Project

Philosophy of science; Biology; Mechanism; Economics; Scientific explanation

Lead
La constitution mécaniste dans les sciences spéciales : analyse philosophique et logique de la découverte
Lay summary
Une partie importante de la recherche scientifique vise à découvrir les mécanismes qui sont responsables d’un phénomène, par exemple, le traitement d’informations dans le cerveau ou la genèse d’un type de cancer ou bien d’une bulle immobilière. Mais qu’est-ce qu’un mécanisme ? D’où vient sa force explicative ? Et quelles règles de raisonnement permettent aux scientifiques d’identifier les composants d’un mécanisme ? Par contraste à la causalité simple, la constitution mécaniste n’a auparavant pas fait l’objet d’une logique de découverte, c’est-à-dire d’algorithmes qui peuvent inférer des mécanismes à partir de données empiriques. En étudiant des exemples biologiques, neuroscientifiques et économiques, ce projet vise à développer de telles algorithmes pour la découverte de mécanismes.

Direct link to Lay Summary

Last update: 10.01.2017

Active participation

Self-organised

Number	Title	Start	Funding scheme

The notion of constitution has always been of great interest in philosophy. For a long time, philosophers have been concerned with material constitution. Is the statue of David identical with the lump of clay co-extensional with it? Or is it something over and above the lump? While this sort of question is still much debated, in more recent times another notion of constitution has been gaining much attention, especially among philosophers of science. This is the notion of mechanistic constitution, namely the relation between a phenomenon of scientific interest (e.g., the temperature in a gas’s container, a mouse finding her nest) and those component parts and behaviors that constitute the phenomenon (e.g., position and momentum of gas particles, NMDA activation at the synapses of the mouse’s neurons). According to many philosophers, the notion of constitution is central to mechanistic explanations of phenomena of interest to disciplines such as biology, neuroscience, and the social sciences. In a mechanistic explanation, a mechanism’s parts and behaviors explain the phenomenon in virtue of constituting that phenomenon. This relation between phenomenon and mechanism is assumed to be non-causal, and so is the corresponding explanation. What remains unclear is how mechanistic constitution is to be understood, if not causally, especially in the light of the heavy criticism recently received by the most prominent account of constitution in the literature, namely Craver’s “mutual manipulability” theory. Moreover, since constitution is a necessary component of mechanistic explanation, understanding the former is necessary for understanding the latter. In particular, interpreting scientific case studies---including those indicated by philosophers as exemplary cases of mechanistic explanation---as in fact instances of scientific explanation depends on the availability of sound criteria for constitution. Indeed, we may go as far as to claim that understanding the very conditions for an arrangement of parts and behaviors to count as a mechanism depends on criteria for what count as the mechanism’s constituent parts and operations. Finally, understanding constitution is also necessary for systematizing our study of constitutional dependencies, by way of methodologies for constitutional discovery, whether observational or experimental. Methodology crucially depends on conceptual clarity. One may usefully contrast this case with the case of causation, where successful protocols exist for discovering causal relations. In particular, such protocols are made possible by definitions (whether explicit or implicit) of an operational kind, namely definitions that link (albeit not reduce) the meaning of a concept to what would count as evidence of the object signified by it. In the same way conceptual clarity on causation was pivotal in advancing empirical research, conceptual clarity on constitution, too, promises to systematize and advance a kind of empirical investigation, which is still not well understood by philosophers but important to many sciences. In a first module, the project will explore interpretations of constitution alternative to Craver’s mutual manipulability theory. Among other things, this first module will cast light on the role of interventions and multiple realizability in the analysis of constitution. In a second module, the project will use classical examples of mechanistic explanation---viz. of action potential in physiology and of spatial memory in cognitive neuroscience---to test and revise the adequacy of the analysis of constitution, as well as examples of putative mechanistic explanations---viz. of cognitive or neurological disorders by means of network models, and of the stylized facts of finance in non-classical macroeconomics---to use the analysis to decide whether or not they are concerned with establishing constitutional hypothesis, and thus with providing mechanistic explanations. Finally, a third module will operationalize the proposed interpretations into---observational and experimental---protocols for constitutional discovery, which may be used in conjunction with available causal discovery methods in an integrated methodology for mechanistic discovery, that is, for discovering structures containing both causally- and constitutionally-related variables.