Do extended selves have an experiential unity?

Does the extended mind imply an extended self? In the penultimate paragraph of their groundbreaking essay The Extended Mind (1998), Andy Clark and David Chalmers answered this question in the positive. For them, the extension of mentality beyond the boundary of the skin entails an analogous extension of the self beyond that boundary (p.18). Not only are cognitive states and processes constituted by external components, but agents themselves are best seen as “spread into the world” (p.18).

Although somewhat tangentual to their main argument, this claim by Clark and Chalmers has stirred a significant debate in philosophical circles, a fact hardly surprising given the relevance of the issue both in philosophical and social domains. In her recent article Persons and the Extended Mind Thesis (2009), Lynne Rudder Baker has criticized Clark’s and Chalmers’ bold claim, arguing that the coryphaei of the EM thesis have gone “a step too far” in the direction of extension. Although, Baker claims, we are entitled to view cognitive processes as determined by both internal and external components, the same does not apply to persons. Strictly speaking, there are extended minds and cognitive systems, but there are no such things as extended selves.

Here, I will sketch the basic arguments proposed by Baker, focusing on her ‘qualms’ against the extended self thesis and her account of persons as constituted by bodies. I will argue that Baker endorses a methodologically cheap variant of internalism, rendering her criticism of “self-externalism” decisively trivial. I will claim that the principle of “bodily constitution”, as proposed by Baker, is not a serious obstacle to externalism, and is widely compatible with all variants of the “extended self” theory. Conclusively, I will argue for a positive answer to the central question: extended minds do imply extended selves.

THE TWO ‘QUALMS’ OF BAKER

Lynne Rudder Baker presents herself as a traditional externalist. She explicitly accepts Putnam’s idea that the contents of thoughts are partly determined by features of the environment (p.2.). Following Hurley, she calls this view content-externalism. But, Baker does not stop there. She also accepts the principal idea of the EM thesis: “not only is the content [of thoughts] determinable by features of the environment, but the vehicle [the ‘carrier’ of thoughts] may also be spread out into the environment” (pp.3-4). Vehicle-externalism, then, is also accepted as true – neither the brain nor the skin sets a boundary on the vehicles of cognition (p.5). In the vocabulary of Clark and Chalmers, Baker is both a passive and and an active externalist, endorsing both the individuation of thoughts by external contents and the idea of cognition extending to tools and social equipment.

But, when it comes to the ‘self’, the similarity with the vehicle-externalists ends very abruptly. While accepting that cognitive processes “loop out” into the world, Baker is adamant that the EM thesis cannot be applied to the ‘self’:

On my view, there are no extended persons – persons – who extend beyond their bodies. However, there are enduring persons – subjects of experience, agents, who can think reflectively of themselves throughout much of their existence (p.9).

Baker presents her argument in the form of two ‘qualms’ against the idea of the extended self (p.7). The first one is centered around the idea of the experiental persistence of persons. Even if we accept the existence of an extended cognitive system comprised of a human being and an external tool (“Otto and his notebook”, in the famous argument by Clark and Chalmers), this existence in no way implies the existence of an extended experiental agent that would correspond to the novel cognitive system (p.7). In other words, the transitory cognitive coupling between Otto and his notebook does not generate a new experiental self (“Otto + his notebook”). Otto does not expand to become an extended entity – “[he] does not dissolve or disappear into a cognitive system” (p.7). Strictly speaking, transitory cognitive couplings do not produce transitory selves. There is an experiental persistence of persons which is immune to the cognitive ‘dissolving’ of the self. In the vocabulary of Baker, Otto the person is a concrete particular, and regardless of the extent of cognitive extension, the extended cognitive system cannot be viewed as such. To put it succinctly, persons have points-of-view, and cognitive systems emphatically do not.

The second ‘qualm’ that Baker brings up against the extended self thesis is the distinction between the personal and the subpersonal levels in the analysis of mentality. For her, the distinction between the two levels is ontological; a point, she claims, that is not appopriately recognized by Clark and Chalmers (p.8). There is a fundamental ontological difference between the subpersonal states and processes of a cognitive system, which can be partially constituted by external elements, and the personal states and processes which are always internal. For example, a cochlear implant is a subpersonal device and can be an external component of the extended cognitive system. On the other hands, a tool for conscious manipulation (such as a pencil) is always outside the conscious self, since the personal experience never extends to incorporate it. For Baker, a state is at the personal level “if the person can come to acknowledge the state as her own” (p.10). Defined as such, personal states presuppose consciousness; any state that does not presuppose consciousness is, by definition, subpersonal. And, while neural mechanisms or properties, undoubtedly, ‘underlie’ the conscious states, the states themselves are still at a personal level (the neural mechanisms are subpersonal) (p.10). Baker insists on the strict separation for a reason:

Keeping personal and subpersonal levels distinct is important to me because I do not believe that there are extended agents or extended persons; however, I shall propose a modest hypothesis that recognizes us as enduring persons, whose subpersonal states may have nonbiological parts that play essential roles in cognitive processing (p.8).

So, Clark and Chalmers are right about the possible extension of subpersonal processes beyond the skin, but are wrong about the extension of selves beyond this boundary. Personal states may be comprised of external subpersonal vehicles, but they are themselves states of an enduring subject of experience, a ‘person’ with a persistent, non-dissolving point-of-view.

In order to ground her objections in a unifying account, Baker proposes an ontological model, according to which persons are material beings wholly constituted by bodies, but not identical to the bodies that constitute them (p.12). Persons differ from bodies in persistence conditions: persons endure as long as they have first-person perspectives, while bodies endure as long they maintain certain crucial (typically biological) functions (p.12). Therefore, persons are not identical to bodies: a person can have different bodies at different times and these bodies can be comprised of constituents both organic or bionic. And, while these bodies are usually biological organisms, they are only contingently so – a person may be constituted by a body that is not an organism, as traditionally defined. A body, Baker claims, can be defined by a simple formula: “necessarily, x is y’s body at t if and only if y is a person and x constitutes y at t” (p.12).

With this ontological formula, Baker extrapolates her main argument against the extended self thesis: while bodies can be comprised of biological or non-biological constitutents, persons cannot be ‘extended’ in any recognizeable manner, since they operate on a different ontological level. However they are constituted, persons are enduring experiental agents, not contingent upon transitory subpersonal couplings. The extended mind, therefore, does not imply an extended self.

VALID AND CHEAP INTERNALISM: TOWARDS AN EXTENDED SELF [1]

In order to understand properly Baker’s criticism, we have to go back to the central question that we are concerned with. Namely, what would it mean for the self to be ‘extended’? Is an extended self simply a self whose causally efficacious constituents (vehicles) expand beyond its body (constitutive account)? Or is the extension of the self fundamentally experiential, describing the conscious acknowledgment of external states as one’s own (experiental account)? A careful analysis sees Baker as consistently conflating these two distinct uses of ‘extended’, arguing both that constitutively “there are no persons who extend beyond their bodies” and that external states cannot be experienced as “one’s own”. Consequently, to properly isolate the distinct claims, we have to examine the arguments one by one.

If Baker’s “modest proposal” of self-internalism is to be understood in the constitutive sense, then her assertion that “there are no persons who extend beyond their bodies” is open to serious objections. Namely, if personal states are constituted by subpersonal processes (which Baker grants as being extended), and if personal processes are individuated by external contents, then in what sense are personal states constitutively internal? Baker’s answer at this point hinges on an employment of her ontological model: persons are constituted by bodies, and never by environmental external features. And, since they are constituted by bodies, they cannot, by definition, be extended.

To understand the problem with this argument, we have to distinguish two ways in which one can be an internalist about the self. First, one can equate the body that constitutes the self with the biological organism (bounded by the skin) and claim that only internal subpersonal processes (processes not lying beyond the boundary of the skin) can constitute the self. This is a valid form of internalism. It insists on a profound difference between internal and external subpersonal processes, a difference which can be grounded in various ways (for an example of this approach, see Adams and Aizawa’s The Bounds of Cognition (2008)). But, this is emphatically not how Lynne Baker argues. In order to capitalise on the ontological difference between persons and bodies, she insists on not equating the body with the biological organism, opting for a constitutive formula of the person’s body: “necessarily, x is y’s body at t if and only if y is a person and x constitutes y at t” (p.12). But, by defining the body as that which constitutes the person at a particular time, Baker completely eschews the criterion of the skin in determining the ‘externality’ of the constitutive processes. That is to say, if the body simply is what constitutes the person at the time, then it is trivially true that there are no persons “who extend beyond their bodies” (p.9). If the body is not bounded by the skin (and includes subpersonal processes outside the skin), then any form of externalism can be accommodated into a constitutive internalism. This is a cheap form of internalism, which simply begs the question against the externalist, by determining per definitionem what is fundamentally at issue.

But, this is not the whole story of Baker’s argument. Insisting on a ontological abyss between the personal and the subpersonal space, she insists that persons cannot be extended since they possess an experiental persistence: they have a “first-person perspective” that is not contingent upon transitory cognitive couplings. Persons, in short, do not dissolve to form experiental analogues of extended cognitive systems. But, having isolated the problematic form of constitutive internalism Baker endorses, we may again legitimately ask: just in what sense are the experientally persistent persons internal? If the criterion for the determination of personal states is the possibility of acknowledgment of states as “one’s own” (i.e. the first-person perspective), then a lot of beyond-the-skin processes can easily be classified as personal. As I write this essay, I have a notebook on which I sketch my general argument. Even if I presuppose the experiental persistence of my person, I can recognize the argument sketched on the paper as my own, despite its lying beyond the boundary of my skin. The argument presupposes conscious thought, and therefore is decisively at the personal level. But, that does not make it any less external (to insist that the argument is internal just because it is mine is again to proliferate a cheap form of internalism which begs the question against the externalist).

This example points directly to the profound misunderstanding that lies at the core of Baker’s account of the extended self. Baker repeatedly stresses that persons are enduring, i.e. they do not dissolve to form novel experiental agents of extended cognitive systems. But, to insist on this point is to radically misunderstand the idea of the extended self. Namely, it is not the case that Otto (our Alzheimer-stricken friend) disappears in order to form a new, extended agent (Otto + his notebook). Quite the contrary, Otto himself is already an extended experiental agent. In other words, the experiental analogue of the cognitive system Otto + the notebook is noone else but Otto himself. This is a fundamental point to any proper understanding of the extended self: the extended self should not be viewed as an independent layer “added to the periphery of some internal… proto self” (Malafouris, 2008, p.1997). The extended self simply is the self, the ‘person’ always already constituted by subpersonal processes lying both inside and outside the skin. Insisting stubbornly on the constitutive relationship of the body, Baker fails to appreciate the central insight of the extended self hypothesis: the idea that the existence of environmental and social cognitive tools was the necessary pre-requisite for the formation of the self. That is to say, we are not primiordially persons that can be extended by an addition of external elements; we are persons only because we are already extended in the social environment.

In conclusion, Lynne Baker’s criticism of the extended self hypothesis falls radically short of its intended goal. Both in its constitutive and its experiential variant, the theory of Baker fails to establish an enduring internal self. Her two ‘qualms’ against the extended self hypothesis reveal two different aspects of what we have characterized as “cheap self-internalism”, a theory which renders self-internalism trivially true and compatible with any variant of vehicle externalism. The person may indeed coincide with the body, but if that body includes beyond-the-skin states and processes, then the person is extended in any theoretically and practically relevant sense. The extended mind does imply an extended self.

[1] The valid-cheap distinction was inspired by D.S. Wilson’s distinction between valid and cheap individualism in natural selection. For Wilson’s account, see Levels of Selection: An Alternative to Individualism in Biology and the Human Sciences (1989).  In Social Networks, 1989, 11:257-72

This message was brought to you by...  ASN: a post-rocker, a post-roller, a Hegelian-Deleuzian, a street philosopher, a George Carlin wannabe, a phenomenological bullshitter, an Arsenal freak...