Reading Guide to: Habermas, J  'A Positivistically-Bisected Rationalism', in Adorno, T-W et al (1976) The Positivist Dispute in German Sociology, London: Heinemann

by Dave  Harris

Habermas's criticisms are aimed at '[positivism's] false consciousness... [of its]... correct practices' (page 198), including its lack of reflection on what counts as the boundary of science, and its tendency to banish other approaches beyond that boundary. It is important instead to self reflect, reveal the broader context for science, and demonstrate the possibilities of rational discussion. Popper's is a critique of positivism, but only 'at the first level' (199), demanding more self reflection. He still insists on preserving the status of 'facts', and does not reflect on technical and cognitive intentions, especially the 'pragmatic' ones inherent in positivism. Albert [who offers an earlier intervention in this Dispute -- don't worry about details but try to pick up the arguments] misunderstands Habermas on the following four issues:

The critique of empiricism.

Albert allows the point that experience has a crucial role in the development of science, but demands only that experience results in hypotheses which then become testable -- as soon as this stage has been reached, the role of experience is restricted. However, it is clear that tests have a pre-defined role here, and the whole procedure is viewed very narrowly. For example, psychotherapy offers a non-empiricist test procedure in human sciences, which clearly is an alternative. Partly, the problem lies in disagreeing about what counts as experience in the first place, what should be questioned, and what should just be accepted -- 'What I regard as a problem, he [Albert] continues to accept without question' (201).

Popper's stance is different. He takes on the claims of other positivists that there can be immediate sense certainty, claims which have been criticised since Kant demonstrated the importance of prior categories in informing such certainty. There are always interpretations, in other words, and 'statements of fact' are every bit as hypothetical as the hypotheses they are supposed to test (202). Popper knows this and develops a radical role for conjectures, arguing that science can never be finally accepted. He also sees verification as a process involving some authoritative prioritisation of sense data [which is agreed in the scientific community]. But the standards of falsification are really grounded themselves in some original 'critical tradition' which is not examined. In science, this critical tradition is itself deeply positivist, insisting on some independence of facts and theories, and promising to be able to run independent tests. Popper does agree that facts can only be grasped by theories, that facts are some product of language and reality, but runs into difficulties with his 'correspondence theory' [which argues that science eventually must correspond to agreements about what is known objectively about the world] -- if this is so, argues Habermas, we should simply test theories in terms of their correspondence, not bother with decisive testing. To be consistent, Popper should see 'facts' as things produced by theories which become fetishised, a process sometimes concealed by some transcendental view -- but he stops short of doing this.

Pragmatic interpretation of the difference between analytical and empirical modes of inquiry.

Popper offers the idea of a 'basic statement' as the basis for some decisive test of hypotheses [roughly, a 'basic statement' is a minimal statement that any reasonable member of the scientific community can agree with, such as 'that there is a moon in the sky']. These are single statements, but universals are clearly presupposed, and an agreement is based on the acceptance of rules. These rules themselves gain their force from being embedded in institutions, rather than possessing some universal logic. The whole argument shows that the crucial area of the empirical fact is 'established in advance by means of theoretical assumptions and testing procedures [which express some intent or interest]' (205). There are no separate 'facts', but rather a interplay of argument and empirical test, guided with a 'view to controlling predictions', and 'an implicit pre-understanding of the rules of the game' ( 205).'Applying' Popper's procedures means invoking these pre-understandings, in an act of hermeneutics [a procedure in social science which positivism rejects as 'unscientific' , of course].

The whole thing depends on an understanding of the research process in general as well as the specific aim -- but why not continue to reflect upon this process itself? The universal properties of this general level have been hypothesised before any empirical evidence gets gathered. What makes it really universal is that it  'corresponds to the elementary requirements of behavioural stability [in the progressive nature of science]. Feedback-regulated actions can only be secured for a long period of time if guided by information as to empirical regularities' (206) [ That is, science appears as universal, global even, because of its success in generating technological applications].

Habermas's own notion of a 'technical cognitive interest' incorporates all of Popper's criticisms and avoids the weakness of his argument about falsification. It also explains other interesting issues, such as the evident temporary nature of scientific knowledge, but the permanent utilisation of scientific techniques: this provides evidence of the regularity and feedback in action and how technology helps to overcome the logical misgivings of science (207). This is the real way in which science gets verified [in the development of technology]. Popper is far too interested in technique to see this [Kuhn doesn't exactly spell this out in his analysis of commitment to scientific paradigms either]. Scientific assumptions are held to be untrue if they fail to guide feedback-regulated action, as much as if they are falsified by other scientists (208). It is the information that theories yield, not the theories themselves, but are of interest to other scientists. Technological implications may not be evident in the research process, but they are built in to the business of testing propositions and applications. Indeed, there is no pure theory -- the relevance to instrumental action permeates the very choice of problems to be solved (209).

Critical justification and deductive proof.

Albert accuses Habermas of confusing science's logic of inquiry with its links with social contexts. Such a distinction can be traced to a post-Kantian split between the psychological and the logical. Although it is quite correct to separate questions of genesis from questions of validity, positivism reveals that there are significant connections between the logical and the empirical. Thus scientific criticism is never just formal or axiomatic. Instead, it shows a much more 'unreserved discussion of propositions' (210), where all the available techniques of refutation are used, not just test results, not even for Popper, and not just statements and their logical relations. It is possible to classify these into different types of refutation, but trivial to do so, especially if we are only trying to revive one of Kant's dualisms.

Popper talks of 'logically compelling deductions' as characterising science, but it is clear that he has additional normative arguments, and that he likes science for its reforming stance directed at old attitudes and standards of judgment (211). The hermeneutic role for science is as important as its deductive one: all distinctions, even the most fundamental ones in science, such as the split between logical and empirical argument, is subject to this kind of discussion. These distinctions presupposed criteria which are not just found in reality, and argument never just substantiates or falsifies them, but does something more like strengthening or weakening them. Methodological decisions in science are really 'quasi moral', not just logical, and therefore need justification as in the old discipline of rhetoric. Justification of this kind is never expressed in clear procedures and conventions.

Being critical involves attitudes as well, embracing a moment which transcends language and logic', and which cannot be deduced from the way scientists use language. There is a need to stay open to rational discussion according to the strength or weakness of propositions. Since arguments presuppose a rational motivation, they require rational attitudes. Even meta-theoretical arguments can also touch on empirical matters [disobeying the convention that separates them]. Popper prefers to describe science in terms of his logical structure rather than its argumentation in this fuller sense, although such argumentation is presupposed. Popper himself uses such argument, appealing to philosophical traditions, or political issues [the connections between science and an 'open society', or between hypothetico-deductive processes and 'piecemeal social engineering']. It is clear that even with him, falsification is only a part of a much more general critical project.

Albert challenges Habermas to establish his ground or foundation, but avoids any attempt to do so himself. There seems to be no obvious need for [Albert's and Popper's ] critical rationalism, and people adopt it out of a sense of commitment, apparently. But are the standards of such critical rationalism beyond criticism? Do they represent some 'comprehensive rationality'? (214). Critical rationalists actually argue by using a 'circle of reflexive self justification' (214). What this reveals is that some more general notion of rationality or argument is the horizon within which the particular standards of validity found in critical rationalism are located and determined.

Criticism is best understood, therefore, in this general sense -- as a process aimed at the progressive resolution of disagreement and the general consensus of all participants. Such a rational discussion is presupposed, and thus 'methodological questions can no longer be meaning fully separated from questions of communication' (215).

Standards and facts.

Albert says these are separated because they are tested in different ways, by correspondence to scientific laws or cultural norms in the first case, and by falsifiability in the second. This is insufficiently reflexive for Habermas: there are important differences, but the positivist approach is not the right one to grasp them.

Popper argues that when we accept a scientific proposal, a standard is implicit, even created, in it, yet when we accept a proposition we are not creating a fact -- hence the difference. But discussions do not create standards, Habermas argues, nor do they ignore facts. In real actual arguments, sometimes facts are a problem and sometimes standards are -- there is the same logical structure, and differences are only ones of emphasis. Science and ethics share the same forms of argumentation.

The apparent independence of facts is at the heart of the distinction between science and non-science for Popper, but other matters, including standards, are also involved, in the form of implicit pre-understandings [as above]. These matters are often not precisely defined, but are subject to a continuous process of self correction and explication, precisely as in attempts to understand any text. There is a dialectical relationship, between [implicit ] standards and [explicit] descriptions. Even the open and critical discussion of standards presupposes these implicit standards! (218). Even Popper's correspondence theory involves an implicit standard which requires critical justification. All critical justifications have descriptive or postulatory rules of procedure which affect the conduct of actual arguments. Further, as we have seen, 'No proposition concerning reality is capable of rational test without the explication of a connection between arguments and attitudes' (218).

Such connections cannot be deduced by logic alone, and require reflection even to make them conscious. However, once the real nature of arguments are brought to consciousness, the constraint of trying to be exclusively logical disappears. The aim of reaching agreement takes over, and dissolves any distinctions between logical and empirical argument. Popper's insistence on these separations is a way of repressing critique, and prioritising some formal logical connections between the logical and the empirical. However, even he realises that this can never be fully achieved -- hence the permanently provisional nature of scientific knowledge.

Two strategies and a discussion.

Habermas wants to defend his systematic approach and framework, which suggests the importance of a 'comparative communicative context' for science. Such a context establishes the 'transcendental preconditions of possible knowledge', but is itself affected by, and indeed 'created under empirical conditions' (220). Neither the sociology of knowledge, nor some focus on pure methodology is fully appropriate on its own, therefore. The two have to be combined in an overall critique of ideology, especially when directed at analysing 'knowledge-guiding activities' [research? theory generation? paradigm shifts?]. It is true that science is tied to the technical cognitive interest, but this is not to denounce it as completely worthless, as Albert [and Popper ] thinks.

The way forward is not to maintain the old disputes, like those between the methods of natural science versus methods of understanding. There is no reason to accept the restriction of dialectical analysis to social science and not to natural science -- this would be to 'immunise' the force of the dialectic. There is one great advantage to be gained, the ability to reflect on attitudes in research and science, and to learn from them. Thus it is interesting that science is not used to help social groups reflect on given historical situations.

Social science cannot abandon this task, however, even if it wants to become like natural science. It is necessary for social science to use methods which provide details of empirical regularities, but not to restrict methods to these, and abandon all other interests as 'ideology'. If ever it did abandon this task, there would be fewer opportunities for reflection, and a kind of re mythologisation would ensue instead. This would not even be very useful, since there are lots of problems in social life which are not just adaptive ['technical'] ones -- such as the maintenance of identity, the role played by communication in staving off anomie, and the twin 'dangers of reification and formlessness' (223) [a hint here of work to come on crisis tendencies as the 'lifeworld' gets colonised].

Using the dialectic does not mean abandoning the rules of formal logic -- we need these even when we do self reflection (ironically, positivists argue we have to abandon them to do this separate business of self reflection). A dialectical approach does open up reflection on suppressed questions.

The strategy of the participants in this Positivist Dispute is really the issue (224). Albert says that Habermas is 'masking' issues of validity, but Habermas thinks he is discussing the very conditions of validity. Habermas thinks of his own strategy as a matter of outflanking positivists: 'You have to make it clear to the positivist that you have already taken up a position behind his back' (224), questioning the very presuppositions upon which positivism's attack is based. You also have to be aware of positivist tactics, apparently, such as pretending to be stupid to force people to accept positivistic language [ a form of excessive operationalism?], or relying on the great success of natural science (really a matter of solving problems within a subculture, says Habermas). Any flaws in the intelligibility of Habermas's argument must not be traced to failures of the dialectic. Mutual unintelligibility arises because the disputants belong to different traditions. They also vary in their knowledge of different traditions -- Popper is sound, but Albert's knowledge of Hegel is clearly 'second-hand' (225). [A nice final demonstration of the tactics and manoeuvres in real arguments between philosophers!]

more guides to social theory