Do we need to explain initial conditions? It is common to think of the universe as a grand time-directed process that started out in some initial state — call this the ‘time-directed universe’ hypothesis (TDU). On TDU, the initial state is explanatorily unique — it is the only one that did not evolve from some prior set of conditions. Some have appealed to this explanatory uniqueness to suggest that it is misguided to seek an explanation as to why the early universe was extremely low-entropy, and so argue that TDU plays an important explanatory role in physics. But what if we reject TDU? This talk considers the options for those that assume a temporally adirectional metaphysics, which I call the ‘C theory’. Given the C theory holds there is no intrinsic difference between ‘initial’ and ‘final’ states of physical systems, it is unclear what we are to make of the explanatory demands of the low entropy early universe. I assess a series of options for the C theory, arguing that the rejection of TDU leaves us no worse off with regard to explaining the low entropy early universe.
Before the talk, there was a reading group with the speaker. The paper we discussed was “Measures, Explanations and the Past: Should ‘Special’ Initial Conditions be Explained?” by Craig Callender, available here: https://doi.org/10.1093/bjps/55.2.195
On Sunday 23 June, Katie Robertson gave a talk in Groningen at a workshop on Probabilities in Cosmology, as part of a stellar lineup of speakers including Sabine Hossenfelder and Robert Wald.
Katie’s title was “Stars and Steam Engines: To What Extent do Thermodynamics and Statistical Mechanics Apply to Self-Gravitating Systems?”. The conclusion was an irenic resolution to recent debates about the astrophysics of elliptical galaxies – statistical mechanics does apply to them, but thermodynamics doesn’t.
On Monday 17 June, Katie Robertson spoke at the Sigma Club seminar at LSE, on the ‘holy grail’ of philosophy of thermal physics: how to reduce the second law of thermodynamics to statistical mechanics.
Katie’s conclusion is that once we get properly clear on the target of the reduction – on what grail it is we’re seeking! – then the Gibbs entropy of statistical mechanics can be shown to play the role of the thermodynamic entropy.
A supplementary FraMEPhys Seminar was given by project visitor Dan Marshall (Lingnan University) on Monday 10th June 2019.
Dan’s title was “Facts and Grounding”.
Abstract: The most popular theories of the individuation conditions of facts are the coarse-grain theory, according to which facts are identical if and only if they are necessarily equivalent, and the structure theory, according to which facts are structured in the same kind of way sentences are structured. Despite their popularity, both these theories have serious problems. In this paper, I propose a new moderate-grain theory of facts that avoids these problems by individuating states of affairs more finely than the coarse-grain theory and more coarsely than the structure theory. I then defend the proposed theory from the objection that it is incompatible with widely accepted principles of grounding by arguing that these principles of grounding are false and should be replaced with alternative principles.
The next in our series of FraMEPhys Seminars was given by Mark Pexton (Durham) on Monday 3rd June 2019.
Mark’s title was “Contextuality, Emergence and Unification in Physics”.
Abstract: A contextual account of emergence and unification is presented. It is argued that emergent phenomena can be thought of as the consequence of system/context interactions. Contexts often involve modal relations not contained in the first order level of the system in question, hence although the system itself may appear reducible, the combination of system and context is not. Unification as an explanatory strategy can sometimes be seen as linked with reductionist intuitions – by considering their reduction base, disparate systems can be shown to be different manifestations of the same underlying phenomena. However, unifications do not proceed via reduction bases alone. Sometimes they involve moving up a ‘level’ of modal space to unify disparate microphysical phenomena by considering unifying features of the properties of aggregates (such as in universality in critical phenomena). It is argued that unification itself as an explanatory strategy (and therefore putative guide to ontological commitments) is itself highly contextual. Unifications can proceed by shifting the demarcations between systems and contexts to provide new system/context boundaries that create different sets of ‘similar’ unified physical phenomena. As such unification does not easily fit a standard paradigm of reduction or emergence.