Semiotic Machines - an Introduction
The matter of private Experiences and their Communicability
AUTHOR: Dr. Gerd Döben-Henisch
FIRST DATE: Aug-6, 1996
DATE of LAST CHANGE: Sept-29, 1996
KANT was the first thinker who made in his Kritik der reinen Vernunft (Critique of poor Reason) (1981/87) the unbridgeable gap between the supposed objects of experience (things as such, 'Ding an sich' = X) and the data of conscious experience (= C_c) explicit. He introduced many different terms to describe the content and the form of the conscious and the unconscious (=transcendental) part of our knowledge. He speaks for example of the 'pure forms a priori of time and space', names the perceptive material 'appearances' and introduces terms like 'categories' and 'regulative ideas'.
But for the further discussion I will like to refer rather to the book Ideen zu einer reinen Phänomenologie und phänomenologischen Philosophie by Edmund Husserl (1913). It demonstrates one way, how one can try to speak about private experiences.
Condensed in a few sentences Husserl tells us that we should start our analysis of our whole experience with the experience as such. Everything what we can know as such is a conscious fact and is called a phenomenon.
He divides the field of phenomena in those which represent concrete material like sense-data, emotions, drives etc., packed under the label of matter or hyle, and in those which represent unifying forms which reveal general structures called Noema.
The important point here is that Husserl is convinced that everybody is able to recognize elements of his consciousness as such and that one can distinguish these elements according to properties inherent in these elements.
What Husserl does not take into account is the fact, that in every case, when somebody wants to communicate his findings, he has to use a language. In the extreme case, he would have to develop a language for this communication. And it is this language-dependency which makes the case of private experience not a trivial case.
In his Philosophical Investigations (Philosophische Untersuchungen) (1945) it is Wittgenstein who deals with the question of private experiences (Empfindungen) thereby reflecting about the special point of the language-dependency. In the richness of his fragmented thinking it is not possible to distil an absolut clear position, but there are numerous passages which unravel for me the following position:
(P1) He doubts that the different individual internal states are such, that one has a clear criterion to distinguish and recognize internal states in a satisfying way (cf. Wittgenstein 1945:258, 265, 322).
(P2) He doubts that an internal state which cannot be correlated with an observable habit can be communicated succcesfully with another person (cf. Wittgenstein 1945:256, 261, 269, 272).
(P3) One has private experiences; they are more then nothing, but they are less than a certain something (cf. Wittgenstein 1945:246, 304).
The postulate (P3) reflects a bit the dialectical character of his investigations: at one side he sees problems with the usual manner to speak about internal states of an individual, on the other side he will not abandon with the manifested experiences completely; they are 'more than nothing'.
Thus we will concentrate on (P1) and (P2).
I agree with (P2). If there are no habits or situational properties which correlate in a 'natural' manner with internal states of an individual A, then A has no chance to communicate his -then- private states with someone else.
I do not agree completely with (P1). There exists a substantial part of my internal states -how much?- where I claim to be able to distinguish these states, to recognize certain properties and to compare them sufficiently clear. For example I can imagine different numbers and formulas, can deduce some consequences, can remember situations and the like. Of course my internal thinking does not necessarily tell something about the 'empirical world around me' (Wittgensteins argument, that the consequences of a thought-experiment are not identical with the real consequences of that experiment (cf. Wittgenstein 1945:265)), but these 'internal states as such' -Husserls phenomena- are distinguishable and have decidable properties. This version of (P1) I call (P1+).
If (P1+) is true -what I assume-, then the problem of the communication of internal states reduces to the question, whether and how it is possible, that an individual A can make use of an observable habit or observable properties of the actual situation, to 'instruct' another individual B by these habits or situational properties, that he, A, is actually focussing on his internal state X_A and that he wants to name this state X_A with the wordtoken W.
The problem could be solved if one would add the following two postulates:
(P4) With regard to all 'important' internal experiences of human sign users one can assume, that they are structurally 'sufficient similar'.
(P5) Every 'important' internal state/ process is either connected in a 'natural' way with a certain habit or with certain properties of the actual situation.
Postulates (P4) and (P5) are very strong, but they have a strong backing in the biological facts of humans sign users as well as in the fact of daily communication. The biological apparatus of a human sign user is to such an extend similar, that it is highly plausible, that the types of experiences enabled by this apparatus should not be to different. There are numerous cases where we are communicating individual states very succesfully.
Let us have a short look to the example of recognition.
(1) For two individuals A and B it holds: A & B 'see' a single Object o with propertiy P1 at time t1 the 'first time' (P1(t1,o,A) & P1(t1,o,B)).
(2) For two individuals A and B it holds: A & B 'do not see' the Object o with propertiy P1 at time t2 > t1 'any longer' (~P1(t2,o,A) & ~P1(t2,o,B)).
(3) For two individuals A and B it holds: A & B 'see' a single Object o with property P1 at time t3 > t2 'again' and they can 'remember' that they have seen the object o with property P1 'before' (P1(t3,o,A) & P1(t3,o,B) & REMEM(A,P1(t1,o,A)) & REMEM(B,P1(t1,o,B))).
Under 'normal' circumstances A & B will report at time t3 that they have seen the object with property P1 already. If we could assume that they both had the internal experience of recognizing the object with property P1 as an event co-occurring with the presentation of the object itself, then in this case some property of the actual situation would function as a kind of a pointer or index with respect to the co-occurring internal processes in the individuals A & B. These internal processes have to be considered as 'objects in their own right'. Presupposing co-occurence of external event and internal event as well as similarity of induced internal processes it would be understandable that two individuals can introduce a new term w with a commonly shared designatum REMEM() consisting of two different internal processes REMEM(A,*) and REMEM(B,*).
After such an situation-mediated introduction of an objective-subjective descriptive term w the situation-properties Sit, which have assisted in the introduction of the term w, would in principle be superfluous for future use of the term w. At any time when A would utter w could B 'refer' to the encoded designatum D without being forced to rely on the original situation property Sit.
Comments are welcomed to doeb@inm.de
Daimlerstrasse 32, D-60314 Frankfurt am Main, Tel +49-69-941963-10, Fax +49-69-941963-22