I am close to completing a book Knowing Science which brings together my work in epistemology and philosophy of science. Its principal themes are: (i) radical anti-empiricism and (ii) the centrality of knowledge to philosophy of science. The latter places my work within ‘knowledge first’ epistemology, as applied to science (which has hitherto ignored knowledge first). I propose a functionalist account of evidence, which entails Williamson’s equation, Evidence = Knowledge. This implies that anything that can be known can be evidence, and hence there is no special role for perception in evidence. I argue that insofar as the concept of observation is important for understanding science, it is not a perceptual concept. Indeed, observations can be made by machines. The latter suggests that the impersonal notion of ‘scientific knowledge’ is not to be reduced to the knowledge of individual scientists. And so I articulate an understanding of social or group knowing that does not supervene on the mental states of scientists, and in which the infrastructure of science plays a role. I develop a view of theoretical inference in science according to which knowledge is gained by through ‘inference to the only explanation’ (or ‘Holmesian inference’). Inference to the best explanation is an approximation to Holmesian inference that allows us to assess the plausibility of a hypothesis and which bears a heuristic relationship to Bayesian inference.
My work in the philosophy of medicine is in part an application of the work described above under ‘Knowing Science’. A long term project is a book combining the philosophy and history of medical methodology. Of past and present doctors and their practices, we can ask ‘What did they know? Did they know that this practice would work? And if so how did they know?’ In my view modern medicine starts in 1721—not because of any novel medical practice but because of an innovation in methodology. The 1720s saw the first systematic project for assessing the efficacy and safety of an intervention (smallpox inoculation). Thereafter further methodological progress led to an increasing ability of doctors to distinguish what works in medicine from what does not. I trace this through the nineteenth to the twentieth century and the advent of the randomized trial and the recent Evidence-Based Medicine movement, while giving an philosophical analysis of the epistemological contribution made by each. This project includes publications on Hill’s ‘criteria of causation’, the EBM hierarchy, and James Jurin’s analysis of smallpox inoculation.
The replication crisis raises questions regarding the reliability of science, methodological norms, and research ethics. It therefore ought to be of central interest to philosophers of science, although to date only a small number have engaged with it. My particular interest focuses on the role of null-hypothesis significance testing in explaining (in part) the existence of failures to replicate. Roughly, improbable hypotheses plus a significance level that is not negligible (e.g. the standard 5%), can be expected to lead to a high proportion of false positives (which then fail to replicate). If that is an important part of the diagnosis, then what should be done? Require more stringent statistical tests? Acknowledge a lower degree of confidence in successful hypotheses? Or reduce the proportion of implausible hypotheses put forward for testing? My interest is not merely theoretical: I am a founder member of the UK Reproducibility Network, which aims to improve research practices in UK (and global) scientific research.
Standard accounts of creativity take creativity to be a disposition to create novel works that are valuable in some way. Creativity itself is therefore always to be valued highly. In a series of collaborative papers, Alison Hills and I argue that these views are mistaken. Creativity can sometimes be exercised in the production of works of no value. In our view, what is distinctive about creativity is that it involves exercise of the imagination. Since creativity is not necessarily valuable on our view, it is important to ask the question: under what circumstances does creativity produce value? We emphasise the importance of a tradition of valuable work as providing exemplars that guide the imagination. Our work on creativity is applied to both the arts and the sciences. My particular interest concerns the relationship between creativity and scientific rationality. These are usually held to be entirely distinct—a manifestation of the distinction between context of discovery and context of justification. Whereas our work leads us to reject this distinction.
Here my interest has focussed mainly on metaphysical questions. For example, do natural kinds exist? And if so what sort of entity are they? Can a posteriori essentialism be defended? What is the relationship of natural kinds to the laws of nature? I argue that natural kinds exist, not simply in the sense that there are natural divisions of things into kinds, but in the strong realist sense that natural kinds are themselves a species of entity. Although, kinds are entities, they are not fundamental entities. Katherine Hawley and I argue that kinds are complex universals. Natural complexes of universals can have essences. I am also interested in how the very possibility of a particular natural kind constrains non-fundamental laws concerning that kind. I argue that the these constraints mean that there is much less contingency in the world than there appears to be. Future work will integrate the metaphysics, philosophy of language, and philosophy of science of natural kinds. For example, the view that natural kinds are complex universals provides an answer to the question: what is rigidity for natural kind terms?
POWERS AND PROPERTIES
In my Nature’s Metaphysics (2007) I argued that the fundamental properties of physics are essentially dispositional properties (powers). In the last decade the ontology of powers has become very popular, and powers have been advertised as the solution to a wide range of philosophical problems, from morality through causation to free will to ethics. Many of these ‘solutions’ require the powers in question to be non-fundamental properties. Nonetheless, I regard the case for non-fundamental powers (in addition to fundamental powers) as not yet made—and so these applications of the powers ontology are spurious. So I am interested in the question: which non-fundamental properties are powers, if any? And if there are non-fundamental properties that are not powers, what characterises the nature of those properties?