The comparative evaluation of scientific theories is one of the central tasks of working theoretical scientists. In the course of the 20th century, it became an increasingly important problem for philosophers of science. What is at stake is nothing less than the nature of scientific progress, and the problems of relativism and scientific realism.
The 20th century saw the development of a number of new scientific theories with unexpected characteristics, which broke ontologically and methodologically with earlier theories. This was most notably the case in physics with the theories of relativity and, especially, quantum physics. The result was a considerable enrichment of the range of theories and inter-theoretic relations available for examination.
In part as a consequence of the perceived methodological and ontological breaks between theories, a number of relativistic theses have arisen within the philosophy of science. On the one hand, the variation in the content of scientific theories in the history of science leads to the denial of scientific realism, and even of the progress of scientific knowledge. On the other hand, given the variation of the norms of scientific method, it has been denied that there may be rational grounds for the judgement that one theory is objectively better than another. A number of different approaches have been proposed in response to these radically sceptical theses: forms of naturalism, revised conceptions of transcendentalism, new experimentalism… These approaches lead to non-foundationalist and non-realist - or realist in a shifted sense - accounts of scientific progress. The debate about these issues continues to be very lively to this day.
The aim of the colloquium will be, in view of these attempts: (a) to list and analyse the specific difficulties that may arise concerning the task of comparing scientific theories, and (b) to discuss the stakes of these difficulties for the question of scientific progress, rationality and realism.
The focus will be essentially on the case of physics. Detailed studies of the history of science will be undertaken in conjunction with philosophical and conceptual analyses.
Examples of fundamental problems to be dealt with will include the following:
1. What kinds of possible "common measure" between rival theories? (kinds
of commensurability/incommensurability)
It seems desirable, to avoid the confusions often surrounding debates related to incommensurability, to elaborate and refer the discussions to a taxonomy of possible similarity/dissimilarities between physical theories to be compared.
In the usual paradigmatic cases of incommensurable rival theories, are we justified to conclude that two physical rival incommensurable theories, or fragments of them, are "translatable", in a broad sense of that term, from one into the other? Can any invariants be shown to obtain, or correspondences established, between the claims of these theories? If so, what is the nature of the relations in question? Equivalence of meaning? Identity of reference? Formal (mathematical) approximation, or conceptual approximation in some sense to be explained? Anchorage in (at least some) 'same phenomena', some same procedures of measurement, some same specifiable actions on material things…?
2. Framing the problems in terms of variation of meaning and reference?
(semantic incommensurability?)
Many authors try to characterize theoretical and observational variations of content arising between two physical theories to be compared, in terms of variations of meaning and reference of correlated linguistic items involved in the two theories under scrutiny.
In such a framework, arise questions of the following sort: which kind of linguistic theory is the more adequate to capture the transformations at stake? (contextual descriptivism/causal theories of reference; local/global holism.…); how to cope with the underdeterminations of judgments of sameness/differences between meanings and references? What about the nature and degree of dependence of reference on meaning?
And what are the consequences for the problem of an ontological trans-paradigmatic continuity? What are the implications for the traditional theory of 'reduction' and the associated idea of a progress by means of successive approximations? (May we allow conceptual incommensurability arising on grounds of formal reduction?) What are the impact on the problem of the common empirical measure between rival physical theories? (May we conclude to commensurability of observational languages and statements, despite of incommensurability of theoretical languages and statements?)
3. Variations/stability of the norms of scientific research (methodological
incommensurability?)
To claim that the norms of scientific research vary, has appeared as a threat to rationality. If the aims of science, together with the methods to reach these ends, change with time and place, relativism seems unescapable: as we need norms to justify norms, we are then unable to find an ultimate methodological ground to ensure our comparative judgments.
Many strategies have been developed to avoid this skeptical conclusion, the most spreading one certainly being the naturalist one. An evaluative overview of the very disparate existing attempts is very desirable. Questions related to this level, which involved both descriptive and normative dimensions, are of the following sort: actual changes/conservations of specific norms of research in the history of physics; relations between changes at the theoretical level and changes at the normative level; meta-principles which lead or should lead to variations of scientific norms (models governing the evolution of norms); implications for scientific rationality and progress (genuine good reasons for adopting new norms/propaganda)…
4. Comparing the predictive power of competing scientific theories?
A number of philosophers, who reject the idea of scientific progress as increasingly accurate representation of reality as untenable, or who feel the need to secure scientific progress despite the threat of these realistic conceptions, have attempted to show that relativism does not arise, since there may be weaker forms of progress: they claim that science provides increasingly powerful instruments of prediction.
In this context, the problem becomes one of determining the nature and strength of scientists' arguments for predictive superiority. It is of course linked to the questions of scientific norms and methodology. Nobody supposes that the arguments might attain the strength of rigorous mathematical proof. However, progress remains to be made with respect to the disagreement between those who take there to be a rational process of scientific debate resolution and those who take such debate to reduce to displays of force and techniques of propaganda.
5. Dissolving some traditional problems linked with theory comparison,
by focussing on the experimental level?
"New experimentalists" claim that many philosophical problems, like semantic incommensurability and it's alleged undesirable consequences on progress and realism, arise from the neglect, by philosophers, of the level of practical operations characteristic of laboratory sciences. Large part of the trouble is supposed to vanish if the philosopher moves from theories, semantics and ideas, to material things, concrete manipulations and actions always connected to scientific progress. Incommensurability is sometimes still invoked in these accounts, but incommensurability of a new kinds (in particular: a so-called 'literal' or 'machinic's one), supposed to be relevant instead of the previous semantic one.
In the context of the conference's focus, such works urge to revisit questions like: what is the exact function of high-level theories, and of competition between them, in scientific research? Are they only one element among many others, no more important than the others, or indeed, sometimes at least, far from less important? How may be redefined scientific progress, and eventually some new forms of realism ('realism of entities', 'pragmatical realism'…), when material things and actions involved in the practice of science are put in front of the scene to the detriment of theoretical representations? What become traditional conceptions of progress and realism confronted with ideas such as a human different instrumental mapping of the world? Is literal incommensurability, defined as no body of instruments to make common measurements (with the result that there is no common measure, in the very literal sense, between two physical theories), really disqualify semantic incommensurability and dissolved the afferent epistemological difficulties connected with it? Are we not confronted with two different types of possible situations, both of epistemological importance?
6. Reconsidering the crucial problem of under-determination (Duhem-Quine's
problem)
The Duhem-Quine's problem has been of central importance in the 20th century's philosophy of science. It stresses an under-determinacy that draws attention on the conventional aspects of science and has been, for this reason, seen as a threat to jugulate.
The thesis is unescapable on the level of principles, but many strategies have been tried to limit it on a practical level. In the last decade, has in particular been suggested, both an extension of the thesis (notably taking into account material aspects), and a urge to shift our philosophical focus, from the under-determination-in-principle, to the task of explaining the strong determination experienced by practitioners of science.
On these matters it is today important, at least, to take stoke of the situation. The conference will address these issues.