QUESTION :

Dear professor Von Glasersfeld,

 

Thank you. Allow me to reply to your December answer to my question

related to the epistemological status of protocols.

 

First, in as far as "accepted protocols of data analysis" and "an

existing theoretical framework for the analysis of observations and

interviews" are not constructed out of an explanation of an unexpected

experience - for example inferential statistics isn't - they have

never lead to any viable conceptual structure or any viable theory.

Not anyone working by that method has, to my knowledge, come up with a

viable conceptual structure or viable theory. Why not? Because that

method is based on an unviable understanding of the method in the

natural sciences e.g. physics. The method in physics however did lead

to viable conceptual structures or viable theories. Why?

Natural sciences proceed by the method of explaining an unexpected

experience. Just as how Piaget worked.

 

Second, what expressions in my letter made you think that I was

"looking for something like absolute viability, which would make it

all too similar to ontological truth"? I wrote "and when this

explanation holds empirically (is viable) it is accepted as a theory,

as long as not a new unexpected experience comes up". So, I was

neither saying or impliying anything absolute! The expression

"insight" I used was followed by "knowledge" as a synonym, and thus

only meaning 'an explanation of an unexpected experience that holds

empirically (is viable) as long as not a new unexpected experience

comes up" and this makes the whole cycle going round and round again

(endlessly, who knows?): explaining an unexpected experience,

constructung an explanation for this unexpected experience, and

finding experiences (again nothing absolute implied by "finding")

which make this explanation viable, just for now!

 

Third, you wrote "Whether a conceptual structure or theory is

considered viable or not depends on the context and the purpose for

which it was constructed". That purpose cannot but be to explain unexpected

experiences; or do you have

examples of viable conceptual structures or viable theories that did

arise out of another purpose, say, to systematize (in an unexplanatory

sense) experiences?

 

Fourth, I surely do not want to bother someone applying knowledge of

limited viability - like that of the "flat earth theory" - with

epistemology. Did you read that anywhere in my letter? I do however

want to bother scientists who are applying unviable methods to

construct viable knowledge, in particular inferential statistics!! My

question for them is: did it lead to any viable conceptual structure

or viable theory? Is the basis for your method viable?

 

Piaget never and never again used things like inferential statistics.

I hated them.

 

Please respected comments.

 

R. Kooyman

The Netherlands

 

ANSWER :

Dear Mr. Kooijman,

I am no longer sure what you mean by "inferential statistics". I took it

to mean the practice of making inferences from the statistics of

observations. This is a common procedure leading to the formulation of

theories, for instance the theory that all swans are white, which will be

disproven as soon as a black swan is observed.

I don't know much about "accepted protocols of data analysis" and it

never struck me that experimental psychologists should construct them

"out of an explanation of an unexpected experience". It seems to me that

such protocols were established so that different

observers/experimentalists could compare their statistical results more

or less fairly. The inferences they draw from the statistics may or may

not lead the psychologists to formulate a theory and the viability of the

theory then depends not on the statistics, but on its usefulness within

the domain of observations that led to its formulation.

 

The intelligence test is, I think, a good example. The people who

designed it collected sets of questions the answering of which, they

thought, required intelligence. Then they statistically evaluated the

answers of students and produced an "intelligence quotient". The fact

that it does not represent what we normally think of as intelligence

does not impede its use and it will be considered viable as long the

users find it a useful way of classifying students.

 

One expression in your letter of December 8th that suggests you are

looking for "absolute" viability is the following: "when this explanation

holds empirically (is tenable, viable) it is accepted as a theory, as

long as not a new unexpected finding comes up." - Newton's theory of

gravitation and the structure of the universe is perfectly viable for

NASA's enterprises, although it has been superseded by Einstein's theory.

Also when dealing with light, physicists use two incompatible theories as

explanatory devices; both are viable depending on the experimental

context. (That's why I mentioned the flat earth theory.)

 

I do not see how you connect any of this with hypothetical explanations

of unexpected experiences. And I do not agree with your statement that

the purpose for which a conceptual structure or theory was constructed

"cannot but be to explain unexpected experiences". The concepts of levers

and pulleys and the mathematical theories to which they led were

constructed for the simple purpose of moving or lifting heavy objects.

The various theories of aerodynamics used in the building of cars and

planes were developed for the purpose of reducing air resistance, not for

explaining any unexpected events. (There was nothing unexpected about air

resistance, it is a common experiential fact.)

 

When he has an unexpected experience, a scientist will try to generate an

hypothesis of how it could have happened, i.e. he invents or, as Peirce

said,"abduces" a rule; but he can call that rule a theory only when he

has tested it. The results of these tests are inductively produced and

they may well include statistics.

 

Best wishes,

Ernst von Glasersfeld
 

BACK | TORNA INDIETRO