Past talks | Eleanor March

Trinity Term 2025

Week 1 (1/5): Paolo Faglia (Oxford) - Perspectives in Special Relativity: against fragmentalism

Abstract: Things appear to be different from different perspectives. The standard reply to this perspectival variance is to claim that such differences are mere appearances and not part of reality. Lipman (2016) argues that special relativity (SR) radically extends this perspectival variance to most of the ways in which things appear to us. Given the wealth of properties subject to perspectival variance, Lipman argues that denying their reality is problematic. On the other hand, by adopting a picture of reality as fragmented, one can embrace the reality of such variant properties (Lipman 2018). I argue that in so far as such a fragmentalist metaphysics may be motivated, it is not motivated by SR. Instead, it is possible to understand SR within a single unified reality, and it is unclear whether the fragmentalist approach to SR delivers the advantages Lipman advertises.

I will start by explaining that SR does not establish the kind of perspectival variance envisaged by Lipman. As a consequence, Lipman's arguments are left severely weakened. I will then sketch an alternative to the fragmentalist account of SR, thus illustrating how one can can understand SR within a single, unified reality. A comparison between these two approaches shows that fragmentalism fails to deliver the advantages Lipman advertises.

Week 2 (8/5): Aditya Jha (Cambridge) - Thermodynamics has never been subjective, classical or quantum (CANCELLED)

Abstract: In a recent paper, Robertson and Prunkl (2023) - Is Thermodynamics Subjective? (Philosophy of Science, 90 (5)) - address a crucial concern with respect to thermodynamics: the claim that including epistemic factors - such as agents/observers and their varying degrees of credence - renders the theory subjective or anthropocentric, as it appears that fundamental thermodynamic distinctions (for instance, between heat and work) are observer-dependent. The authors contend that quantum mechanics can liberate thermodynamics from this issue as quantum probabilities are inherently objective; they allow for agent-independent assertions that are consistent with the thermodynamic description of a system, thereby establishing the objectivity of thermodynamics. However, contrary to what the authors argue, I argue that an observer's role is overstated and largely unnecessary for achieving objective thermal system descriptions. In typical states, thermal systems evolve toward their attractor distribution in an objective manner, independent of whether observers employ classical or quantum credences, indicating that the use of objective expectation values does not play a critical role in making objective thermodynamic statements. I further explain why factors such as coarse-graining, resource-relativity, and choice of measure - which although involve agent-dependent considerations - do not inject subjectivity into thermodynamics. The overall conclusion is that quantum considerations do not remove the anthropocentrism inherent in thermodynamics since there was none anyway.

Week 3 (15/5): Pascal Rodríguez-Warnier (Western) - Fluctuation theorems and (ir)reversibility in statistical thermodynamics

Abstract: In this talk, I show that the Jarzynski equality implicitly contains an alternative method for deriving equations of state for thermal systems through processes that are far from being thermodynamically reversible.

The thermodynamic state of a system is characterised by a small number of parameters, known as state functions (e.g., total energy, temperature, pressure, and entropy), which are well-defined only in equilibrium. Finding the so-called equations of state, which relate state functions, provides thermal physics with predictive power. Deriving these equations necessitates equalities between thermal quantities, which typically hold only for the special case of thermodynamically reversible processes. Reversible processes are conceived as the limiting case of irreversible processes in which imbalances between thermal quantities tend to zero.
The fluctuation theorems are a family of results in statistical thermodynamics which enable the derivation of equalities between the probabilities of thermal quantities and hold irrespective of the nature of the process. I propose that Jarzynski's equality - a fluctuation theorem equating the expectation value of the exponentiated work with the exponential of the free energy difference between a pair of states at the same temperature - offers a practical method for deriving equations of state without invoking reversibility (Jarzynski 1997, 2011). For this, I outline additional steps that allow us to estimate the free energy difference (and hence the partition function) at different temperatures. This means that the equations of state can be derived by analysing sets of processes that are arbitrarily far away from equilibrium. Such a move is possible only if we take the statistical fluctuations seriously. Reversibility is recovered in the limit as the set of irreversible processes tends to infinity, although no assumption is made about the actual existence of such a set.

Hilary Term 2025

Week 1 (23/1): Marta Bielińska (Oxford) - Prolegomenon to an epistemology of orbifolds

Abstract: In general relativity, spacetime is usually taken to be a (3+1)-dimensional Lorentzian manifold that is Hausdorff, paracompact, time-orientable, and so on. However, it seems that there are currently no known experiments that directly test all these spacetime properties. This raises the question of whether it is possible for our spacetime to have a different structure - for example, that of an orbifold, which is, roughly, a manifold quotiented by some group. In our talk, we will address two questions. First, is it possible to empirically verify whether one lives on a manifold or on an orbifold? Second, following up on Read and Bielińska (2022), can one identify the specific type of orbifold on which one is living; for example, is it globally or locally orientable? Our considerations begin with an examination of the topological and differentiable structures of orbifolds. We then discuss the possibility of defining field theories, such as general relativity or quantum field theory, on orbifolds. Finally, we conclude with broader philosophical reflections on epistemology of orbifolds. (Based on joint work with James Read.)

Week 2 (30/1): Jonathan Fay (Bristol) - On Reissner's hypothesis: Historical proposals for a Machian unification of gravity and inertia

Abstract: The combination of Mach's hypothesis concerning the material origin of inertia and Einstein’s equivalence hypothesis on the unity of inertia and gravity raises the intriguing possibility that gravity may be explained entirely as the dynamical part of a relativized law of inertia. Shortly after the development of general relativity, it became clear that Einstein’s general covariance based approach does not reduce inertia to mass interactions. While Einstein subsequently treated "Mach's principle" as a selection criterion for models of his theory, there is an alternative research program that implements Mach's idea explicitly in its foundations. The aim of my talk is to bring attention to some of the key 20th Century papers that (retrospectively) form part of this research program, focussing in particular on Reissner (1915) and Sciama (1953).

Although the approaches of Reissner and Sciama are quite different, they are unified insofar as they both embody this intriguing hypothesis. By looking at the common feature of such models, I draw out some of the key implications of the hypothesis, including the epochal variability of the gravitational 'constant' and a potential guarantee of the critical density condition in cosmology.

Week 4 (13/2): Maren Bräutigam (Köln) - Leibniz's Principle and similar particles

Abstract: In this talk, I will give an overview of the debate on whether Leibniz's Principle of the Identity of Indiscernibles (PII), according to which qualitative difference is necessary for numerical difference, is violated by similar particles (like, e.g., two electrons). According to so-called orthodox views, PII is always violated or, at best, valid in a weak sense only. According to heterodox views, however, (a strong version of) PII is valid for similar particles in at least some cases. The main motivation for heterodox views is that orthodox arguments for the violation of PII (tacitly) rely on a certain background assumption - called factorism - which can be questioned from a physical point of view. Factorism is a semantical view concerning the quantum mechanical formalism; more specifically, it is the view that indices within this formalism have physical (not just mathematical) meaning. However, factorism yields some implausible consequences; e.g., according to one prominent argument, it hinders quantum particles from becoming localized in the classical limit. Another important concept on which heterodox views rely is a certain understanding of entanglement - called GMW-entanglement, because it goes back to physicists Ghirardi, Marinatto and Weber - which is argued to fit better for similar particles than the usual understanding in terms of non-factorizability. Since heterodox views are relatively recent, they suffer from a problem - the so-called ambiguity problem - to which a complete and satisfying solution is yet to be found. The ambiguity problem shows that individuation via properties (i.e., what heterodox views want to accomplish on an ontological level) is not unique, i.e., ambiguous. I will discuss different aspects of the ambiguity problem and propose some potential solutions.

Week 5 (20/2): Eleanor March (Oxford) - "the logical sequence of his Principles": Understanding Du Châtelet on Newton's law of gravitation in the Principia

Abstract: Émilie Du Châtelet (1706-1749) is perhaps equally well-known for her magnum opus, the Institutions de Physique of 1740, and for her later French translation of and commentary to Newton's Principia (first published posthumously in 1756, with the corrected edition in 1759). One of the few topics which Du Châtelet addresses in detail in both the Institutions de Physique (chapter 15) and the commentary to her translation is Newton's arguments for his law of gravitation in the Principia. To date, however, no systematic comparison of the two has been undertaken (and very little has been said on either of them separately). I reconstruct and compare these two accounts. This offers a new perspective on Du Châtelet's developing thinking on the justification of Newton's law of gravitation within the Newtonian system.

Week 6 (27/2): Rupert Smith (Bristol) - Scientific representation and the hole argument: Insights from a structuralist approach

Abstract: Recent trends in this literature can be broadly characterised by the way they begin either from the mathematics or the metaphysics. I take Weatherall (2018) as the primary example of the mathematical approach, in which he argues that if one is only to reflect on the correct use of the mathematics, one can block or undermine the hole argument. For the metaphysical approach, I focus on a line of argument I identify in Pooley and Read (2021) in which they claim that arguments from mathematical practice fail to eliminate the relevant space of metaphysical possibilities. Specifically, they maintain that the hole diffeomorphic manifolds can represent haecceitistically different spacetimes. In response to these trends, the primary aims of this research are twofold: first, I articulate a structuralist account of scientific representation, building on Van Fraassen's structuralist empiricism (2008), that answers the demarcation problems posed by Frigg and Nguyen (2022); second, I aim to use this account to diagnose the reasons for, and provide solutions to, the recent impasse in the voluminous literature on the hole argument. While I am sympathetic to the arguments from mathematical practice, I argue that without a structuralist account of scientific representation Pooley and Read’s counterargument cannot be defused. While I am conceding this point, however, I am doing so in a minimal way: no such representation can meet the criteria of being a scientific representation. Further, if the points p and p' in the hole argument are to represent different events in spacetime, it is merely by means of stipulation. If, as I argue, scientific representation is to be understood structurally, for a manifold to be considered a model of a spacetime, it needs to include a representation relation, understood as some variety of structural mapping. Moreover, an isomorphism acting on a model would also need to have a corresponding transformation on the representation relation, which is at least sensitive to the strongest notion of structure in the manifold being considered. I conclude that, while mathematical practice alone fails to exclude the possibility of the hole diffeomorphic models representing haecceitistic differences, a structuralist account of scientific representation excludes the possibility of these classifying as scientific representations.

Week 7 (6/3): Wessel Vinke (Oxford) - Is there a hole in the hole argument?

Abstract: The hole argument can be interpreted as a problem for the substantivalist. Weatherall (2018) argues that this argument derives from an incorrect use of the formalism of general relativity. It is in essence a charge of equivocation against the hole argument, as Pooley and Read (2021) have called it. Here, I present two criticisms of Weatherall's account. Firstly, interpreting the term 'sameness' or 'equivalence' as being used in different senses is unfair to the original presentation of the hole argument. Secondly, Weatherall's objection depends on an unwarranted identification of the metaphysical issue at stake with the identity map in mathematical practice. (I may also be able to present an attempt at reconciliation with Weatherall's insight.)

Week 8 (13/3): Margot Stakenborg (Oxford) - Notions of causality in physics

Abstract: This will be half presentation, half discussion about different notions of causality. I will go over the notions of what causality is in philosophic literature, statistical definitions of causality, and discuss whether a viable notion of causation exists in fundamental physics. Critics of fundamental causation, like Bertrand Russell, argue that fundamental physical laws are time-symmetric and describe correlations rather than causal mechanisms. They claim that causation is an emergent, high-level concept rather than an intrinsic feature of physical reality. This skepticism is supported by the time-reversal invariance of classical mechanics, quantum mechanics, and general relativity, which challenges the idea of a directional causal structure at a fundamental level. However, there are forms of time-directed processes in physics suggest a causal structure, like the cosmological arrow of time, entropy, or in the definition of initial conditions of systems. I will also discuss the interventionist account of causality, by Judea Pearl. I will add to this discussion that we can also use atemporal definitions of causality to describe certain phenomena, e.g. matter causes space-time to bend, which raises the question if causality, in principle, should be time-dependent if its definition can be broadened beyond temporal constraints. I would like to facilitate a discussion on what we deem to be correct notions of causality at the end.

Michaelmas Term 2024

Week 1 (17/10): Eleanor March (Oxford) - What is minimal coupling, really?

Abstract: Minimal coupling is a heuristic used to construct dynamical equations for matter fields in general relativity. But as often been noted in both the physics and philosophy literature (since at least Trautman (1965)), standard explications of 'minimal coupling' are either unclear, ambiguous, or 'hyperintensional'. In this talk (based on joint work with Jim Weatherall), I provide a precise explication of minimal coupling which avoids these problems. I also discuss how this explication can be used to resolve other puzzles about minimal coupling in the literature.

Week 2 (24/10): Bryan Cheng (Oxford) - Empirically grounding representation

Abstract: Under the semantic view of theories, structural accounts of representation that invoke the mathematical notion of isomorphism are prevalent. However, isomorphism accounts face significant objections, with the most problematic question being how the physical world can be placed in the mathematical relationship with abstract models.

I compare two accounts of representation presented by Bas van Fraassen and David Wallace. While van Fraassen's account is based on isomorphisms, he supplements it with an appeal to the intentions of scientists, indexically locating users in the logical space of the theory. On the other hand, Wallace's 'theories and theorizing' account attempts to forgo the aforementioned issues by arguing that any scientific theory in use already has its own theory of representation built in - and that this process is not viciously circular.

To evaluate these proposals, I will focus on the role of reference, indexicality and empirical grounding in these accounts. I will argue that, because of disanalogies between languages and scientific theories, Wallace's account is less persuasive than van Fraassen's despite strong thematic similarities - and that it may be insufficient for Wallace's metaphysical view of maths-first structural realism.

Week 3 (31/10): Ray Pedersen (Oxford) - For one dendritic world

Abstract: Despite DeWitt’s claim that Everettian quantum mechanics (EQM) offers its own interpretation, there is no consensus regarding EQM’s best ontology. I organize the possible ontologies according to the two following choice dimensions: one versus many concrete histories, and overlapping versus diverging histories. Accounts such as Conroy’s (2018) Everettian actualism, according to which there is one concrete world among many possible worlds that together form the branching structure, fail to account for precisely how our world earns the honor of being concrete. Regardless of whether the worlds themselves are diverging or overlapping, such an account requires the introduction of stochastic processes to EQM which would deprive EQM of its appealing deterministic nature. On the other hand, metaphysical accounts of EQM that posit many concrete worlds, such as Wilson’s (2020) quantum modal realism, face the serious objection over ontological extravagance, for they demand an infinite number of concrete worlds with identical initial conditions that just happen to together be governed by the global wave function. If one desires a parsimonious and genuinely deterministic account of EQM, one must entertain alternative options.

Fortunately, in my taxonomy, one type of ontology remains: many overlapping concrete worlds. This resembles early metaphysical readings of EQM, such as that of Dewitt (1970) and what Wallace (2014) calls the hydra view. Initial versions of the hydra view, however, cannot withstand the identity problem: post-branching, if there are now multiple agents where there once was one, which agent is really the original agent? How can one possibly make sense of the identity of objects on such an account? To dissolve this issue, I argue for one dendritic world. To any observer within the branching structure, it appears that measurements yield determinate outcomes, and this novel approach demands a new mereological account of the universe and the objects within it. In addition to material parts and temporal parts, I propose branch-like parts (BLiP). Each object in this singular branching world is the mereological sum of its BLiPs; when an agent-BLiP undergoes decoherent branching, it is simply the case that they split into additional agent-BLiPs. This new mereological relation allows one to avoid common objections against one world metaphysics of EQM by increasing the structure of our world, while simultaneously decreasing the number of worlds that we must posit from an infinite number to only one.

Week 4 (7/11): Will Wolf (Oxford) - Permanent underdetermination in dark energy and inflationary cosmology

Abstract: I argue that both dark energy and inflationary cosmology represent cases of so-called 'permanent underdetermination' of theory by evidence. I then present (a) a taxonomy of possible responses to underdetermination, and (b) an understanding of both dark energy and inflationary cosmology from an effective field point of view, in order to argue that while there is a response that can partially alleviate some of the concerns about underdetermination in the dark energy sector, these concerns persist in the inflationary sector.

Week 5 (14/11): Maria Avramidou (Oxford) - The problem of chaos in philosophy of science

Abstract: Science typically operates under the assumption that with more research, regular and predictable patterns will be discovered in areas where regularity is not yet evident. When scientists encounter a complicated pattern, they often search for underlying periodicities that may be obscured by random noise.

Chaotic systems are characteristically hard to predict despite being governed by deterministic dynamics. They appear random due to their non-periodicity and exhibit sensitive dependence on initial conditions; a small change in the initial state of a deterministic nonlinear system can result in large differences at a later time. This suggests the critical importance of precise measurements for making accurate predictions. In practice, the reliability of predictions is compromised by three principal factors: i) the inherent inaccuracies in measurements; ii) the limitations in computational capacity; iii) the necessity of approximating continuous dynamics using discrete time series. These factors result in uncertainty about the state of the system, leading to a form of underdetermination of theory by data. Following Li and Yorke's (1975) definition of chaotic systems, I discuss the problem of building and testing faithful models of chaotic systems.

Week 6 (21/11): Frank Cudek (Oxford) - Localising singular structure of spacetime

Abstract: I argue that the singular structure of spacetime is best conceptualised as a non-local, but localisable, property of spacetime regions, and that there is no unique, best definition of a singular region of spacetime (there is, however, a schema that any adequate definition instantiates). I also discuss to what extent, assuming the definition of singular structure in terms of b-incomplete curves, one can make precise sense of notions such as ‘x is close to singular structure’ and ‘x is closer to singular structure than y’, where x and y range over spacetime points.

Week 7 (28/11): Paolo Faglia (Oxford) - Relational quantum mechanics does not solve the measurement problem

Abstract: It’s been almost 30 years since the relational interpretation of quantum mechanics (RQM) was introduced by Carlo Rovelli (Rovelli, 1996) and, over time, the interpretation has changed and developed significantly. Still, I claim that current formulations of RQM do not solve the problem of measurement. Moreover, I argue that the prospects of articulating a version of RQM which both meets its own goals and solves the measurement problem are not promising. The problem is the following. All versions of RQM explain the world in terms of an ontology of systems and events, where an event consists of a variable of a system taking a value relative to another system. I argue that, in order to solve the problem of measurement, RQM needs to offer a specification of the circumstances in which events occur. Current formulations of RQM claim that events occur whenever interactions occur. However, no precise characterisation of the notion of interaction is given. I develop the most plausible ways of understanding the notion of interaction in the context of RQM, but I show that they fail to provide a satisfactory specification of the circumstances of the occurrence of events. In light of the failed constructive efforts, I conclude that the prospects for formulating a version of RQM which both satisfies its aims and solves the problem of measurement are dim.

Week 8 (5/12): Marta Bielińska (Oxford) - Prolegomenon to an epistemology of orbifolds (CANCELLED)

ELEANOR MARCH