A CRITERION FOR REJECTION OF OBSERVATION REPORTS?
In The Logic of Scientific Discovery, Popper endorses the following view of Reininger and Neurath, while cautioning his reader that the error of omission committed by these writers may lead to very grave consequences.
Observation reports, or protocol sentences, or basic statements, once affirmed or declared by the world of science to be true, need not always be so affirmed. Popper cautions his reader that unless criteria for the denial of a once affirmed basic statement are given, we are in danger of dogmatism, and in particular of the overthrow of empiricism, since without such criteria anyone is free to deny those basic statements which contradict his pet hypotheses.If Popper had said no more on this point, one might have raised the following question: ‘can we do without criteria for the denial of once- affirmed basic statements and not become dogmatic?’ We may have criteria which would disallow the denial of basic statements for the purpose of dogmatizing; we may avoid dogmatizing even without such criteria - say by sheer accident or by grace. Obviously, we may identify the criteria with the demand not to dogmatize: we may permit the denial of basic statements for any purpose and for any reason except the desire to uphold a theory. Our last criteria may not be good; for instance, dogmatism may turn out to be the unintended and unnoticed consequence of some other policy (such as pleasing the grand-old-man). This, however, is the fate of a criterion; moreover, even if criteria help to avoid one kind of dogmatism, adherence to those criteria may amount to some other kind of dogmatism.
The operative word in the preceding paragraph is the word ‘may’ : its purpose was to indicate a variety of relevant possibilities. All of these are implictly rejected - or so it seems - by Popper who says that ‘Neurath fails to give any such rules and thus unwittingly throws empiricism overboard’ (p. 97).
Of course there is only a subtle difference between ‘allows people to throw empiricism overboard’ and ‘throws empiricism overboard’. The fact remains, however, that I understand Popper to be advocating the view that we urgently need a rule or criterion for the denial of once-affirmed basic statements so as to avert dogmatism.Two or three questions press themselves on us here. First, ‘is the risk imminent?’ Secondly, ‘does Popper provide a satisfactory rule?’ And thirdly, ‘does this not lead to a higher level dogmatism?’
The risk of dogmatism may be understood in a variety of ways. It may mean that unless the legislature provides such a rule, the commonwealth will suffer. This is definitely a misinterpretation. Neurath and Popper are not arguing as legislators, and they know full well that the commonwealth of learning was faring fairly well without any rules such as the ones they were trying to formulate. On the contrary, in a sense Neurath and Popper, when discussing the issue at hand, may be viewed as social anthropologists who observe the commonwealth of learning like Malinowski was observing the Trobriand Islanders. It is a fact, one may hear Popper say, that the commonwealth of learning is non-dogmatic and empiricist, and I wish to find out or make a hypothesis about the rule they are following which keeps them from foundering in the wake of the dogmatist.
This interpretation, though much better than the first, is still not very satisfactory. I think the truth lies somewhere in between the two interpretations but I find it impossible satisfactorily to get matters clear. Therefore I do not know in what sense it is urgent - Popper says it is - to legislate a rule to tell us when we are permitted to deny the truth of a once affirmed basic statement, and why the absence of such a rule not merely permits dogmatism but even leads to it in some sense or another.
V.