An important recent development in evolutionary biology recognizes niche construction as “a second major participant in evolution, after natural selection” (Odling-Smee, Laland, Feldman 2003, 12). Niche construction is an ecological inheritance: along with genes and epigenetic resources from their parents, organisms inherit a transformed environment exerting different selection pressures via the cumulative effects of other organisms’ activities on their developmental and selective environment. Niche construction is often regarded as primarily abiotic, but behavioral niche construction occurs wherever organismic behavior affects the next generation’s developmental environment in ways that reliably reproduce that behavioral pattern. Recognizing the biological significance of niche construction thereby also blurs traditional boundaries between biological and cultural evolution.
In a forthcoming book (Rouse 2015), I argue that language and other aspects of human conceptual understanding arose and are sustained in significant part through behavioral niche construction. This paper brings together four important, interrelated consequences of this account of conceptual capacities for a broadly naturalistic understanding of scientific practice:
The traditional philosophical idea that a scientific theory “represents” the world is a metaphor, grounded in other epistemic activities that are actually representational. For example, if we make a drawing of an object that we see, it can be said straightforwardly that the drawing represents the object. The relationship between a theory and the world (or the relevant part of it) is not truly representational. To the theory, we have full direct access; to the world, we do not. In contrast, in real representation there is clear accessibility to both sides. The very idea of the “external world” is a metaphor (“external” — outside of what?), imagined after the phenomenal objects which we observe and make representations of. The rest of the metaphorical structure follows easily: the theory represents the imagined object, with a correspondence between various aspects of the theory and various imagined properties of the object. This metaphorical correspondence is the elusive truth sought by scientific realists.
In line with the program of the study of scientific practice articulated previously (Chang 2011), I propose to consider what we do when we represent something. In the simplest kind of case, we take note of some particular observed features of an object, and create another object that has those same features. Something re-presented has to be present (or be presented) to us in the first place. Representing is the construction of an artificial object (which may be a formal system) that serves to express specific features of an observed natural object, in order to facilitate the achievement of certain epistemic aims. When a theory “represents” the unobservable world, we do typically begin with some observed features of the situation but the theoretician introduces many other features. In order to check the faithfulness of the “representation”, we would need to have independent access to the alleged features of the world, which we do not. So, rather than representation, what we have here is the activity of construction. These points will be illustrated through the case of the Rutherford–Bohr atomic model.
The external world, representation, correspondence — these concepts, as they normally occur in epistemological discussions, are metaphors. Moreover, they are dead metaphors, in the sense that they are by now so ingrained in the discourse that they are not even recognized as metaphors and routinely get mistaken as literal expressions (see Bowdle and Gentner 2005; Goldberg 2011, ch. 4). The problem with dead metaphors is that they no longer serve the creative and exploratory functions of metaphors (on which see Hesse 1966), which require an awareness that the application of the expression in question is not literal, or at least uncertain if taken literally. Dead metaphors are at best useless and harmless, at worst misleading. I will finish with a discussion of how we might move beyond the dead metaphors of representation and correspondence. One option is to remove them, in the hopes that we may see more clearly what the non-metaphorical situation is. Or would there be benefits in keeping them but rendering them actively metaphorical?
The shift in philosophy of science from a model of science as a body of propositions to a model of science as systems of practices allowed for novel perspectives on many old philosophical queries about the scientific enterprise. One such philosophical query was the creation of novel scientific concepts. In a philosophy of science that focuses on the propositional structure of science, a fundamental distinction is maintained between the activities that bring about a scientific concept and the theoretical role of a scientific concept—and the former are taken to be irrelevant for a philosophical understanding of a given concept. Such a view is shored up by a treatment of concepts as (in principle) fully graspable entities, in at least some invariant core, which provided the basis for treating the conceptual realm as its own independent object of philosophical analysis, separate from practice.
The acceptance of scientific practices as a subject worthy of analysis in philosophy of science completely transformed how the problem of novel concept formation in science is dealt with. Firstly, a philosophy of scientific practice turns the question of how a scientific concept is formed from a non-starter into something philosophically fruitful. Secondly, whatever answers there might be are likely to be found in the various practices scientists are involved in—from practices of experimenting and observing to practices of visual and/or mathematical modelling, and so on. On these lines, Nersessian (2008) and Rouse (2011) have argued that instances of scientific concept formation are not purely intra-linguistic and sudden events, but that making conceptual sense of scientific experiential situations is a tortuous process that is lengthy, difficult and which appeals to diverse methodological strategies in order to articulate a communicable and well-supported scientific concept.
Thus, on the practice reading, scientific concepts are taken to be context sensitive insofar as (1) the practices are the loci where new scientific concepts are formed; (2) scientific practices are integral to the concept formation process; (3) the loci to which a scientific concept is projectable beyond its context of formation are extensions and continuations of precisely those scientific practices that afforded the concept to be formed in the first place. In this paper, I focus on (2). I first show that the formulation of a concept of magnetic field was the result of experimental and representational practices that treated magnetism as a spatial array of dispositional properties—practices of mapping phenomena in controlled settings. That scientific practices are integral to concept formation is well established in the literature (e.g. Chang 2011; Rouse 2011). I note however that such accounts show how scientific practices influence concept formation, but not that they do. I argue that the latter is necessary if one wants to avoid the charge that scientific practices are already presupposed as integral in reconstructions of historical cases rather than accounted for, and I conclude with an attempt to offer such an argument.
Consider the following sketch of 20th and now 21st century philosophy of science. As Reisch (2005) argues, the original impulse behind the Weiner Kreis was social as well as epistemological in nature. Nonetheless, by the post-war period mainline philosophy of science had become strongly internalist in orientation. One effect of the failure of mid-century philosophy of science to take the larger cultural effects of technoscience seriously was the creation of science and technology studies in the 1960s.
At the same time, Kuhn’s Structure (1962) initiated the long slow march of the philosophy of science away from an internalist focus and toward taking history and culture seriously. The founding of SPSP can thus be seen as the next logical step in this process—a response to the deficiencies of mainline 20th century philosophy of science by emphasizing questions attendant to the actual practice of science in the real world.
But does SPSP actually practice its practice? Who is the audience for its insights—philosophers, or the wider world? Who comes to its meetings, or submits papers for consideration? Has SPSP managed to break out of the charmed circle of what I have called disciplinary philosophy (Frodeman 2014)?
This talk raises these issues by asking, what are the consequences of putting the philosophy of science into practice? This question can be broken down into two elements, what I will call the institutional and the theoretical. While loathe to separate the two—this separation, I will argue, is much to blame for the irrelevance of the philosophy of science to the larger world—I will focus my remarks on the latter, and ask: what are the theoretical consequences of actually practicing the philosophy of science?
I will argue that the first consequence is that philosophic rigor must itself be seen as pluralistic in nature. The rigor of disciplinary work (i.e., work directed toward other philosophers) is different from, but neither better nor worse, than the philosophic rigor appropriate for real world exigencies. This is a point that bioethicists have long understood. In his 1973 article “Bioethics as a Discipline,” Daniel Callahan already saw that doing philosophical thinking with physicians, scientists, and other stakeholders demands “rigor…of a different sort than that normally required for the traditional philosophical or scientific disciplines.” Bioethics today (de facto, if not de jure) exists in disciplinary and in non-disciplinary forms that synergize with one another.
This suggests that we should not be forced—as a matter of general principle, and as a matter of gaining tenure and promotion—to value one standard of rigor over another. In response, over the last decade I have offered the neologism of ‘field philosopher’ to describe what an alternative sense of philosophic rigor might look like. Field philosophy is addressed primarily to non-disciplinary peers in evolving contexts of use. And its disciplinary activities are oriented toward sharing lessons learned in order to improve non-disciplinary contributions. In addition to disciplinary criteria of success, field philosophers are judged by their contributions to policy processes and public debates. And rigor is defined by balancing epistemological thoroughness with other criteria such as timeliness, cost, and relevance.