Ronald Lemmen (1994) Taking Experiences Seriously. Psycoloquy: 5(28) Evolution Thinking (3)

Volume: 5 (next, prev) Issue: 28 (next, prev) Article: 3 (next prev first) Alternate versions: ASCII Summary
PSYCOLOQUY (ISSN 1055-0143) is sponsored by the American Psychological Association (APA).
Psycoloquy 5(28): Taking Experiences Seriously

Book Review of Sheets-Johnstone on Evolution-Thinking

Ronald Lemmen
School of Cognitive and Computing Sciences
The University of Sussex
Brighton BN1 9QH England


In showing how meaning and human thinking and practices are rooted in the tactile-kinesthetic body, "The Roots of Thinking" provides an interesting and welcome new perspective on cognition. Maxine Sheets-Johnstone (1990) leaves no room to doubt the need for a thorough "corporeal turn" in cognitive science.


analogical thinking, animate form, concepts, evolution, tactile-kinesthetic body.
1. Many cognitive scientists will heartily agree with Sheets-Johnstone's (1990, 1994) conviction that for a proper understanding of the mental, a "concerted turn toward the body" is needed (p. 19). However, Sheets- Johnstone goes much further in arguing for the unity of mind and body than most cognitive scientists would dare. "The mental" and "the physical" have traditionally been studied separately, leading to the propagation of two unnatural species: not just disembodied minds but also mindless bodies. We should do more than pay lip service to embodiment while essentially still "[keeping the body] in its usual place with regard to a self, mind, or subject" (p. 304). The living body is more than a device that can make certain discriminations and can move in certain ways. If we are really to acknowledge the mind/body unity, we have to think of the body as an experiencing and experienced body.

2. Sheets-Johnstone's main objective is to show that the origins of human thought and concepts are to be found in the living body, and how. The main message for cognitive scientists is that experiences really matter for cognition, and that they should therefore be taken seriously. Although some will claim that since the "other minds problem" is insoluble in principle, the use of notions like "what is available in experience" should be banned from cognitive science; in my opinion, the notion is useful and necessary. Sheets-Johnstone has done an admirable job in showing how it can be used. The fact that cognitive science does not routinely deal with experiential values and their significance for cognition is a telling one concerning the shortcomings of cognitive science rather than those of Sheets-Johnstone.

3. In the philosophy of mind, a few recent theories analyse meaning as "what is available in experience." Cussins (1990) speaks of "non- conceptual theorizing," because the purported meanings cannot in general be analysed in conceptual terms, that is, as standing for objects, properties and situations. Lakoff and Johnson (Johnson, 1987; Lakoff, 1987) even call their program "Experiential Cognition." It is interesting to notice how well these theories match that of Sheets-Johnstone. All agree that meaning is primarily tied to what is experientially available to a whole cognizer, and that traditional (conceptualist) meaning theories are fundamentally incomplete. They also agree that understanding consists in ways of perceiving or moving, and that these are fundamentally tied to bodily capabilities. Sheets- Johnstone is most adamant of all in emphasizing that these capabilities need to be understood as more than mere perceptual-motor pathways. Tactility and kinesthesia are modes of knowing the world (p. 291) because they are our fundamental ways of experiencing it. In fact, Sheets-Johnstone argues forcefully that it is the tactile-kinesthetic body that is inherently meaningful, and that the meanings in vision and in language are rooted in and dependent on it.

4. There is now, especially in the Artificial Life community, a growing interest in explaining behaviour in terms of interacting behaviours rather than in terms of functional decompositions (e.g., Brooks 1991; Hendriks-Jansen forthcoming). Insofar as this is a way to treat cognitive systems as organic wholes, that is good news. However, cognitive creatures are still treated as empty shells of "doings." The eight case studies that Sheets-Johnstone presents are all evidence of the need to deny "the largely unspoken scientific split between corporeal experience and corporeal behavior" (p. 279). For example, "to walk means not merely to have a certain anatomy or to put one leg in front of the other in a certain way for all to see, but to experience a quite particular gait" (p. 369). This experience allowed a simple binary periodicity to be brought into human awareness; one that, according to Sheets-Johnstone's analysis (chapter 3), this was was one of the crucial steps in the invention of Number.

5. The analyses will look rather alien to many cognitive scientists. It Sheets-Johnstone is therefore to be applauded for carefully spelling out the whats, hows and whys of her methodology. Although she is primarily concerned with reconstructing the life-world of our hominid ancestors, the insights that she arrives at have everything to do with the concerns of those who study (human) cognition. A reader who wants to find out quickly how the relationships lie may want to start reading the last chapter on tactile-kinesthetic invariants. Having the book at hand, the reader is strongly recommended to take a look at the fascinating case studies in part II.

6. Tactile-kinesthetic invariants play an important role in Sheets- Johnstone's account of communication. She argues that human language is on a par with the ways other animals communicate, in that it did not originally consist of arbitrary elements, but made use of "corporeal representations." Using examples ranging from the Tanzsprache of bees through bipedal sexual display of male early hominids to primordial language, it is shown that it is a common theme that "animate bodies represent by symbolizing the spatio-kinetic dynamics of their own experiences" (p. 123). What is symbolized and communicated is not a matter of facts or objects, but one of "motional-relational complexes" in the world (p. 126). This is achieved by relating these iconically to the animal's own prior tactile-kinesthetic experiences. This notion of corporeal representation appears to be a very useful one with respect to explanatory projects in cognitive science, especially when linked to animals' awareness of their own behaviour, that is, their own sensory- kinetic experiences.

7. All of this does not mean that we now know how to model the mental processes of communicating animals. First of all, we still have no idea what it takes for a system to have experiences. Experiences are being taken for granted by Sheets-Johnstone, as biological givens. Yet it is not clear at all that biology (or Artificial Life) will be able to tell us which creatures have experiences and which have not. The only useful suggestion Sheets-Johnstone makes is that it seems "biologically axiomatic that, to the degree an animal has to learn to move itself, it is self-aware" (p. 371). Another nontrivial matter with respect to modeling is the ability to transfer sense analogically from one bodily part to another. Many animals make gestures (e.g., with the hand or tongue) that are iconically linked to an experience involving another part of the body. Many display behaviours are like this, as when erect body posture corporeally represents penile erection (chapters 4 and 7 make up for a traditional scientific blindness to sexual behaviours and experiences in early hominids). It is not clear how such analogical transfer of sense is achieved, apart from there having to be a sameness in the ways two behaviours are experienced. But Sheets-Johnstone does make clear that "such transference is at the root of concepts" (p. 61).

8. Displays or communicative acts are not only experienced by the behaving animal, but also by the animal(s) that the behaviour is directed to. For the latter to understand them as meaningful, they have to understand them on the basis of their own corporeal experiences. This is a direct consequence of the "biological disposition to use one's own body as a semantic template" (p. 308). Successful communication between conspecifics thus hinges on the existence of tactile-kinesthetic invariants (it is only through these invariants that the animals can "agree" on a meaning) and on "analogical apperception." The latter notion refers to the perception of other bodies being filled in in terms of the ways one's own experiences would be, if one's own body behaved like the perceived bodies. Again, Sheets-Johnstone does not indicate how analogical apperception might be achieved, but she does make clear that it is an essential element of cognition.

9. A final problematic issue concerns the relation between corporeal representation and natural meaning. On the one hand, Sheets-Johnstone appears to claim that an animal's understanding of movements and orientation of other animals (non-conspecifics) is continuous with its understanding of natural meaning, as in the connection between smoke and fire. This seems necessary, because there are many cases in which an animal cannot rely on tactile-kinesthetic invariants being shared with the thing or organism that is understood. On the other hand, she forcefully drives home the point that even our understanding of things that can happen to a matchbox is corporeally grounded. On several occasions she mentions work by Piaget in which he describes how an infant progressively opens its mouth as it is opening a matchbox. It is not clear whether (and if so, how) this use of the own body as a semantic template mixes with the grasping of natural meanings. Sheets-Johnstone's claim is quite clear: meaning arises from what is inherently meaningful, namely, the animate body. There is much to be said for that view, but as it stands, it leaves natural meaning unexplained.

10. In summary, "The Roots of Thinking" is a fascinating book. The author not only makes a powerful case for the "corporeal turn," but also alerts those who are already concerned with issues of embodiment that it is insufficient to embody the mind while leaving bodies mindless. Experiences have to be taken seriously, because they are causally efficacious in cognition and even in evolution. The theory has much in common with the work of Cussins, Lakoff and Johnson. Ironically, whereas Lakoff (1987) and Johnson (1987) put much effort into fighting against "objectivism," Sheets-Johnstone presents cultural relativism as a natural antagonist (chapter 10). This is because Lakoff and Johnston are talking to the heirs of traditional epistemology and logical positivism, whereas Sheets-Johnstone is addressing an audience of anthropologists and anthropologically inclined philosophers. It is therefore not surprising that her book is lacking in clues concerning actual mental mechanisms. This, however, in no way detracts from its value for cognitive science. First, because there can be explanation without implementation (see also Hendriks-Jansen, forthcoming). Second, because it provides a fresh and useful perspective on cognition, making us aware of the basic roles of such often overlooked phenomena as tactile-kinesthetic experiences, analogical transfer of sense, and analogical apperception.


Brooks, R. (1991) Intelligence Without Representations, Artificial Intelligence 47: 139-159.

Cussins, A. (1990) The Connectionist Construction of Concepts. In The Philosophy of Artificial Intelligence, M. Boden (ed), 368-440, Oxford University Press.

Hendriks-Jansen, H. (forthcoming) In Praise of Interactive Emergence. In Proceedings of the Fourth Artificial Life Conference, R. Brooks (ed).

Johnson, M. (1987) The Body in the Mind. University of Chicago Press.

Lakoff, G. (1987) Women, Fire, and Dangerous Things. University of Chicago Press.

Sheets-Johnstone, M. (1990) The Roots of Thinking. Philadelphia: Temple University Press.

Sheets-Johnstone, M. (1994) Precis of The Roots of Thinking. PSYCOLOQUY 5(8) evolution-thinking.1.sheets-johnstone.

Volume: 5 (next, prev) Issue: 28 (next, prev) Article: 3 (next prev first) Alternate versions: ASCII Summary