of an originating agent. For example, in physics the generic concept is 'charge' of which there are several species, such as electric, magnetic, and gravitational.

2. Taking the idea of persons as the primary actors surrounded by a penumbra of skills and dispositions . . . we can see how people play a logically very similar role that is played by charges as the primitive material beings of the physical world. While people are the sources of temporally and situationally differentiated skills and dispositions to act on and to interact with other suitable beings, magnetic poles, electric charges and so on surrounded by their fields, spatially distributed dispositions to act on suitable test bodies.

Harré's development of Bhaskar's exposition of the failure of regularity determinism has taken us into its replacement—ubiquity determinism.

     From the import of the proposal that the sciences are to participate in a shared metaphysic of powerful particulars, we can see a significant connection between the physical and the human sciences: in the exercising of causal powers worlds are brought into being, such that, for instance, in the case of electromagnetic forces, the material world is brought into being and in the case of the forces of human beings, the conversational world is brought into being. From this proposition, Harré derives the idea for his theory of agentive discourse (Harré 1996: 120–35):

The conversational world, like the physical world evolves under the influence of real powers and forces, dispositional properties of the utterances that arc the real substrate of all interchanges. (Mülhäusler and Harré 1990: 24)

Implied in this principle is the concept of the person as the powerful particular whose engagement in various discursive practices is the exercising of specifically human causal powers (see Varela 1994b).

     In other words, persons are pure agents in the special sense that they cannot be reduced either to the natural laws that naturally ground them or to the social rules which (since they themselves have been humanly invented) can only culturally ground them. The point is that their domesticated causal powers permit human persons to determine both the conditions of the applicability of natural laws and the conditions of the applicability of social rules. Thus, the conception of ubiquity determinism becomes the direct basis for the recovery of human agency in virtue of the conception of agentive causality. The latter leads to the agentive discourse of embodied persons.

Causal Powers Theory and Principled Determinations

That ascriptive decisions can be made that generate distinctions between genuine and freakish powerful particulars follows straightforwardly from the systematic organization of the conception of causal powers. The systemization shows that such ascriptions can be principled instead of being left to the hermeneutic intentions of a theoretician from ideological commitments that often define the commitments. The logic of causal powers theory has several components that should be understood one at a time.

Causal Activity and the Powers and Particulars Schema

     The principle of causal activity and its distinctive powers and particulars schema may usefully be thought of in terms of four propositions:

1. The principle of causation is the strict conception of the production of activity and its consequences.

2. Causation is the activity of powerful particulars doing forceful work.

3. The doing of forceful work is the signature of agency.

4. Agency is, therefore, the production of activities by virtue of powers that constitute the nature of particulars.

     Particulars are any of the natural kinds of real entities (or units) that are the necessary sources of necessary activities. However, this is qualified by the principle that causation is materially (that is, necessarily) related to its effects, but its effects are statistically (hence, contingently) related to its causation. Consequently, powerful particulars also entail liabilities in virtue of the real world of multiple, interacting, and competing other powerful particulars.

     The principle also specifies that such particulars are constituted by an exclusive and special relationship between a power and its given particular. This is the power of a particular schema, and never a power and a particular schema, because this idea precisely denotes the structural, thus the functional integrity of a given particular which then allows one to conceive of that particular as a genuine entity. Note, too, that agency can strictly be ascribed only to genuine entities by virtue of their structural-functional constitution. A concise example will clarify the point.

Choice of Predicates

     Consider two identical inclined planes and two identical mass-objects with opposite shapes—a perfect circle and a perfect square. Using each plane for each object, so that they are released from the raised ends of the inclined planes, is the description of their falling movement down the inclined planes accomplished by an indiscriminate choice of predicates? Are there no natural kind-dependent properties of the mass-objects to be theoretically (thus, empirically) respected?

     A referential realist's choice of descriptive predicates, that is, 'rolling' vs. 'tumbling,' would be principled because the choices would appropriately fit 'rolling' to 'round-shape' (circle) and 'tumbling' to 'square-shape' (square) in virtue of the scientific principle that the nature of a thing determines what it can and cannot do. Structure determines function.

     The masses in question are objects; substantive because they are real natural kinds (that is, genuine entities) and bona fide powerful particulars. To fail with regard to choice of predicates would be to tolerate descriptive ideas such as, for example, "a 'square-shaped' object rolls down the inclined plane" or "a 'round_shaped' object tumbles down an inclined plane. Moreover, one would have to live with the idea that the nature of a thing (its structure) has nothing to do with what it can or cannot do.

Nonquestions

A strict understanding of the constitution of the idea of natural kinds of causally empowered entities makes it conceptually impossible seriously to ask certain kinds of questions as if they were theoretically viable, for example, "Where is the explosion before a stick of dynamite has been detonated?" Or, "Where are the photons located?" The point is this: whether it is explosions before they occur, photons before they are in evidence, motives before they are negotiated in action, words before they are uttered, intentions before they are identified in social actions, or, finally, structures before human social relations are in fact patterned in their activities, there are no 'photon bags' nor any kind of social or psychological 'structure bags' that will help materialistically minded individuals to come to grips with such concepts. Unfortunately, the theoretical vacuity of this kind of question is often found at the very heart of social science thinking. It lies at the root of its propensity to reification and can further be seen in the remainder of the critical features of the conception of causal powers—in violation of the principle of causal activity.

     To violate the principle of causal activity is to transform its power of a particular schema into a power and a particular schema. There is a direct consequence of this shift in schemas: the structural-functional integrity of a particular (the theoretical reason for regarding it as an entity), whose constitution as a powerful particular enables it to do forceful work, has had its integrity destroyed.

     When this happens, what one has, in fact, is a commitment of the fallacy of bifurcation. That is, powers and their particulars have been decoupled with the conversion of a 'power' into a sui-generis, free-floating, abstract activator, which leads to the fallacy of activation. The latter denotes cases where the causal activity of powerful particulars is transformed into powers activating particulars. The crucial consequence is that two kinds of error of causal powers ascription are open for commission:

1. One can commit the ascriptive error of externalization. An activator is mislocated by being projected outside of any one or more particulars, thus treated as an external-independent activating determinant. Note that the point of this error is that, in principle, no power is located as a structurally constituted property of a given particular. Therefore, it can have no theoretically coherent function in the causal activity of a given particular in the real world. In view of this, recall our inclined plane example: the properties of 'rolling' and 'tumbling' cannot be ascribed indiscriminately in contradiction with the movements of round and square objects.

2. One can commit the ascriptive error of internalization. An activator is mislocated by being introjected inside of any given particular, thus treated as an internal-independent activating determinant. In this regard, recall the stick of dynamite: the 'explosion' is not located 'in' the stick itself, as if there is an 'explosion' waiting to activate the stick of dynamite. The stick simply explodes.

What about People?

     When causal powers theory is applied to the human case, certain conceptual distinctions which can be forged can then be used to make principled determinations in the assignment of agency. The application presumes that, to regard human beings as genuine and not freakish powerful particulars, structural-functional integrity must be respected.²

            The application also requires a critical assumption concerning powerful particulars and evolution, thus bringing to the foreground the special character of the recovery of human agency. When Harré declares that "the power of speech to alter consequences is something which in a way mocks causality" (1984: 162), the point is that agentive causality is the recovery of human agency.

     The performativity of talk³ in relation to time, unlike the activity of physical things in relation to time, is such that "social talk as social action can create the future by making promises, commitments, contracts [even revolutions], . . . [and] it can be used to recreate the past (Harré 1984:161). What we have here is the first major breakthrough that takes us beyond the prison house of the actualism mentioned earlier: namely, that the liabilities of a given particular in motion or at rest imply the powers of other causal agents, thus presume the agentive powers of that given particular.

     The next major breakthrough is the crucial assumption that the evolution of physical forms of powerful particulars continues so that we have the evolutionary stratification of complex organic powerful particulars. In the case of human beings, we have the species-specific advance beyond primate social brains to cultural brains emerging between Homo habilis/erectus and Cro-Magnon. As 'cultural brains,' human brains are structured like a language (Harré and Gillet 1994: 80–96).

     In view of the presumptions of applications of causal powers theory, the distinction must now be made between natural powers and cultural powers constitutive of human beings. Natural powers are grounded in the material individual and so belong to the acultural organism. The natural powers of the organism are the conditions of the facilitation of action. Self-mobilization, the very process of agency, is a social act. It is a social relation of persons. In this strict sense, the social relations of persons are agentic. The act is a mutual process whereby persons consider how the other will, can, or could act in response to their own act. This is done in order to direct themselves to act in such a way that a joint act is accomplished (see Varela and Harré 1996). Cultural powers are grounded in the social relations of persons and so belong to persons. Thus, they are the conditions of the enactment of actions.

     As the natural individual is transformed via the social symbolic relations of socialization (psychological symbiosis) into a cultural person, the organism is transformed into a body. Note that the deep point of these crucial distinctions is that (speaking ontologically, not empirically) persons do not dangle at the end of their social relations, since the rules or conventions defining the substance of these relations do not use people, but rather it is people who use the rules and conventions that make relations possible.

     All of this fortifies us against the error of externalization found in Durkheim's worst scenario: namely, the classic deterministic view that the problem of social order is the problem of the necessity of order (and therefore ordering), so that people must be seen to be living deterministically within it. The Wittgensteinian distinction between using rules and being used by rules allows us definitively to block the theoretical predilection to effect the subtle shift from the belief in the necessity for the rules to the belief in the necessity of the rules in order to understand patterned social activity. Speaking theoretically (not ideologically), there may well be the necessity for social order, but never the necessity of social order.

     Finally, we will note that these same critical distinctions regarding conditions, transformations, and the proper location of agency fortifies us against the error of internalization found in the worst scenarios of Freudianism and cognitivism (and their variations) on the idea of mentalizing the organism (Harré and Gillet 1994: 1–36). In those scenarios, person, culture, and embodiment have been swallowed up in the naturalization of human agentic life.

Beyond Turner: Agentive Embodiment to Agentive Discourse

I don't think that a theory of agency generally is necessarily about human agency, and I don't see any problem in conceptualizing social classes or organizations or even societies as exercising agency. (Turner 1992: 247)

This telling comment is contextualized in the last chapter of Turner's book to a conversation with Fardon concerning the entire corpus of his work on embodiment, recovering human agency, and revitalizing sociological theory.⁴

     The theme is Fardon's declaration that Turner's work on agency is ultimately rhetorical. His critical observation is Turner's conceptual failure to distinguish human activity and human agency (1992: 246–48). Turner redundantly replies that agency is "the capacity of knowledgeable human agents to bring about changes." Again, we see that the ontological grounding of the idea of an active body (thus, the effective agency of persons) is absent. Turner cannot connect 'activity' and 'agency' because he does not have a systematic conception of causal powers, with the fatal consequence that he actually ascribes agentive causation to the 'body,' insisting that the body is active and persons are effective.

     At the heart of Turner's difficulties, we find the commission of both errors: externalization and (with regard to the body) internalization. On the one hand, he fails to distinguish the pre-embodied organism and its materially grounded natural powers of facilitation from the embodied person and his or her culturally grounded powers of social enactment. On the other hand, he conflates natural and cultural powers. Turner's location of human agency entails treating the body as if it is the organism—a necessary upshot of his preference for a phenomenological solution to the problem of determinism and the recovery of human embodied agency.

     Given that Turner regards assigning agency to collectivities unproblematic, he also declares that his prescription for an adequate sociology of the body entails the dismissal of Durkheimian sociologism as an example of "simple reductionism." However, unable to recognize the contradiction, Turner can only mean that Wrong's thesis of the rejection of the oversocialized conception of man in traditional Parsonian theory (see Holton and Turner 1986) is to be extended to the rejection of "an entirely over-socialized conception of embodied persons"(Turner 1984: 246). This is the crux of Turner's rejection of language or discourse as the model of social action.

     Turner suggests that Foucault promotes a determinism that is implicit in the fatalism embedded in Nietzschean irony and implicated in the explanatory formalism of structuralism (1984: 242–44).⁵ His counterthrust to Foucault is to declare that "bodies may be governed, but embodiment is the phenomenological basis of individuality (1984: 234). In other words, the historical social practices of cultural discursive life are performed by individualized embodied persons.

     Such individuals "practice in, on, and through their bodies" (Turner 1984: 245). Thus, it is this reality of the individual's practice of having a body that confers on being human the status of causal agency. And he points out that it is the phenomenology of agency for which discourse determinism does not provide (1984: 248). According to Turner, the deep consequence for Foucault's view of embodiment is that the body is conceived to be the effect of discourse.

     Turner also insists that Marx's idea of the 'sensuous potentiality' of the body is lost (1984: 250) if the body is no more than an effect of discourse. Rather than the phenomenological fact of the language of ownership—"this is my body, my pain, my disease," we have the objectified language of ownership of the body—"this is its body, its pain, its disease" (ibid.). After all, if language speaks the subject, it must also speak the body. While Turner is correct in his renunciation of the idea that language speaks the embodied person, once again, the theoretical means he uses to justify the idea is suspect.

     From a new realist standpoint, Turner's appeal to phenomenology involves a fundamental error. It must be the case that he has missed the idea that, to avoid a 'simple reductionism,' it is not the experience of an active body but the enactment of bodily activities of any kind that defeats the determinism. We are obliged to understand Turner's allusion to agency as a practice that is effected 'on,' 'in,' and 'through' the body with a significant difference: 'having' a body is being a body, to be sure, but this is merely locative—being at 'this place' at 'this time.' Having a body which is being a body is the case in virtue of agency. The doing of agency is enactment.

     The experience of the activities of bodily techniques, movements, and gestures presuppose their enactment. In fact, human action is constitutively discursive, conversational, communicative, and linguistic. Action itself is 'talk': talk about the body (representation), talk of the body (experience), and talk from the body (enactment).⁶ It is my judgment that a proper theory of dynamic embodied action—which Turner clearly comes to call for in his later work⁷ —requires Farnell's idea that the body is not just 'physical being' but moving being (Farnell 1994: 939). The idea must be that talk from the body as the enactment of speech act systems and action sign systems is the primary generative act. The principle must be that conversational or discursive practices are performatively grounded in and conventionally a structuring of a suitable gestural region of the body. This is a realist reading of Merleau-Ponty's insight (cited above) that the ontological framework of the perceiving and mute body "is such that all of the possibilities of language are already given in it" (1968: 155).

     It is not bodies that 'enact agency' or 'experience an enactment,' but people. In the beginning is the person, therefore the body. It is only the embodied person who enacts bodily activities, never the body enacting the person. In that case, the 'body' is, in fact, the 'organism'—but the organism only facilitates; it does not enact.

Turner's Critics

Shilling has failed in his attempts to provide an adequate theory of the agency of embodied persons, and, in condemning Foucault's alleged discourse determinism, he gives the unfortunate reason that it denies "the mind's location within an active body." He then insists that from this determinist standpoint, the body's "own powers of production" are derived from discourse. Subsequently, he concludes, "[T]he body is dissolved as a causal phenomenon into the determining power of discourse (Shilling 1993: 80). The conceptual separation of the body and its causal agency from discourse reveals that Shilling conflates the pre-embodied facilitating organism and the enacting embodied person. As an 'organism' (not as a 'body'), there is neither enactment nor facilitation. For only the body is real, since only as a person can an organism be a body, hence facilitate the enactment that only persons perform.

     The same consequence is to be found in a paper by Frank (1991). He correctly wants to move from Turner's "social systems problems and the body" to "social action problems confronting the body," but he does this in order to locate the body at the intersection of the "equilateral triangle" of "institutions, discourses and corporeality." While this idea of location may or may not be fruitful, Frank then fatally casts the idea in an identical Turnerian manner: "Bodies, of course, do not emerge out of discourse and institutions; they emerge out of other bodies, especially women's bodies (Frank 1991: 49, emphasis provided).

     Turner, Shilling, and Frank fail to see the scientific issue of determinism and human agency as central to the problem of how embodiment is going to "resolve basic issues in the nature of social action and social structure" (Turner 1992: 259).⁸ In the end, all three scholars turn away from philosophy of science and proceed to an enriched phenomenology of the lived body as the solution to the problem of assigning causal agency with regard to the human levels of body, person, and society. As I have attempted to illustrate, Harré neither made nor repeated this mistake.

     Harré's realist theory of science has provided the concept of agentive causality which (when suitably understood and systematically worked out) is the recovery of human agency. The recovery takes the human form of agentive discourse; thus, once and for all, the difficulty of discourse determinism can be set aside. The new realist view is the philosophically fruitful way to conceptualize the effectiveness of persons in sociocultural life, for it is the primacy of persons enacting speech acts and action sign systems—both of which are appropriately constituted by gestures and other bodily movements (see Williams 1982, 1995; Farnell 1994, 1995).⁹