This paper considers the implications of recent accounts of emergent phenomena for the question of the unity of the sciences. I first offer a historical account of physicalism in its different guises since the mid 19th century. Two threads connecting these otherwise quite different views have been the rejection of emergent phenomena and the commitment to the unity of science. In section two I provide an exposition of emergence as presented in recent philosophy of science, where the key claim is that “parts behave differently in wholes”, based on the empirical finding of what Gillett (2016) calls “differential powers.” Gillett argues that the empirical evidence does not yet support the strong emergentist claim that there is downward causation or any other form of influence from the whole system to its constituent parts, but that such evidence might be obtained. In section 3 I propose instead that the question of whether or not the finding of differential powers is taken to provide overwhelming evidence for strong emergence depends on the further interpretation of differential powers, and ultimately on very broad metaphysical commitments. The interpretation of differential powers that is most resistant to objections from opponents of strong emergence involves a rejection of substance ontology, and hence the rejection of physicalism. Thus, as I conclude in section 4, philosophers should not wait in expectation for empirical results that will settle the question of whether or not there is strong emergence.  I offer a preliminary costs/benefits analysis of the different ontologies of differential powers, intended to aid the reader in their decision over the status of strong emergence. On the most radical interpretation, the usual physicalist conception of the unity of science must be rejected, while a different kind of metaphysical wholism stands in its place.

In this talk, I explore what might follow for the metaphysics of fundamentality if we take seriously certain reasons to believe there is anything fundamental in the first place.

There is a line of thought gathering momentum which suggests that just like causal laws govern causation, there needs to be something in metaphysics that governs metaphysical relations. Such laws of metaphysics would be counterfactual-supporting general principles that are responsible for the explanatory force of metaphysical explanations. There are various suggestions about how such principles could be understood. They could be based on what Kelly Trogdon calls grounding-mechanical explanations, where the role that grounding mechanisms play in certain metaphysical explanations mirrors the role that causal mechanisms play in certain scientific explanations. Another approach, by Jonathan Schaffer, claims to be neutral regarding grounding or essences (although he does commit to the idea that metaphysical explanation is ‘backed’ by grounding relations). In this paper I will assess these suggestions and argue that for those willing to invoke essences, there is a more promising route available: the unificatory role of metaphysical explanation may be accounted for in terms of natural kind essences.

 Physicalism is the thesis that all mental facts are ultimately grounded by physical facts. There is an explanatory gap between the mental and physical, and many see this as posing a challenge to physicalism. Jonathan Schaffer (2017) disagrees, arguing that standard grounding connections involve explanatory gaps as a matter of course. I begin by arguing that Schaffer and others mischaracterize the explanatory gap between the mental and physical—it chiefly concerns what I call cognitive significance rather than priori implication or related notions. The upshot is that standard grounding connections normally don’t involve explanatory gaps. Then I consider two grounding-theoretic proposals about how to close explanatory gaps in the relevant sense, one involving structural equations (Schaffer 2017) and the other mechanisms (Trogdon 2018). While each of these proposals seeks to illuminate grounding connections, I argue that neither is helpful in closing the explanatory gap between the mental and physical.  

The relationship between a mechanisms and its working parts is known as mechanistic constitution.   In this paper we review the history of the mechanistic constitution debate, starting with Salmon’s original account, and we  explain what we take to be the proper lessons to be drawn from the extensive literature surrounding Craver’s mutual manipulability account.  Based on our analysis, we argue that much of the difficulty in understanding the mechanistic constitution relation arises from a failure to recognize two different forms of mechanistic constitution — corresponding to two different kinds of relationships between a mechanism and the phenomenon for which it is  responsible.  First, when mechanisms produce phenomena, the mechanism’s parts are diachronic stages of the process by which entities act to produce the phenomenon.  Second, when mechanisms underlie some phenomenon, the phenomenon is a activity of a whole system, and the mechanism’s parts are those of the working entities that synchronically give rise to the phenomenon.  Attending to these different kinds of constitutive  relations will clarify the circumstances under which mechanistic phenomena can be said to occur at different levels.

 Levels terminology is employed throughout scientific discourse, and is crucial to the formulation of various debates in the philosophy of science. In this talk, I argue that all levels are, to some degree, autonomous. Building on this, I claim that higher levels may be understood as both emergent from and reducible to lower levels. I cash out this account of levels with a case study. Nerve signals are on a higher level than the individual ionic motions across the neuronal membrane; this is (at least in part) because the nerve signals are autonomous from such motions. In order to understand the instantiation of these levels we ought to identify the mechanisms at the lower level which give rise to such autonomy. In this case we can do so: the gated ion channels and pumps underwrite the autonomy of the higher level.

The most familiar approach to scientific compositional explanations is that adopted by the so-called “New Mechanists”. This approach focuses on compositional explanations of processes of wholes in terms of processes of their parts. In addition, the approach focuses on the use of so-called “interlevel interventions” as the means by which compositional relations are investigated. By contrast, on the approach I adopt, we see that there are compositional explanations of individuals in terms of their parts and properties of individuals in terms of the properties of their parts. In addition, I draw attention to the use of abductive methods in investigations of compositional relations. I illustrate my approach by use of Robert Hooke’s microscopic investigations of the cork and the development of the theory of the action potential.