Notes on: Bhaskar, R.  (1980) 'Scientific Explanations and Emancipation'.  Radical Philosophy 26, Autumn 1980.

Dave Harris

Emancipation can be defined as a 'transformation from an unwanted to a wanted source of determination' (16).  This notion is presaged and entailed by explanatory theory, but only effected in practice.

Explanatory schema and transcendental realism.  Explanation involves a subsumption of a problem under new concepts, rather than under a more general problem.  Empirical science necessitates an empirical referent, free from extraneous inferences, something 'enduring and transfactually active'. This necessarily involves us in ontology rather than epistemology, investigation of the domains of the real and the actual.  [if the empirical is not to be consumed by the concept, it must refer to a separate reality].  There is a difference between theoretical and practical problem solving.  Theoretical problem solving is not just a matter of induction or deduction, but is 'iteratively analytical, retrodictive', a dialectic between the cognitive resources which help us to model and then empirically check out our applications (17). We should understand the concrete as the result of many determinations, and pursue a process that is the reverse of abduction [?]. There are different types of science operating between abstractions and the concrete.

Ontology is therefore necessary as the study of the presuppositions of science.  Epistemology by contrast reduces the discussion [to method].  The world as we know it is multiplicity determined and emergent.  Social science is not the same as natural science, however: it operates through an notion of the causality of social forms with human agency as a mediator.  Nevertheless, the social world is still emergent, and this is why we cannot reduce it simply to human action.  We need instead a 'critical naturalism' [an Aristotelian notion—see below].  This is not positivism, because it preserves the differences between the natural and the social sciences.  Nor is it naively 'objectivist', because it sees reality as the result of historical practices and processes.  There is a break with approaches operating with the old structure/action dichotomy, or the syntheses between them [which are also epistemological].  The approach develops a realist stance, which makes it emancipatory in the same sense as natural science can be, as in the definition above.

Social structure, human agency.  Critical naturalism posits an ontological structure of praxis as in Aristotle, involving the transformation by agency of preexisting material causes, both natural and social.  Social causes exist and persist only through agency, so we can see the social as both a condition and outcome of agency.  This leads to the notion of a duality of structure [as in Giddens?  Note also that DeLanda sees a lot of parallels with the position of Deleuze].  There is also a duality of praxis in that agents produce unmotivated conditions for their own conduct (18)—this is not a natural process, but it is a necessary aspect of developing any social science.  Explanations and social science are also never closed.  Instead, they follow explanatory non-predictive criteria of confirmation and falsification [I'm pretty sure this is a quote, although I have not kept the page number] 

Splitting social life between structure and action is too reductive.  At the level of the actor, unintended consequences and unmotivated conditions are found, and unconscious motivations.  Analyzing these provides an emancipatory role for social science.  Action is generated by the conative [involving will or desire] as well as the cognitive, and we need to reject theoreticism.  The limits of social science do not arise from epistemological errors, but from ontological ones—it is common not to provide an account of the full reality of action.  The conditions for action include Giddens's rules and resources which are unmotivated and unacknowledged, hence the limits of social phenomenology.  Historical resources are also important, and historical constraints  can be superceded, hence the limits of approaches like ethnomethodology which have no historical dimension.

Interpretative fundamentalism.  There are no certainties, no non-historical Archimedian points from which to observe social life.  Referring to 'facts' is no good, since these are already social, they are objects apprehended in experience.  Science must operate as a network without foundations (19), something which is historical, seeing experience always in a context.

Language and beliefs are now central and constitutive of human reality [which is where DeLanda parts company], so can we take beliefs as fundamental?  This would characterise hermeneutic stances, and their well-known uncritical sides.  Fundamentalism must always presuppose some unvalidated knowledge, and cannot establish its own legitimacy (19).  In hermeneutics, interpretations alone are constitutive, but this ignores the non-conscious aspects as well: it is impossible to express ideas and institutions by means which are customarily used to describe them [almost certainly another quote, 20].  Language cannot describe its own constitutive power.  And this means there are necessary fundamentals in discourses—how should we interpret them?

We can attempt to dismiss or replace this problem, or simply assent to actual first person accounts [which is ethnomethodology's approach].  We get adequate interpretations only by referring to agents and their self formation.  In this sense, adequacy refers to 'saving a maximum of significant phenomena and...  showing...  the degree and type of "irrationality"' (20).  [Hard to see how this would actually work].

Facts and values.  The human sciences operate with notions of causal motivation, and also logically entail value judgements.  This raises implications for their role as prescriptions for action.  We should think of facts and values as related in a 'helix' [values lead us to select particular facts, which leads to particular theories, which leads to a second set of values, 20].  We need to avoid reification in positivism or theoreticism, recognizing that theories are also motivated, conative and cognitive.  An excessive emphasis on practice also provides for a 'practicalist' error, which ignores the cognitive, and is therefore 'irrationalist' (21) [with a hint also of decisionism].  Theoreticism denies practice any role in the generation of theory. The relations between facts and values are not symmetric, however.  Facts and theories logically entail values and practical judgments, but not the converse.  This is why we conceive of their relations as a helix and not a circle, to permit progress.

The subject matter of the human sciences includes social objects and beliefs about these objects (21).  Emphasizing either objects or beliefs can produce empiricism or idealism respectively.  Again, their relations are really causal and cognitive—they generate actions and they can be critiqued.  But the causal or ontological is prior to critique.  So, if a belief P is false, and S explains P, then S can be critiqued, and we can generate a policy intended to remove it.  Indeed, the critique of policy is 'mandatory' (21), unless S is dissoluble, or is not maintained by some false consciousness [in cases where we know what oppresses us but are powerless to remove it?].  This would provide us with an example where a fact leads to a value.  This procedure also tries to explain the source of false beliefs, as in 'explanatory criticism'.  This refers back again to the notion of emancipation as a matter of shifting determinants rather than pursuing, saying, isolated moral critiques of one determinant.  Overall, we can develop a notion of levels of critique as 'so many ratchets of reason'.

Instrumental vs. critical rationality.  Human sciences can still be emancipatory even with mild levels of critique.  Thus, at level one, they can provide us with techniques to increase the efficiency of means to an end.  At level two, they can operate contextually, to situate instrumental rationality, so that knowledge is seen as a technique which breaks the hold of dominant groups and can be used to express the wants of subordinate groups.  At the third level, human science can be intra-discursive, but still not explanatory—it can critique other theories including the consequences for action.  In these first three senses, all sciences are critical and evaluative. There is a fourth level though, involving explanatory critical rationality.  Natural science sometimes tries to explain false beliefs as part of its activity, but humans sciences must do so.  The problem that emerges is how to move from diagnosis to political action.  There is clearly at least freedom from [as a first step?], involving achieving a less constrained position.

All these provide a role for the social sciences in diagnosing cognitive, practical and communicative ills (25) [shades of Habermas here?  Is the argument that social science needs different forms for these different contexts, or is there some underlying model of the critique of beliefs?]

The explanatory critique of reality involves a kind of depth rationality.  We find applications of this model in matters like psychological rationalization or ideological mystification.  In these examples, misrepresentation is necessary for P to exist, even for the object of belief to exist or persist [for example psychological states].  In social life, we can explain the persistence of an object and a false presentation of it because the object itself necessarily generates false appearances [commodities looking like things?].  This is the characteristic mode of production of capitalism, yet there is probably no single category of contradiction, rather specific concepts of contradiction instead.  Thus Marx's Capital criticizes theories, practical consciousness, and the conditions explaining consciousness.  The split between science and ideology also sees consciousness as inadequate, asocial, something reified, hypostasized or fetishized.  This is rooted in alienation (25). Even though Engels saw a split between facts and values, Marxism generates a necessarily evaluative stance from its cognitive critique, requiring no external moralities or ideals.

Marx and Freud can be seen as examples of deployments of depth of rationality, but is depth critique transcendentally necessary in any sense?  Critique can be rejected on  rational grounds, and this would lead to further depth investigations of the reasons for such a block.  This would need to be cooperative, and involve a necessary link underlying the theory and various practical unblockings that might be possible [pretty much what Habermas says about Freud?].  Critique in this case could take different forms according to whether the block was cognitive or communicative.

The desire for emancipation can not be predicated on the basis of the rationality of agents or of historical development.  Nor is it a priori.  It needs to be investigated. This can encounter dissonance rather than liberation, where people might develop a cognitive awareness of the block but still desire to keep it.  Critique is really grounded in the belief at the beginning, that sciences must be critical.  A 'paramorphic'relation exists between the natural and the social sciences, in that social science investigates the underlying structures producing events just as does natural science.

In conclusion, for science to be emancipatory, it must take reasons as causes, if discourses not to be 'ontologically redundant' and 'scientifically inexplicable'(27).  Its emancipatory values are immanent, as in Habermas's argument about speech-constitutive universals.  However, there are problems with Habermas because in his terminology, any cooperative speech is likely to be seen as liberating [making connections with actual speeches problematic].  Ethics need no [foundational] linguistic formulation.  Critique should be internal to or conditioned by objects, making self-reflexive auto-critique the only possible option.  The emergence of knowable laws will prevent reductions to either materialism or idealism.  If materialism is deterministic, emancipation is impossible without some breakthrough, and if it is idealist, emancipation will be either entirely intrinsic to thought, or some extraneous oddity.  The notion of emergence is essential, and only then can policy depend on a realist science (28), although the development of emancipatory policy is not the only reason why the concept of emergence is true.

more social theory