What is the ‘past hypothesis’, and what is it trying to explain?
Submitted for PH373 (Philosophy of Time) Take Home Exam
The past hypothesis is a cosmological proposition suggesting that the universe began in an extraordinarily low-entropy state. First articulated by Ludwig Boltzmann and later developed by philosophers and physicists like David Albert and Roger Penrose, the hypothesis addresses the thermodynamic arrow of time – that is, the unidirectional increase in entropy observed across the universe. By positing an initial condition of low entropy, the past hypothesis explains why entropy rises over time, which in turn explains our perception of temporal asymmetry – that there is distinction between a fixed past and an open future.
Central to the hypothesis is the insight from statistical mechanics that low-entropy macrostates – highly ordered configurations, like a gas concentrated in one corner of a container – are realized by a small number of specific microstates. In contrast, high-entropy macrostates, such as a uniformly distributed gas, correspond to vastly more possible microstates. This discrepancy makes low-entropy states statistically improbable and high-entropy states overwhelmingly likely (Boltzmann, 1897). The laws of physics are time-symmetric, and temporal symmetry implies that entropy should statistically increase toward both the future and the past. In spite of this, as (Price, 2004) observes, the universe exhibits a striking temporal asymmetry, with entropy only increasing toward the future. Thus, entropy’s continuous increase over time demands an explanation.
This is the issue the past hypothesis seeks to resolve, by treating the low-entropy state as a boundary condition of the universe’s initial state. From this starting point, the statistical tendency for entropy to increase explains the observed thermodynamic asymmetry. For example, (Callender, 2004) notes that the second law of thermodynamics naturally follows from a low-entropy initial condition, clarifying why systems evolve predictably toward a higher-entropy equilibrium.
Entropic asymmetry means that entropy increases towards (what we call) the future, thereby allowing structures to form and persist in ways that encode information about the past, such as fossils or photographs. This explains why we can have knowledge of the past but not of the future: the processes that preserve information are themselves shaped by the entropic gradient. Unlikely and ordered states tend to arise and persist in the past-present direction, but dissipate in the present-future direction. This epistemic asymmetry, in turn, grounds other temporal asymmetries – causal, practical, and metaphysical – that structure our experience of time: we think we can affect the future but not the past, and view the past as fixed and the future as open, because we believe we can have knowledge of the past but not the future (Dummett, 1964). The past hypothesis provides a foundation for explaining all these temporal asymmetries by accounting for the entropic asymmetry that underpins them.
There are some challenges with the past hypothesis as an explanation, in particular how improbable it is. (Penrose, 1989, p. 344) calculates its probability as
, suggesting that such a low-entropy state is extremely unlikely under standard statistical mechanics. Though critics like (Price, 2002) maintain that such an improbable boundary condition must be justified, Callender argues that the past hypothesis does not itself demand further explanation. He defends it as being a law-like “brute fact”, which when integrated into our existing scientific understanding, constitutes the best available explanation for the entropic and temporal asymmetries we observe.
To conclude, the past hypothesis posits an initial low-entropy state to explain entropy’s rise and temporal asymmetries, offering an explanation for the direction of time and our perception of it.
Bibliography
Boltzmann, L. (1897). Lectures on Gas Theory.
Callender, C. (2004). There is no Puzzle about the Low-Entropy Past. In C. Hitchcock, Contemporary Debates in Philosophy of Science (pp. 240-256). Blackwell Publishing Ltd.
Dummett, M. (1964). Bringing About the Past. The Philosophical Review, 338-359.
Penrose, R. (1989). The Emperor's New Mind. New York: Oxford University Press.
Price, H. (2002). Boltzmann's Time Bomb. The British Journal for the Philosophy of Science, 83-119.
Price, H. (2004). On the Origins of the Arrow of Time: Why there is Still a Puzzle about the Low-Entropy Past. In C. Hitchcock, Contemporary Debates in Philosophy of Science (pp. 219-239). Blackwell Publishing Ltd.
Result
Mark: 88% (High First)
Feedback: Your second answer again provides a very good account of the issues at stake in discussions of the past hypothesis. Some of the material on how the entropic arrow explains the other arrows of time tends a little too much towards the merely descriptive. Here it might again have been an idea to be slightly more selective and cover one point in more detail. (You can always signpost the fact that you are doing so.) This might also have left space to expand a little more on Callender's solution as to how we can treat the past hypothesis as a law.
How to do even better in the future (generic across this and another answer):
This is very good. As indicated above, perhaps sometimes even more thought could go into trying to explain things in a way that even a complete novice to the relevant literature would understand. Another thing I would recommend is having a sentence at the beginning that states what you aim to argue for in your answer. This gives your answer a sense of direction.