Introduction
John Sutton’s book (note 1)is a work of high scholarly attainment, spanning a wide range of topics requiring different kinds of academic expertise, an intellectual pleasure for historian, psychologist, and philosopher alike, not in the least because of its intriguing, not to say provocative thesis. Indeed, Sutton’s book implicitly or explicitly has a triple focus: first of all it aims at descriptive historiography of psychology and philosophy. Both internal (rational) and external (cultural-historical) influences are taken into account to portray the dynamics of theory development in the domain of early and modern memory research and theorizing.

In addition, however, Sutton’s book can profitably be regarded as a historiographical specimen in the tradition of Lessing’s Rettungen. For the second challenge the book is trying to meet is nothing less than to rehabilitate a much neglected if not ridiculed part of Cartesian philosophy, viz. Descartes’ theory of animal spirits rushing through the hollow nerves and acting, per impossibile (such is the received opinion [note 2]), as intermediaries spanning the unbridgeable chasm between Cartesian body and mind. Yet, notwithstanding the intellectually defiant course Sutton’s main argument is bound to take given its rehabilitative objective, his treatment of opponents and of critical material is admirably evenhanded.

In the third place, finally, Sutton’s book also includes a systematic defense of a theory of memory traces widely deemed to be refuted among modern philosophers of a Wittgensteinian bent as well as among direct realists invoking Gibsonian psychology in their defense. Thus this book is certainly recommended reading for all those who believe that good history of philosophy should be good philosophy simpliciter, if only in the minimal sense of being philosophy teaching by example. Moreover, Sutton writes in a captivating style throughout the book, with flowery, yet remarkably regimented prose, although at a few less fortunate occasions the phraseology stretches from the beautifully economic to the forbiddingly dense to the downright enigmatic.

Central Themes
The two central themes of the book are, first, that the theory of animal spirits has a long history and that, notwithstanding the common derision it has been abundantly exposed to, it has not at all entrammeled the progressive development of the mind/brain sciences (p. 23). On the contrary, Sutton goes on to argue in the second place, Descartes’ ‘neurophilosophy’ managed to exploit animal spirits flowing through brain pores in tentatively suggesting a surprisingly ‘modern’ distributed model of memory employing superpositional storage! This model was much less common than plain animal spirits theory which most contemporaries accepted. Yet it is clear from critics and followers of Descartes alike that this was the way Descartes’ philosophy of the brain was interpreted during the 17th century (pp. 23, 65). Thus this aspect of Descartes’ philosophy did not initially go unacknowledged. On the contrary, it had important historical influence. It is safe to conclude, therefore, that there is a much longer background to dynamic views of memory than is commonly acknowledged (p. 15).

Methodology
This admittedly anachronistic reading of the Cartesian corpus (pp. 15, 23, 155) Sutton boldly defends against several major objections. Yet his hermeneutic stance raises profound questions of historiographical methodology. Should intellectual history be described only in terms available to the historical subjects themselves, thus sticking to a narrowly positivist, yet widely applauded conception of historiography in the tradition of Ranke’s adage to write history ‘wie es wirklich gewesen ist’? This is a very restrictive methodology indeed. And it certainly runs the risk of blinding the historian to the emergence of creative theoretical hunches and insights that his historical subjects might have wanted to describe in more enlightening terms now familiar to us if only they had been able to wrestle such concepts from the unadapted conceptual environments given to them in their own time. Here it would seem the historian is justified in exploiting the advantage of hindsight. Thus a less restrictive historiography may have the potential of being the more enlightening and unraveling, yet stands in danger of the opposite risk of writing no more than mere Whig history.

Sutton is acutely aware of this risk and bravely confronts it. Thus when Harrington notes that ‘the historian must make every human effort to discipline his or her culturally colored subjectivity and take the historical evidence on its own terms’ (Harrington, 1987: 5), Sutton decides to ‘refuse this sensible historian’s caution, and […to] flirt throughout with the twin dangers of nostalgia and present-centredness.’ (p. 15) Indeed, in a slightly different yet analogous context he goes so far as defending a deliberately partial reading (p. 57), as when he decides to bypass Descartes’ separate ‘intellectual memory’ (which philosophers have traditionally focused on to the exclusion of Descartes’ disconcerting neurophilosophical ruminations) and to concentrate instead on Descartes’ theory of corporeal memory. Here again, rather than just letting the historical subject ‘speak for himself’ by taking the whole undiluted text as the historian’s assigned object of inquiry in whatever confused first-person actuality, Sutton prefers what anthropologists would call the third-person or ‘etic’ stance over the first-person or ‘emic’ one. The former methodology would encourage adopting a hermeneutic style in which a corpus of texts might legitimately be interpreted in terms that are foreign to its own conceptual environment. In contrast, an emic stance would allow no more than faithfully bringing to life the ideas of the historical subject in terms as close to the subject’s familiar intellectual vernacular as possible. I think Sutton’s ‘etic’ methodology yields a fruitful perspective. However, rather than deciding the underlying methodological issues a priori and on principle one had better discuss such matters on a case–to-case basis in terms of the historiographical harvest actually obtained or in terms of the validity, the plausibility and the novelty of the resulting historiographical insights.

Naturalism
Sutton’s one-sided emphasis on Descartes’ theory of corporeal memory as well as his general resolve to read pertinent Cartesian texts from an overall connectionist perspective are two cases in point which would seem to presuppose an etic methodological stance in historiography for their justification. Now these two points may seem to Sutton to shade into a third one, which thus might be held to be analogous, yet which I hold to be distinct. It concerns the question whether or not we are entitled to adopt a naturalist stance in our interpretations of the Cartesian corpus, as Sutton advocates (in good company). Of course, if one takes for granted the traditional philosopher’s interpretation of Descartes according to which rationalist anti-naturalism is already part and parcel of Cartesian philosophy, then a naturalistic approach towards Descartes’ philosophy could be justified only on ‘etic’ methodological grounds. However, as I have amply argued elsewhere (Meyering, 1989), there are good grounds for rejecting the premise of this argument. For anti-naturalism is not only (a corollary of) an untenable dogma of analytic philosophy (Quine, 1953, 1971), it also constitutes a seriously distorting perspective if applied to the historiography of philosophy. So on this point I am in wholehearted agreement with Sutton when he criticizes anti-naturalist history which “either neglected Descartes’ science or treated it only as an adjunct to first philosophy.” (p. 51) Similarly, Sutton is completely right when he seeks “to counteract historiographical obsession with metaphysics” (p. 51). The only, minor, point I would wish to make, then, is that I wouldn’t call this a question of “retrospective reading” (ibid.; italics mine), but rather a question of the correct approach towards the history of philosophy in general, and certainly towards Cartesian philosophy in particular. In brief, this is not a question of (etic or emic) methodology, rather it is a question of historical content.

Coherence
Notwithstanding my positive assessment of the legitimacy of Sutton’s approach in general, his skewed his-torical slant on the topic of Descartes’ theory of memory and his tendency to portray Descartes as a latter-day neurophilosopher faces, as one may well expect, virtually insurmountable difficulties. For how are we to reconcile the Descartes of the Meditations with Descartes the neurophilosopher and the psychofysiologist of L’Homme as seen through Sutton’s eyes? Are we to relegate what are traditionally taught to be crucial tenets of Cartesian philosophy — Descartes’ metaphysical dualism, his seemingly professed rationalism, his expositions on intellectual memory, his view of body as purely extended and thus as utterly inert, etc. — are we to relegate all that to that perennial subterfuge of the rational reconstructor’s embarrassment, viz. ideological mauvaise foi or dissimulation on Descartes’ part? It is not that Sutton’s interpretation of Descartes’ scientific work is simply falsified by such contrary observations. It is just that Sutton surely owes us an account of how the various seemingly incongruent pieces of Descartes’ intellectual legacy fit together into a coherent whole.

Indeed Sutton’s reading of Descartes yields a fascinating mirror image of the received view. Formerly we had a univocal rationalist philosopher Descartes whose message was that of metaphysical dualism. Now we are presented with an equivocal or ‘dualist’ Descartes, the crypto-neurophilosopher, whose true message is distributed processing in the brain! Yet Sutton’s mirror image stands in danger of making the mistake opposite to that made by traditional interpreters, viz. that of now treating Descartes’ metaphysics as a mere adjunct to his scientific theorizing! Are we then to believe, perhaps, that Cartesian metaphysics was designed merely to please the theological Faculty of the Sorbonne and that it should not be regarded as deeply relevant to a reliable picture of Descartes’ true philosophical significance? That would certainly burden our credulity a little bit too much. Fortunately, that isn’t Sutton’s answer, to be sure. But for all the illuminating answers Sutton does give (pp. 67-99) — and highly illuminating and intriguing these answers certainly are — the final upshot of his response is an admission of a fantastic incoherence, a “key historical crux in Descartes” (p. 101), a “terrible internal pressure” in his theory (ibid.), a “recurring desire [on Descartes’ part] to distance himself from his own theories of physical-cognitive processes like remembering” (ibid.). On Sutton’s account it is Descartes’ longing for the notion that somebody or something is in charge, capable of asserting control over the wayward body, that forces him to postulate “the fragile authority of a central executive, whether the incorporeal intellectual memory or the rolling pineal gland.” (p. 100). In other words, Descartes’ insuppressible need for inner discipline, for a rational agent and for a locus of global control makes him piously mention the importance of intellectual memory and succumb to the posit of a central executive, incompatible with the distributed model of memory, with the autonomous “folding, sieving, and commotion of the physiological fluids and fibres” (p. 100), and with the “winds and flames of the animal spirits” (p. 101).

However, one may well wonder whether this is sufficient explanation for the mighty metaphysical edifice Descartes erects in the Meditations and elsewhere! Is a philosophical monument of such immense and self-assured proportions called for, and thus possibly explained, simply by a certain lingering unease on Descartes’ part over a possible lack of executive control and inner discipline? It is hard to believe how such an exuberant celebration of the rational soul as orchestrated in Descartes’ metaphysical writings, with virtually every statement in the ostensibly deductive development of Descartes’ argument in explicit contradiction with what the crypto-neurophilosopher Descartes would really have us believe about the inner workings of the mind/brain, could have been instigated with such seemingly unshakable confidence and in such elaborate detail if the motive for it was apparently so slight while the felt sense of doctrinal incoherence must have been so acute and agonizing, especially in a thinker of Descartes’ measure of rigor and clarity!

Executive Control
Apart from these larger issues Sutton’s interpretation also raises more detailed questions as to the relationship between the kinds of control exercised by the rational soul and the pineal gland respectively. To be sure, Sutton has very interesting things to say about self-mastery (pp. 99-102) and about control by the rational soul (pp. 80-81). But how exactly does this kind of executive control relate to the quite different ‘control’ exercised by the pineal gland? Though the latter kind of ‘physical-cognitive’ control is the more interesting from the modern point of view, it is not clear how it could really contribute much to bridging the gap between body and mind, or between physical and rational cogitation — assuming, that is, rather unconventionally, that that was what Descartes was after in the first place! Sutton’s intriguing suggestion (pp. 76ff.) that Descartes in fact interposes ‘complex automatism’ in between simple automatism on the one hand and true action on the other doesn’t seem to be very effective in alleviating the tensions generated by his special neurophilosophical reading of Descartes. Complex automatism, on this view, would constitute a hybrid source of intermediary control over, possibly, long temporal gaps, yet would involve no more than the pineal gland without any control on the part of the rational soul being required. Yet it would seem that it is a little misleading to speak of control here. At any rate, the genuine control exercised by the rational soul seems to be quite opposed to the ‘control’ exercised by the pineal gland. For the former consists to a large extent in trying to unlearn some of the associative links set up by the automated conditioned responses in the pineal gland. So I am not at all sure it is correct for Sutton to say that Descartes, rather than conceptually isolating the patterns of conditioned responses emerging in the rolling motions of the pineal gland, in fact couples them with rational control and self-mastery and considers them equivalent! (p. 80)

Continuity Claims and Reference Potential
In conclusion let me touch upon a more general question in the history and philosophy of science. The connectionist framework Sutton invites us to map onto Descartes’ neurophilosophical theorizing surely provides a highly suggestive model for Descartes’ theory of how the animal spirits may operate in the brain so as to sustain corporeal memory and distributed representations in superpositional ‘storage’ over long periods of time. But how far does the analogy really go and what levels of detail should be required to map onto each other in order for the suggestion to be plausible or real? It is not as if Sutton’s account is evading such queries. On the contrary, the author quite frankly puts his cards on the table and the helpful diagram he provides us with (p. 156) has the very merit of inviting our scrutiny. Yet the items most conspicuously missing from the range of the various mappings suggested in the diagram precisely concern such crucial features of the connectionist architecture of the brain as the activation values of the processing units and the related mechanism for plasticity (note 3). As is well known, in neural networks these features are implemented by spiking frequencies and connection weights. Now, surely, it would be completely unfair to evaluate Sutton’s interpretation by requiring that Descartes’ brain model include even an inkling of the biochemical mechanisms only 20th century brain research has finally managed to reveal to us. Yet it would seem that Descartes’ model should at least provide corresponding mechanisms characterized at some more abstract functional level of description mapping onto these key elements of connectionist theory. Lacking such mappings the proposed analogy doesn’t seem to be sufficiently ‘anchored’ in crucial details of intellectual reality.

This issue nicely ties up with a broader philosophical question concerning theoretical continuities and discontinuities in general. In a chapter entitled ‘the puzzle of elimination’ Sutton discusses the question why animal spirits theory was abandoned in the course of the 18th century while that theory was not just an im-pediment to neuroscientific progress (and never had been). Some authors have suggested animal spirits simply had to be eliminated in the same way that witchcraft or locus naturalis had to go. By contrast, other historians have stressed animal spirits’ continuity with later research by Galvani et al. on animal electricity, claiming that animal spirits were simply reduced to, in the sense of being identified with, neural electrical fluids. Indeed, as Sutton observes (p. 217), it was Galvani’s own view that his research actually explained the nature of the animal spirits.

Now notoriously, debates surrounding conceptual change have raged during the past three decades or so amongst philosophers of a relativist vs. those of a realist bent, each party having invoked correlative semantic theories in support of their claims. Now that the dust has settled somewhat, more pragmatic views seem to have emerged. Thus Kitcher (Kitcher, 1978, 1993) has stressed the contextual nature of determining reference. According to this view, terms have associated with them a heterogeneous reference potential, so that reference is fixed differently in different contexts. This opens up a much more dynamic picture of reference fixation than the classical options would allow. Instead of the “totalitarian” or “inflexible” views (Fine, 1975: 26) implicit in causal-historical and descriptive theories of reference alike, this more sophisticated dynamic view of reference would render claims of theoretical (dis)continuity a matter of contextual and historiographical relevance and evaluation. Consequently, in my view, the scope of such claims would in principle be widened considerably, depending on what characteristics associated with a term one would wish to select as essential for the purposes of historical explanation as well as on what level of abstraction one would wish to highlight for the historiographical task at hand. To be sure, the coarser the grain of analysis or the more general the set of associated characteristics, the looser the correlative claims of continuity or discontinuity, of reduction vs. elimination. Even so, such claims could still yield crucial insights in the dynamics of intellectual history to which the less flexible classical theories of reference would be blind.

I am not sure to what extent Sutton implicitly concurs with this line of reasoning. One indication that he does is when he proceeds to take into account wider spirituous domains (such as culturally permeated ‘descriptions’ of the cognitive phenomena, or the 18th century emergence of homo psychologicus and of the Enlightenment ideal of ‘moral Man’) within the context of his discussion of reduction vs. elimination of animal spirits during the 18th century (pp. 218ff). Needless to say, I applaud this move, and believe it to be legitimate. Indeed, Sutton takes a cue from Edwin Clarke (Clarke, 1968: 139-41), who ascribes the 20th century reintroduction of dynamic conceptions to its tie with the 17th century and with ‘the Baroque demand for movement’ (quoted on p. 218). Similarly the neurobiologist Jean-Pierre Changeux chose ‘Animal Spirits’ as title for the chapter of Neuronal Man which seeks to replace ‘a static description’ of cerebral wiring with ‘knowledge of a different, dynamic order’ (Changeux, 1985; quoted on p. 218).

Now I submit all of this pertains in a profound sense to an evaluation of the main thesis of Sutton’s own book. For, if anything, Sutton’s account of Descartes’ theory of memory itself involves an exciting continuity claim, albeit one of a very general and abstract nature. As I understand it, the point of Sutton’s book is not just to draw attention to a striking structural analogy he discerns between Descartes’ neurophilosophy on the one hand and modern views of memory and of distributed representation on the other. It also emphatically implies the historical thesis that “there is a longer background to more dynamic views of memory than is commonly acknowledged” (p. 15). To evaluate that thesis, then, becomes, in view of what I have just said about an expanded theory of reference potential, a delicate question not just of weighing the historical evidence, but rather of deciding whether certain abstract but highly general, and thus possibly superficial, structural similarities between the two brain processing models under consideration outweigh the striking absence in Descartes’ theory of memory of convincing analogues to such key elements in the connectionist story of parallel distributed processing as activation values and connection weights. That decision Sutton’s otherwise fascinating historical account has left insufficiently motivated and compelling, I am afraid, notwithstanding its highly original insights and its numerous thought provoking aperçus.

Notes
1. John Sutton, Philosophy and Memory Traces: Descartes to connectionism. Cambridge: Cambridge University Press, 1998. References by page number only all refer to this book. [Back to text]
2. Krell refers to animal spirits as those 'ultimate oxymorons' (Krell 1990: 5, quoted on p.23). [Back]
3. Sutton's vague suggestion of 'microstructure' as an analogue for connection weights doesn't seem to amount to much more than empty hand-waving. [Return to text]

References
Changeux, J.-P. (1985). Neuronal Man: The Biology of Mind. New York: Pantheon Books.
Clarke, E. (1968). The doctrine of the hollow nerve in the seventeenth and eighteenth centuries. In Stevenson, L.G. & Multhauf, R.P. (Eds.), Medicine, science, and culture, (pp. 123-141). Baltimore: Johns Hopkins University Press.
Fine, A. (1975). How to compare theories: Reference and change. Noûs, 9, 17-32.
Harrington, A. (1987). Medicine, Mind, and the Double Brain: Princeton University Press.
Kitcher, P. (1978). Theories, Theorists, and Theoretical Change. Philosophical Review, 87, 519-547.
Kitcher, P. (1993). The advancement of science: Science without legend, objectivity without illusion. New York and Oxford: Oxford University Press.
Krell, D. F. (1990). Of Memory, reminiscence, and writing: On the verge. Bloomington, Indiana: Indiana University Press.
Meyering, T. C. (1989). Historical Roots of Cognitive Science. The Rise of a Cognitive Theory of Perception from Antiquity to the Nineteenth Century. (Vol. 208). Dordrecht-Boston-London: Kluwer Academic Publishers.
Quine, W. V. O. (1953). Two Dogmas of Empiricism. In Id., From a Logical Point of View, (pp. 20-46). New York: Harper & Row.
Quine, W. V. O. (1971, ). Epistemology Naturalized. Paper presented at the XIVth International Congress of Philosophy, Vienna: Herder & Co.
Sutton, J. (1998). Philosophy and Memory Traces. Descartes to Connectionism. Cambridge: Cambridge University Press.

TOP OF PAGE

To return to the index page for Philosophy and Memory Traces, click here.

Back to John Sutton's home page.