Francisco Varela
Ethical Know-How
Action,Wisdom, and Cognition
Stanford University Press 1999

Keywords: THE ETHICAL EXPERT - skilled behavior - intentional analysis - we acquire our ethical behavior in much the same way we acquire all other modes of behavior: they become transparent to us as we grow up in society. This is because learning is, as we know, circular: we learn what we are supposod to be in order to be accepted as learners - socialization process - self-conscious or intentional action and self-less or intentionless action - non-unitary self - Computationalism in cognitive science embraces the idea that the self or cognizing subject is fundamentally fragmented or non-unified simply because it postulates mental or cognitive processes of which we are not only unaware, but of which we cannot be aware - unified "point of view," a stable and constant vantage point from which to think, perceive, and act - the self: consciousness is its central feature.


23
THE ETHICAL EXPERT
My main point in the First Lecture was that philosophers and scientists who study the mind have grossly neglected skilled behavior, which is immediate, central, and pervasive, in favor of exploring deliberate, intentional analysis

To counter this imbalance we must impress on ourselves how much of our lives is spent in skilled behavior - working, moving, talking, eating - and how little is spent in deliberate, intentional analysis. Yet it is this latter category that we notice. It is this latter category which has been the focus of philosophers and scientists alike.

It is also clear that we can add responding to the needs of others to our list of skilled behaviors without doing violence to our concept of ordinary life. And if that is so, then it should also be clear that the situations in which we exercise ethical expertise far outnumber those in which we must exercise explicit ethical deliberation.

Nevertheless, even the most subtle of modern writers on ethics continue to tell us that the central issue is reasoning. For instance, no less a light than Alaisdair MacIntyre, in After Virtue, concludes from a reading of Aristotle's Nicomachean Ethics that the moral agent is best described as a competent performer deliberately choosing among maxims: "In practical reasoning the possession [of an adequate sense of the tradition to which one belongs] ...appears in the kind of capacity for judgement which the agent possesses in knowing how to select among the relevant stack of maxims and how to apply them in particular situations."

The fact that this pronouncement translates easily into information-processing jargon betrays the inability of this kind of approach to appreciate immediate coping, which lies beyond the grasp of computationalist description, regardless of the details of its terms of analysis. 

Only the enactivist view-point, which I introduced in the First Lecture, can account for immediate coping.
Furthermore, we acquire our ethical behavior in much the same way we acquire all other modes of behavior: they become transparent to us as we grow up in society.  This is because learning is, as we know, circular: we learn what we are supposed to be in order to be accepted as learners. This socialization process has roots too profound for analysis here.

Still, we can say here that when the matter is viewed in this light, it is clear that an ethical expert is nothing more or less than a full participant in a community: we are all experts because we all belong to a fully textured tradition in which we move at ease. 

pg 35-40
Thus we can distinguish between self-conscious or intentional action and self-less or intentionless action. At first the idea of action without intention seems absurd, but in fact our lives are full of intentionless actions. We dress, we eat, and more important, we exercise consideration for oth-ers. We do all these things without intention, but we do not do them randomly or purely spontaneously.

We do them without intention because we are experts at them. Through appropriate extension and attention and by training over time we have transformed these actions into embodied behavior. But just what is the key element that makes such intentionless learning possible? The answer is right in front of us. Our microworlds and microidentities do not come all stuck together in one solid, centralized unitary self, but rather arise and subside in a succession of shifting patterns. 

In Buddhist terminology this is the doctrine, whose truth can be verified by direct observation, that the self is empty of self-nature, void of any graspable substantiality. Once we are fully able to ride with the enormous openness contained in this sunya of self, the possibilities for further self-understanding become both vast and immediately accessible. This point is crucial. It is the golden thread that unites our self-understanding with an external and scientific account of mental functioning.

ON NON-UNITARY COGNITIVE SELVES
To make this non-unitary self meaningful in terms of our own tradition and from our (Western) perspective, I need only turn to modern cognitive science. Yet we need not confine ourselves to any single trend within the field of cognitive science, for even the more conservative view-points in the field, the classical computationalist perspective, for example, deny the existence of a solid, centralized, unitary self.

Computationalism in cognitive science embraces the idea that the self or cognizing subject is fundamentally fragmented or non-unified simply because it postulates mental or cognitive processes of which we are not only unaware, but of which we cannot be aware. 

In fact, computationalism postulates mental (not just physical and biological) mechanisms and processes that are not accessible to the "personal level" of consciousness, especially self-consciousness. In other words, one cannot discern in conscious awareness or self-conscious introspection any of the cognitive structures and processes that are postulated to account for cognitive behavior. 

Indeed, if cognition is fundamentally symbolic computation, this discrepancy between "personal" and "sub-personal" imrnediately follows, since presumably none of us has any awareness of computing in an internal, symbolic medium when we think.

It is possible to overlook the depth of this challenge to our self-understanding, largely because of our post-Freudian belief in the unconscious. There is a difference, however, between what we usually mean by "unconscious" and the sense in which computationalism means that mental processes are unconscious. We usually suppose that what is un-conscious can be brought to consciousness - if not through self-conscious reflection, then through a disciplined form of introspective analysis such as Freudian psychotherapy. Computationalism insists on the existence of mental processes that cannot be brought to consciousness at all. Thus it is not that we are simply unaware of the rules that govern the generation of mental images or of the rules that govern visual processing; we cannot in principle ever be aware of these rules. One computationalist account asserts that these processes cannot be brought to consciousness without ceasing to function, because by its very nature consciousness is slow and deliberate, not fast and automatic as, say, vision must be to function properly. Another account describes these processes as "modalar," that is, as comprising distinct subsystems that cannot be "penetrated" by conscious mental activity. Thus in this sense computationalism challenges our conviction that consciousness and the mind amount to the same thing, or that there is any essential or necessary connection between the two.

Of course, Freud also challenged the idea that the mind and consciousness are the same. Furthermore, he certainly realized that to distinguish between the mind and consciousness entails the disunity of the self or cognizing subject, a point to which we shall turn shortly. It is not clear, however, whether Freud took the further step of calling into question the idea that there is an essential or necessary connection between the mind and consciousness. Freud, in his argument for unconscious beliefs, desires, and motivations, left open the possibility that these unconscious processes belonged to a fragment of ourselves hidden in the depths of the psyche. 

Although it is not clear the extent to which Freud meant such a fragmentation literally, it is clear that when cognitive scientists postulate a collection of fragmentary, nonunifiable processes, they mean exactly what they are saying. As Dennett puts it:"Although the new [cognitivist] theories abound with deliberately fancifal homunculus metaphors—subsystems like little people in the brain sending messages back and forth, asking for help, obeying and volunteering - the actual subsystems are deemed to be unproblematic nonconscious bits of organic machinery, as utterly lacking in point of view or inner life as a kidney or kneecap."

In other words, the characterization of these "subpersonal" systems in "fanciful homunculus metaphors" is only provisional, for eventually all such metaphors are "discharged" - they are traded in for the storm of activity among such selfless processes as neural networks or AI data structures.

Our pre-theoretical, everyday conviction, however, is that cognition and consciousness - especially self-consciousness - belong together in the same domain.
Cognitivism runs directly counter to this conviction: in determining the domain of cognition it explicitly cuts across the conscious/ unconscious distinction.  The domain of cognition consists of those systems which must be seen as having a distinct representational level, not necessarily of those systems which are conscious. Some representational systems are, of course, conscious, but they need not be to have representations or intentional states

Thus, for cognitivists, cognition and intentionality (representation) are the inseparable pair, not cognition and consciousness. This theoretical division of the domain of cognition was considered by early cognitive scientists to be an empirical discovery of no small importance and indicates, again, the remarkable transformation wrought by the cognitive sciences altogether. 

But now a problem arises: we seem to be losing our grip on something that is undeniably close and familiar - our sense of self

If consciousness - to say nothing of self-consciousness - is not essential for cognition, and if, in the case of cognitive systems that are conscious, such as ourselves, consciousness amounts to only one kind of mental process, then just what is the cognizing subject? Is it the collection of all mental processes, both conscious and unconscious? Or is it simply one kind of mental process, such as consciousness, among all the others? 

In either case, our sense of self is challenged, for we typically suppose that to be a self is to have a coherent and unified "point of view," a stable and constant vantage point from which to think, perceive, and act

Indeed, this sense that we have (are?) a self seems so incontrovertible that its denial - even by science - strikes us as absurd. And yet, if someone were to turn the tables and ask us to look for the self, we would be hard-pressed to find it.

Dennett, as usual, makes this point with flair:"You enter the brain through the eye, march up the optic nerve, round and round the cortex, looking behind every neuron, and then before you know it, you emerge into daylight on the spike of a motor nerve impulse, scratching your head and wondering where the self is.''

Our problem, however, goes even deeper. It is one thing to be unable to find a coherent and unified self amid the furious storm of "subpersonal" activity. This inability would certainly challenge our sense of self, but the challenge would be limited. We could still suppose that there really is a self, we simply cannot find it in this fashion. Perhaps, as Jean-Paul Sartre held, the self is too close, and so we cannot uncover it by turning back upon ourselves. The computationalist challenge, however, is much more serious. According to computationalism, cognition can proceed without consciousness, for there is no essential or necessary connection between the two. 

Now whatever else we suppose the self to be, we typically suppose that consciousness is its central feature. 

It follows, then, that computationalism challenges our conviction that this most central feature of the self is needed for cognition. In other words, the cognitivist challenge does not consist simply in asserting that we cannot find the self; it consists, rather, in the further implication that the self is not even needed for cognition.

At this point, the tension between what science affirms and our own immedlate experience seems to insist upon is tangible. 

If cognition can proceed without the self, then why do we nonetheless have the experience of self? We cannot simply dismiss this experience without explanation. Until recently, many scientists and philosophers of mind nonchalantly shrugged off this problem by arguing that the perplexities surrounding it are just not relevant to the purposes of cognitive science.

To make any further headway in our inquiry we must look more closely at the nature of this fragmentation. As I will discuss in the Third Lecture, the nature of this fragmentation is that of  emergent (or self-organizing) properties from brain mechanism, giving rise to what I shall term a virtual self, a mode of analysis which is very recent in cognitive science and Western thought altogether.
 
Francisco Varela: Ethical Knowhow - Cognitive agents - pg. 45
 
 

 


HOME