The common usgae "I predict event A with probability 0-8 and event B with probability 0-2", is not exception. This statement is a shorthanded version of "I predict (with certainty) that the value of a variable called the probability of A, namely p(A) equlas 0-8, and the value of a corresponding variable for B, namely p(B) equlas 0-2". In other words, we are not predicting events, but certain abstract entities called the probabilities of events which can be variously interpreted, for example, in the present case, as an assertion that if either A or B (but no other event) were able to occur upon many occasions, 80 percent of the time the occurence would be A, and 20 percent of the time it would be B. Thus, it follows from our axiom, that we do not make predictions about a piece of the real world, an "assembly" as such, which is unknowable in detail. Rather, we make predictions about some simplified abstraction from the real world -- some incomplete image -- of which we can become certain (the probability model is, of course, an abstraction of this kind). Subject to some important qualifications, which will appear in the discussion, this simplified abstraction is a "system". (ATC 19)

What is an observer uncertain about? In the first place an observer, absurd as it sounds, may be uncertain about his objective, that is, about the kind of predictions he wishes to make. This is rarely the case so far as a scientific observer is concerned. A scientist usually knows whether he want to make clinically useful observations, commercially useful observations, or observations compatible with the hypothetico-deductive structure of physics. On the other hand, there are cases of dilettante observation, where the objective is not obvious at the outset and only becomes so when some tentative knowledge has been gained. This situation is not readily analysed, for we can only speak about a source of uncertainty relative to some objective or other, i.e. clinical, commercial or physical prediction making and for the moment we shall deal exclusively with those cases where the objective is specified. (ATC 19-20)

The third level of uncertainty encountered in cybernetics is very much more difficult to understand, and is possibly analogous to the principle of indeterminacy of physics. As is known from quantum mechanics, the very measurement of the value of one variable of a microphysical entity produces changes in the value of another of its variables. Thus, for example, regardless of the accuracy of the measuring instruments which might be developed, it is impossible to know both the position and the momentum of an electron exactly, because whatever means are taken to get close to the first value will drive accuracy out of the other value according to a known relation. It may be that the uncertainty resides solely in the process of observations it may be there is a deeper significance in nature. At least, here is an uncertainty of a third kind. (DC 260)

So far as I am aware, there is no skill which is not based on knowledge; conversely, it is hard to conceive of knowledge that exists independently; all knowledge is tied to a skill in the sense that it is useful in acheiving some goal and (apparently, at any rate) in the sense that it is registered as part and parcel of a goal directed system. To be even more emphatic, I do not believe it is possible to "learn facts" without a goal or goals, the achievement of which is contingent upon knowledge of these facts. (TM 5-6)

As orienting comments, I conjecture that generalized intellect is possible and that it does not preclude those affective components of mind that are often brushed aside when machinery is brought into the picture. I definately maintain that a substantial part of the human mind can be encoded for execution in machines more general than a human brain. By way of summary, cognition in the present and broad sense of the word, is a general process. Biological cognition is a rich and important, but nevertheless, very specialized example of it. (CCL 3)

Home Cybeneticians Gordon Pask Beliefs, Models, and Analogies

Symbolic Generalizations | Beliefs, Models, and Analogies | Values | Exemplars | Techniques | Guiding Questions