KIP and DENNETTs 'Consciousness explained'
AUTHOR: Dr. Gerd Döben-Henisch
FIRST DATE: Nov 27, 1995
DATE of LAST CHANGE: Nov 27, 1995


Methodological Presuppositions





A philosopher building an empirical theory of the consciousness?

The aim of the book is to explain consciousness within the framework of contemporary physical science (p.40). And later on he adds that he will setup an empirical theory of the consciousness (p.98).

This is a bit surprising, because for me DENNETT is a philosopher and if he now makes the statement, that he wants to set up an empirical theory, then I ask myself, whether he wants to identify philosophy and empirical science? And if so, then there arises the next question, why not to abandon with the term philosophy at all? If philosophy is the same as empirical science, then we should not use the term philosophy any longer; it would be confusing.

But here comes DENNETT himself: "So, since as a philosopher I am concerned to establish the possibilities (and rebut claims of impossibility), I will settle for theory sketches instead of full-blown, empirically confirmed theories. A theory sketch or a model of how the brain might do something can turn a perplexity into a research program: ... such a sketch is directly and explicitly vulnerable to empirical disproof, but if you want to claim that my sketch is not a possible explanation of a phenomenon, you must show what it has to leave out or cannot do; if you merely claim that my model may well be incorrect in many of its details, I will concede the point." (p.41)

Thus, more to the point would be the formulation that DENNETT, as a philosopher, wants to work out the general possibilities of explanations of the consciousness which are at least not in conflict with the accepted empirical theories. Then one could -and should- ask, what in this case would be the special contribution of philosophy to the debate different to what the natural sciences are offering.



Defeat Dualism

While introducing his own position DENNETT mentions dualism. Dualism is the idea of mind as distinct from the brain, composed not of ordinary matter but of some other, special kind of stuff (p.33). In consequence of this assumption it follows that mind and body (brain) must interact: the brain must inform the mind and the mind must direct the body. This will lead to direct conflict with the laws of standard physics, especially with the principal of the conservation of energy (cf. p.35). Moreover some basic enlargements of the ontology of the physical sciences would be called for in order to account for the phenomena of consciousness (cf. 36). And DENNETT confesses that he -although this is an apparently dogmatic attitude- will avoid in his book Dualism at all costs (p.37).



Materialism

If one excludes a special principle distinguished from matter, then one has to cope with one principle alone: matter. The only sort of stuff we know, is the physical stuff of physics, chemistry, and physiology, and the mind must then somehow be nothing but a physical phenomenon; the mind reduces to a function of the brain (cf. p.33).

But DENNETT does not favour materialism: "Adopting materialism does not by itself dissolve the puzzles about consciousness, ... Somehow the brain must be the mind, but unless we can come to see in some detail how this is possible, our materialism will not explain consciousness, but only promise to explain it, some sweet day" (p.41f).

What does this mean? DENNETT excludes above vividly a second principle besides matter, but he hesitates to state plainly, that then the mind can only be seen as some effect/ function of the brain. What is the missing alternative?

Even if materialism cannot provide a detailed explanation today, it exhibits a general principle which either is true or not. DENNETT himself claims not to provide detailed empirical theories but only sketches of theories to explain the principal possibilities. So, why is he opposing to materialism? What is the third position between dualism and materialism?



Behaviorism

Wheras dualism and materialism are defined through ontologies -presuming the existence of two or one kind of basic stuff-, behaviorism is defined through methodology: behaviorism looks to its subject only from a 3rd-person perspective and does not accept data gained by introspection. In a radical version it excludes explicitly 'mental states' or even a special mind-principle besides matter. Seen from a philosophy of science point of view it is not a contradiction to be methodologically a behaviorist and with respect to ontology making the assumption that there are mental states which are -perhaps- not describable with behavioral means (this last position is near to an epiphenomenalistic position: assuming mental events, which have no physical effects, thus science can`t study them).

Nevertheless it is a problem to speak within behaviorism about mental states. If one denies dualism, then mental states can only be certain functions of the brain, i.e. mental states have to be seen as physiological processes, which can not easily be observed 'from the outside'. And it arises the question what it could be that makes some physiological processes to 'mental' processes? the adjectiv 'mental' presupposes some concept of 'mind', otherwise it has no clear meaning. How would a behavioral theory of the mind looks like?

Without saying clearly, why, DENNETT rejects behaviorism. Nevertheless he proposes an improvement. Presupposing that there are mental states he claims, that it is indeed possible to study mental states scientifically, i.e. from a third-person view (cf. p.71).

Now, what does this mean? What distinguishes DENNETTS proposal e.g. from a behavioristic theory? DENNETT (1) excludes a special mind-stuff, thus opposing dualism; he (2) claims to look to the subject of investigation only from a third-person point of view, thus repeating empirical science in general and also behaviorism; but he is (3) explicitly presupposing mental states in a way that they are describable and explainable by empirical sciences.

Where are these mental states coming from?



Phenomenology

Phenomenology is like behaviorism a methodological attitude: in opposition to behaviorism does phenomenology look to reality from a first-person point of view. By this we have a priviledged access to the content of our conscious experiences, i.e. there exist kinds of experiences to which we are only related from the first-person point of view; if you depart from this view, you will loose these experiences as they are.

There are different approaches also within phenomenology. In socalled 'pure phenomenology' one attempts to give a pure, neutral, pretheoretical description of what we find 'given' in the course of everyday experience (p.70).

DENNETT warns against an uncritical adherence to phenomenology. "What we are fooling ourselves about is the idea that the activity of 'introspection' is ever a matter of just 'looking and seeing'. I suspect that when we claim to be just using our powers of inner observation, we are always actually engaging in a sort of improptu theorizing..."(p.67).

The priviledged access by introspection does not include a direct access to the causes, which are responsible for the experiences as they are, and not to the effects, these experiences have in the body.

This leads DENNETT to the propagation of heterophenomenology (see below). But one can ask, whether this should be all what could be said about phenomenology? To say that introspection can be misleading if used uncritically is a commonplace which also applies equally well to a third-person view, and that it is very dangerous to construct theories related to data is a tautology as well. The only interesting question is, whether a phenomenological approach 'in principle' can not lead to useful theories about our world or not. The discussion of this point do we postpone to the end of these reflections.



Heterophenomenology

For DENNETT is heterophenomenology "a neutral path leading from objective physical science and its insistence on the third-person point of view, to a method of phenomenological description that can (in principle) do justice to the most private and ineffable subjective experiences, while never abandonning the methodological scruples of science" (p.72).

The usage of the term 'phenomenological description' in this cited passage is critical. Only if DENNETT understands it in the more general sense of 'describing events with the descriptive vocabulary of some scientific discipline' without a special relationship to subjective experience, then is this usage compatible with the assumption of a third-person point of view. But then it is difficult to see, how he can cope with the most private and ineffable subjective experiences of people, because these are only given in the manner of the subjective experience itself.

He states explicitly, that "the method I describe makes no assumptions about the actual consciousness of any apparently normal adult human beeings" (p.73). And he continues: "since if consciousness is anywhere, it is in them"(p.73).

Here comes his solution. His main idea is, that it is possible to reconstruct subjective data of a certain human person A based only of recorded behavior (including speech) of A.
  1. The recorded data of A are the raw data.
  2. From these raw data three trained stenographers are preparing independently transcripts of the speech sounds resulting in a text (cf. p.74). The process of 'transcribing' recorded sound waves into a sequence of words of a certain language depends on assumptions about which language is being spoken, and some of the speaker's intentions (cf. p.75). The method, so far described, is 'cut-and-dried' and uncontroversial by scientific standards (p.76).
  3. "We must go beyond the text; we must interpret it as a record of speech acts; not mere pronounciations or recitations but assertions, questions, answers, promises, comments, requests for clarification, out-loud musings, self-admonitions. This sort of interpretation calls for us to adopt what I call the intentional stance" (p.76).
  4. "We must treat the noise-emitter as an agent, indeed a rational agent, who harbors beliefs and desires and other mental states that exhibit intentionality or 'aboutness', and whose actions can be explained (or predicted) on the basis of the content of these states" (p.76)
  5. "We can't be sure that the speech acts we observe express real beliefs about actual experiences; perhaps they express only apparent beliefs about nonexistent experiences" (p.78)
  6. "Our experimenter, the heterophenomenologist, let's the subjects text constitute that subject's heterophenomenological world, a world determined by fiat by the text (as interpreted) and indeterminate beyond. This permits the heterophenomenologist to postpone the knotty problems about what the relation might be between that (fictional) world and the real world. This permits theorists to agree in detail about just what a subject's heterophenomenological world is, while offering entirely different accounts of how heterophenomenological worlds map onto events in the brain. The subjects heterophenomenological world will be a stable intersubjectively confirmable theoretical posit..."(p.81).
Thus, what DENNETT seems to beliefe, is, that we can translate the subjective experience of a person A into an objective theory TH by taking the observable behavior of A by some Bs into a text T, which in turn then will be reconstructed by some Cs as the theory TH.

The decisive step is the step from the compiled text T to the interpreting theory TH.

With these assumptions DENNETT is, with respect to methology, clearly following the steps of modern experimental psychology, especially cognitive psychology. Restricting to what is observable in a third-person perspective, cognitive psychology tries to reconstruct 'cognitive processes' within the subject under investigation. The question to which extend one has to make special assumptions about special 'internal variables' or special 'internal processes' to make the theory fitting the observed data is up to the theory-building psychologist. To postulate 'intentions' is principially allowed, if one can show that such a claim allows the construction of a logically coherent and empirically sound theory. What the psychologist 'means' if he is using a term like 'intention' doesn't matter as long as he can 'embed' this term smoothly in his theoretical framework.

To say, that one can within such a framework reconstruct the 'subject's heterophenomenological world' is difficult to understand. If DENNETT means with this term the conscious world of the subject A as A sees itself in his first-person perspective, then one cannot see how a psychologist B can reconstruct this phenomenology of A only looking to A from the outside. The only solution could be to assume, that B is using his own phenomenology as a 'model', as an 'example' to interpret the data of A in the light of his knowledge of a first-person perspective. If so, then one could ask, why then should not everybody 'articulate his own subjective experience as a formal theory' directly, which then, in a next step, could be compared with other similar theories of the structure of subjective conscious experiences.

The last option states a clear methodological -but shurely not empirical- case. But to say, that one wants to reconstruct subjective experience without saying where to get the right data, is by far more dubious.

Thus as long DENNETT does not explain his position more cleary, either in the sense that he is accepting the label of experimental psychology, or either that he makes clear how he can establish relevant subjective data which are beyond the experimental psychological paradigm, it is not possible to follow him further more.

If DENNETT will stay with the psychological paradigm -and there is nothing to see, which makes his approach different-, then he suffers of the same problems as experimental psychology suffers. The restriction to observable behavior does allow the construction of theories with highly complex internal processes within an agent, but these complex internal processes are completely 'blind' assumptions, because there is no knowledge available about the 'real subjective' data. These assumptions are at best 'projections' 'inspired' by the 'knowledge' of the psychologist which is organizing the available data into a theory.





The KIPs point of view

Because questions of ontology are secondary compared to questions of epistemology we decided in the KIP not to take an ontologically defined position (e.g. materialism versus dualism), but an epistemological one: what is the 'natural' point of departure in the project of 'discovering reality'?

The natural starting point is shurely our conscious knowledge of our perceiving, of our thinking, of our feelings, of all that stuff which can be a 'content' of our consciousness, some distinguishable entity of our knowledge, that, what usually is called a phenomenon.

This 'dynamical stream of phenomena', constantly changing, covers uncountable many implicit relations, functions, laws, hidden structures, which only can be made explicit by a complicated process of explicit symbolic representations, abstractions, deductions, setting up of hypothesis, so called 'cognitive activities', which are not outside of the phenomenoloical perspective, but are a genuine part of it.

The 'limits' and possible 'falsities' of phenomenological cognition-processes -which have been mentioned by DENNETT- do not completely disqualify this kind of cognition; they only reveal the genuine complexity of the human endeavor of constructing an adequate theory of the everlasting process of cognition.

To abandon phenomenological cognition -as DENNETT proposes- and to find help and truth in the realm of empirical sciences is not only unnecessary, but principially impossible.

It is unnecessary because the paradigm of empirical science can be run as a sub-perspective of phenomenological cognition! The only difference between a phenomenological approach and an empirical approach lies in the restriction of the set of allowed phenomena. One can reconstruct the third-person perspective of empirical science as a first-person perspective where the set of allowed phenomena is restricted to a certain subset of socalled empirical phenomena. The selection criterium for this subset of empirical phenomena are certain properties of the phenomena together with some behavioral conditions. Thus the third-person perspective is in fact only a limited first-person perspective. The dichotomy of empirical, objective science and the subjective, 'mind-oriented' sciences is therefore only a methodological artefact without to much meaning; it is 'useful', but not more. And in this sense it is therefore principially impossible to fall outside of the phenomenological perspective.

How You can imagine is the research-strategy, we have choosen for the KIP, substantially different from the one DENNETT proposes. Instead of first denying that there is some object called consciousness and then having trouble to establish a workable relationship to that something, we accept, that the starting point of our process of cognition is already rooted in an individual consciousness, which has the option to examine 'consciously' the available stream of phenomena with the available 'cognitive operations' and 'tools' to built up a collection of partial theories, more or less explicit, more or less formal, more or less applicable to certain subsets of phenomena, especially to the subset called empirical data. The theory-productions of such an endeavor we take as basis for the construction of special tools called 'computer simulations'.

The ontological problem of 'mind' and 'body', of 'mind' and 'machine', vanishes, if one takes into account, that a human person, e.g. DENNETT, has a brain, which produces some 'inner view' induced by his several peripheral stimulations and 'internal' processing circuits. The way the brain can 'look to itself from the outside' is only by 'looking to the outside mediated by the inner view', i.e. the inner view is generating structures which represent the 'outside' from the 'inside'. The 'inner view' is naturally not a 'picture', or some other kind of 'concrete' object, but it is a collection of simultaneous states and complex processes, which all together are producing an 'inner view' which 'as inner view' has its own, specific reality, which is only given 'as such'. Seen from the 'inside' there is, strictly speaking, no 'inner' and no 'outer'; there is primarily only the allembracing 'inner reality' which can in itself be structured, classified, interpreted etc. in manifold ways. This inner view is selforganizing and selfsustaining.

Comments are welcomed to kip-ml@inm.de
INM

Daimlerstrasse 32, 60314 Frankfurt am Main, Deutschland. Tel +49- (0)69-941963-0, Tel-Gerd: +49- (0)69-941963-10