Despite the ubiquity of levels talk in contemporary science and philosophy, very little has been done to clarify the notion. Here I defend a kind of descriptive pluralism about the levels metaphor: it is applied usefully in many contexts to describe different relata, different relations, and different senses in which items might be located at a given level. Because the levels metaphor is so ubiquitous and so promiscuously applied, some vigilance is required to keep the applications distinct from one another. I have discussed only a few applications: levels of science, theory, realization, size, mereology, aggregation, and mechanism. Even in these few key examples, we have found good reason to remain vigilant. The implications of the levels metaphor in one application only occasionally transfer when the metaphor is applied differently.
I have also suggested that levels of mechanisms (or, more generally, levels of organization) are especially important to the explanatory structure of neuroscience and the special sciences generally. If one thinks of levels in this way, one can easily see why interlevel causation should seem so problematic (indeed, it is problematic), one is free to jettison Oppenheim and Putnam’s idea of monolithic levels of nature, and one can see room in the causal structure of the world for the existence and legitimacy of higher-level causes and explanations. Whether the idea of levels of mechanisms truly pays off in such useful ways remains to be seen. I merely hope to have preserved the metaphor, and its application to mechanisms, in the face of problems it inherits only through equivocation.
Acknowledgements
This essay is dedicated to William C. Wimsatt, whose sensitive and pioneering explorations of complexity, levels, and organization have shown how to tame the biopsychological thicket without sacrificing its wildness. Thanks to Anthony Dardis, Jens Harbecke, Donald Goodman-Wilson, Eric Hochstein, Eric Marcus, Lauren Olin, Gualtiero Piccinini, Anya Plutynski, Philip Robbins, Mark Povich, Felipe Romero, and Gary Williams for comments on earlier drafts.