In light of recent criticisms by Woodward (2017) and Rescorla (2018), we examine the relationship between mechanistic explanation and phenomenological laws. We disambiguate several uses of the phrase “phenomenological law” and show how a mechanistic theory of explanation sorts them into those that are and are not explanatory. We also distinguish the problem of phenomenological laws from arguments about the explanatory power of purely phenomenal models, showing that Woodward and Rescorla conflate these problems. Finally, we argue that the temptation to pit mechanistic and interventionist theories of explanation against one another occludes important and scientifically relevant research questions.

Some have defended a Fregean view of perceptual content. On this view, the constituents of perceptual contents are Fregean modes of presentation (MOPs). In this paper, I propose that perceptual MOPs are best understood in terms of object files. Object files are episodic representations that store perceptual information about objects. This information is updated when sensory conditions change. On the proposed view, when a subject perceptually represents some object a under two distinct MOPs, then the subject initiates two object files that both refer to a.My defense of this view appeals to its satisfaction of four constraints that I argue theories of perceptual MOPs should satisfy. Furthermore, I show that some existent accounts of perceptual MOPs fail to satisfy them. The defended constraints also indicate what is unique about perceptual, as opposed to linguistic or cognitive, MOPs.The following discussion gives an account of perceptual MOPs of objects. I ultimately leave it as an open-ended question whether there are MOPs of perceptual attributes and events. 1 Furthermore, on the assumption that there are MOPs of attributes and events in perception, I leave it as open-ended how the proposed account might be applied to them. MOPs of attributes and events raise another set of questions. 2 I do not address these questions in the present paper. In section 1.1, I outline the distinctive epistemic situation that agents are in when they perceive an object under different MOPs. Then, in section 1.2, I distinguish perceptual MOPs from what Crane (2003) calls intentional modes. As I will argue, accounts that conflate intentional modes with MOPs are not feasible. In section 1.3, I argue that perceptual MOPs are not individuated by the representation of distinct properties attributed to their referent. For example, a representation of an object a with the attributed property F does not necessarily constitute a distinct MOP from a representation of a with the attributed property G. In section 1.4, I argue that a theory of perceptual MOPs should elucidate how the content of perceptual experiences enables singular thinking. These four subsections provide a framework for analyzing the plausibility of theories of perceptual MOPs. Furthermore, the framework indicates what is unique about perceptual, as opposed to linguistic or cognitive, MOPs. After providing this framework, in section 2, I give an object files account of perceptual MOPs and I show how the account meets all the specified constraints. Finally, section 3 looks at some existent views on what perceptual MOPs are, i.e., Chalmers (2004) and Schellenberg (2018).In this section, I show that these views fail to meet some of the constraints specified in section 1."MS is a body illuminated by the sun," while rejecting (b) "ES is a body illuminated by the sun."Since it is possible for a rational agent to accept (a), while rejecting (b), given that the only difference is that 'MS' is replaced by 'ES,' these expressions meet Frege's criterion for having different MOPs (1892: 1...