pluralismA
Scientific Pluralism Between Realism and Social Constructivism?
Some Critical Comments
Peter P. Kirschenmann
Prof. Emeritus, Faculties of Philosophy and Exact Sciences
Vrije Universiteit, Amsterdam
Borssenburg 6, NL-1181 NV Amstelveen
31-20-6453336,
[email protected]
Abstract: Scientific pluralists oppose their views to scientific realism, especially in its monistic form. I sketch the problem context that has given rise to pluralisms. I critically discuss these views in terms of a number of questions: Is the monism they oppose not a straw-man? Are pluralists onlookers or philosophers? Are the claims of strong pluralism tenable? Do pluralists not have an unclear relation to truth? I conclude that scientific realists can accept the main point of pluralists, but need not and will not accept that there are ineliminable incompatibilities between scientific approaches to particular phenomena.
1. Introduction
There have been competing theories, like the phlogiston theory and the oxygen theory of combustion. The scientific realist is happy that one of the theories, the oxygen theory, won out and proved to be the correct one. There have also been non-competing theories concerning the same phenomena, raising problems about their relation. For instance, kinetic theory or statistical mechanics has created still unresolved problems in relation to phenomenological thermodynamics, the reversibility and recurrence problem. Further, there are theories developed for different phenomena which are patently hard to integrate in one theory, such as quantum theories and general relativity theory. Theories and models in the physical sciences also are replete with idealizations and approximations. Notice, however, that many of them, like the use of mass points for pendulum or planetary motion, can be proven as exactly valid.
Especially in the non-physical sciences, in psychology and the social sciences, there have been different, often competing, approaches and models regarding the same phenomena. For instance, there have been well-known nature versus nurture debates about the explanation of properties and traits of human beings. And it is especially in the face of the situations in such areas of research that philosophers have defended an explanatory or scientific pluralism. Helen Longino (2006, 2013) has done so in her extended studies of scientific approaches to two kinds of behavior, sexuality and aggressiveness.
This type of pluralism has been characterized as a position between "two extreme views", namely between "a (monistic) metaphysical realism according to which there is in principle one true and complete theory of everything … [and] … a constructivist relativism according to which scientific claims about any reality beyond that of ordinary experience are merely social conventions" (Giere 2006, 26).
This characterization can be put in terms of constraints. For realists in general, the abstract overall constraint for acceptable scientific hypotheses is that they be true. "Constructivists admit an indefinite number of theories, the only constraint being human ingenuity." (Kellert et al. 2006a, xiii). Pluralists hold that there are constraints, cognitive and pragmatic, "that limit the variety of acceptable classificatory or explanatory schemes". (ibid.)
I shall critically discuss pluralist views in terms of a number of questions: Is the monism they oppose not a straw-man? Are pluralists onlookers or philosophers? Are the claims of strong pluralism tenable? Do pluralists not have an unclear relation to truth? I conclude that scientific realists can accept the main point of pluralists, but need not and will not accept that there are ineliminable incompatibilities between scientific approaches to particular phenomena.
2. The Monist Straw-Man
One should note that, actually, pluralists oppose their view not so much to realism in general, but rather to scientific or explanatory monism, the view that the world or any investigated part of it can be described or explained by a single, complete and comprehensive account. For, clearly, there have been realists (me included) who are not monists. For instance, John Dupré (1996, 105) defended a "promiscuous realism", by arguing that "individual things are objectively members of many individual kinds", which all are real, but not reducible to one essential kind. Pluralists tend to reply that such a view "is hard to distinguish from radical relativism" (Kellert et al. 2006a, xiii).
One cannot help getting the impression that the monism pluralists are opposed to is a kind of straw-man. Of course, in the first half or the last century, there was the unified science movement. But this movement advanced the idea of a unity of scientific methods or, even more generally, promoted the scientific attitude as an antidote against metaphysics and ideologies. In its anti-metaphysical attitude, this movement was even similar to the pluralist movement.
Also of course, there is the recurring idea of a "theory of everything". But up to now, the idea has remained very much a scientific phantasy.
Longino (2013, 138ff.) argues by way of examples that researchers of human behavior who "engage in extended defenses" of their own approach or "expose the weaknesses of alternatives" actually "adopt a monist perspective". This is dressing up the straw-man, for I doubt whether any of those researchers would claim that her/his approach gives the final, complete and comprehensive account of the behavior concerned.
3. Pluralist Philosophy or Spectatorship
The monist view that "the ultimate aim of a science is to establish a single, complete, and comprehensive account of the natural world" or the investigated parts of it is considered a metaphysical view since it is said to be based on the metaphysical assumption that " the world is such that it can, at least in principle, be completely described or explained by such an account" (Kellert et al. 2006a, x). Logically, it would follow that the pluralist denial of this metaphysical view is itself also a metaphysical position.
Yet, pluralists do not want to defend a metaphysical position. They "do not assume that the natural
world cannot, in principle, be completely explained by a single tidy account", but rather "believe that whether it can be so explained is an open, empirical question" (ibid.). This is why they say that their "general thesis is epistemological" (ibid.) Here, 'empirical' seems to refer to what can "empirically" be observed in the development of the sciences. In this respect, then, the pluralist stance seems to degenerate into the position of a mere spectator, just watching and witnessing, possibly examining, what is happening in the sciences.
Giere places his form of pluralism, his "perspectival pluralism", in the framework of a scientific naturalism. He recommends to reformulate metaphysical doctrines, like monism, as methodological maxims and to consider naturalism also as a methodological stance. And as a methodological naturalist, "one can wait until success is achieved. And there are good naturalist standards for when this happens." (Giere 2006, 39) Thus, he also seems to advocate a wait-and-see attitude.
However, pluralists are not mere spectators of developments in the sciences, but philosophers at least inasmuch as they put forward arguments. For instance, they identify what they see as the various general sources of the plurality of approaches, theories, and models concerning the phenomena in particular areas of scientific inquiry. "These include (a) the complexity of the phenomena—whether associated with crossing levels of organization or multiple factors within the same level of organization; (b) the variety of explanatory interests; (c) the openness of constraints—whether from above or below; and (d) the limitations of particular explanatory strategies vis-à-vis the phenomena." (Kellert et al. 2006a, xvi-xvii)
Giere (2006, 38) argues that a scientific claim could only be "exactly true is if it uses a complete model that fits the world exactly in every respect." The underlying assumption is "that everything is causally connected with everything else" (ibid.) Thus, in the case of an incomplete theory, "there will be some influences on the subject matter not accounted for" by it (ibid.). The argument seems to invoke "a metaphysical assumption of connectedness in the universe" (ibid.) Yet, Giere restates that it "can be made less metaphysical by assuming only that we do not know the extent" of this connectedness (ibid.). He thinks that this "more modest conclusion is sufficient to support a robust (perspectival) pluralism" (ibid.).
In the next section, I shall briefly consider Longino´s particular arguments for pluralism. She, too, wants to avoid metaphysical assumptions. She states (2013, 138): "Pluralism is best understood as an attitude to adopt with respect to the multiplicity of approaches in contemporary sciences." This pragmatic pluralistic attitude has been labeled "the pluralist stance" (Kellert et al. 2006a)
4. Strong Pluralism
One can distinguish between weak, modest/moderate, strong/substantial, and even radical pluralism (Longino 2013, 137ff.). Weak pluralists regard a plurality of approaches as temporary. Moderate pluralists hold "either that a plurality of questions supports different and nonreducible, but still compatible approaches or that pluralism at the theoretical level resolves into an integrated account at the phenomenological level." (ibid., 137). Both weak and moderate pluralism are said to reduce to monism (Kellert et al. 2006a, xii). Strong pluralists, like Longino, hold that some areas of investigation are characterized by multiple "incompatible", "nonreconcilable" approaches, resulting in a "ineliminable"plurality of "incommensurable", "incongruent" theories or models, each representing only partial knowledge (Longino 2013). Finally, radical pluralists would claim that this holds for all areas of scientific investigation.
Longino (2013, Chs. 2-6) presents and discusses five particular types of approaches to understanding human behavior: quantitative behavioral genetics, social-environmental approaches, molecular behavioral genetics, neurobiological approaches, and several integrative approaches (developmental systems theory, the gene-environment-neurosystem interaction program, multifactorial path analysis); she subsequently (in Ch. 7) adds human-ecological approaches, which do not draw on the biology of individuals, but on the comparative study of populations.
One main argument of hers is that the different approaches parse the space of possible causes of behavior in noncongruent manners. Longino (2013, 125ff.) characterizes this space in terms of the following categories of causes: (a) genotype: allele pairs; (b) genotype: whole genome; (c) intrauterine environment; (d) physiological and anatomical factors; (e) nonshared environment; (f) shared (intrafamily) environment; (g) socioeconomic status. For instance, molecular behavioral genetics takes into account or "measures" only (a) and (b), treating the remaining factors as environmental. Social-environmental approaches, on the other hand, consider (e), (f) and (g) as effective causes of variations, and the remaining factors as uniform. Thus, they consider (c) also as inactive, while quantitative behavioral genetics counts (c) as environmental or as noise. For such reasons she claims that those approaches are incompatible and cannot be integrated into a single complete account.
Another point put forward in support of this claim is that behavior, the object of investigation, is understood in different ways: "as a disposition or as an episode, as a dimension of variation in a population among populations, or as an individual characteristic" (Longino 2013, 102). In addition, concepts of specific behaviors, like aggression and sexuality, vary historically and culturally. (It is strange that Longino takes up the questions of definition of behavior after her discussion of the approaches to behavior.)
Longino is not clear about what integration into one complete account could mean; that is why it is not clear what she is rejecting (Driscoll 2014). The various approaches to study human behavior undoubtedly are in the process of development and incomplete. But the conclusion that integration is ruled out is not justified. Note that Longino nowhere calls the different approaches or their results 'contradictory'. Indeed, to the more moderate pluralist Giere (2006, 37), the approaches or perspectives concerned "seem to be largely complementary and nonoverlapping".
5.Uncomfortable Truth
As mentioned, scientific realists will insist that theories, models, explanation should be checked and assessed for their truth. We saw Giere (2006, 38) arguing that a scientific claim could only be "exactly true is if it uses a complete model that fits the world exactly in every respect." This is an exaggeration or else a confusion of '(exactly) true' with 'complete', or also a confusion of realism with monism. For a realist, a model can be (exactly) true in particular respects.
For Longino (2013, 147ff.), the attitude of non-eliminative pluralism focusses on the "different kinds of knowledge" incommensurable approaches can offer, without "assuming that one, at most, is correct". Yet, what "can it mean to say that two or more [such] approaches are correct"? To answer this question, Longino has introduced the "umbrella concept" of conformation of a successful representation to its object, which is meant to cover truth at one extreme, but also homomorphism, approximation, etc. These other forms come in degrees and in terms of respects. To say, then, that two different representations "are equally correct is to say that for each there is a degree and respect in relation to which it can be said to conform to its subject matter".
No doubt that approximations etc. come in degrees and respects. Yet, truth is an unavoidable general concept. It is implicit in Longino's notion of correctness. And it cannot but figure in answers to questions such as: does this representation conform to this degree and in that respect to its object?
According to Longino (2013, 149), different kinds of conformation are mandated by different pragmatic aims. Thus, one cannot separate something like pure comprehensive knowledge from its application; rather, practical goals and associated constraints shape the choice of an approach and the evaluation of its results. In this sense, she advocates a "pragmatically inflected pluralism". Yet, one can side with the point she makes and still insist that there will always remain a purely epistemic evaluation of the truth of a result.
6. Conclusion
Pluralists have a good point when they insist that understanding the sciences includes understanding the plurality of scientific approaches, models etc. However, they do not need to set up monism as their opponent to make this point. Scientific realists not only easily accept this point, but have also been dealing with it extensively. They will of course take issue with strong and radical pluralists, who assert the ineliminability of certain incompatible models. Incompatibilities in the sciences, for realists, are strong invitations for further research.
For emergentist realists, in particular, there are autonomous causal and thus explanatory relations on higher levels, although not inconsistent with explanatory mechanisms on lower layers. For instance, "the mouse is dead because a snake bit it" is a sufficient explanation by itself, aside from possible further biological and molecular accounts. Emergence, then, leads to a natural plurality of accounts. It seems to me that pluralists tend to neglect this autonomy of accounts on different levels.
Literature
Driscoll, Catherine, "Review of Studying Human Behavior", Philosophy of Science 81, 676-680.
Dupré, John 1996 "Metaphysical Disorder and Scientific Disunity", in Peter Galison, David J. Stump (eds.) The Disunity of Science: Boundaries, Contexts, and Power, Stanford, Calif.: Stanford University Press 1996, 101-117.
Giere, Ronald N. 2006 "Perspectival Pluralism", in Kellert et al. (eds.) 2006, 26-41.
Kellert, Stephen H., Longino, Helen. E., and Waters, C. Kenneth (eds.) 2006 Scientific Pluralism, Minneapolis, London: University of Minnesota Press.
Kellert, Stephen H., Longino, Helen. E., and Waters, C. Kenneth 2006a "Introduction: The Pluralist Stance", in Kellert et al. (eds.) 2006, vii-xxix.
Longino, Helen. E. 2006 "Theoretical Pluralism and the Scientific Study of Behavior", in Kellert et al. (eds.) 2006, 102-131.
Longino, Helen E. 2013 Studying Human Behavior: How Scientists Investigate Aggression & Sexuality, Chicago and London: The University of Chicago Press.
