Freddy Purcell discusses Tyler Brunet’s talk on inevitability in branching models of time and why it may not be so easy to change the past.
In another Phil on Tap first, Tyler Brunet joined us on a quiet night in the White Hart. As an avid fan of Tyler’s second year module on Metaphysics, I was very excited to hear more about a topic I have strangely heard little about since, but enjoy greatly.
Tyler’s work on time travel is partially “philosophy at fiction”, where a philosopher finds an interesting concept within fiction and attempts to make it consistent, with hope that this will also generate some philosophical dividends. Lewis (1976) does this with time travel fiction by explaining how certain tropes, like interacting with a future version of yourself, are strange but consistent. Lewis, however, works on a specific conception of time and time travel that Tyler diverges from in his work.
A typical conception of time is linear, where events are simply organised by relations of happening before or after each other. Lewis works with this model of time, construing it as deterministic. This means he believes that while a time traveller could affect events in history, they cannot change it. So, while a time traveller could be a part of why a historical event like the French Revolution happened, they could never prevent it from happening. Tyler understands this idea of time travel as self-correcting as it prevents the traveller from changing the past. This contrasts to non-self-correcting time travel where you could go crazy and change anything. Non-self-correcting time travel often goes hand in hand with a branching model of time, where if a traveller were to go back in time to change something, they would set the course of history down a different path.
While self-correcting time travel is often more popular in fiction, perhaps because it is taken to be more coherent, Tyler is interested in fictions that sit in the middle of self-correcting and not. One example is Steins;Gate, where protagonist Okabe attempts to use time travel to save his friend Mayuri from dying. He is successful but finds that she dies again in a different way. And again, and again when Okabe keeps trying to save her. Similarly in the 2002 film ‘The Time Machine’, inventor Alexander Hartdegen manages to travel back in time to prevent his fiancé’s murder, in which he is successful, but she soon dies again in a different way. These narratives obviously don’t fit into the self-correcting model as the protagonists are able to change the past. However, they still fail to prevent the death they want to change. Steins;Gate’s explanation for this failure is that time operates in a branching model, but that there are attractor fields that bring branches together on an event, like Mayuri’s death. Okabe’s challenge is therefore to go back in time to find a branch outside of this attractor field that doesn’t lead to Mayuri’s death.
Tyler takes Steins;Gate’s explanation on board, claiming that there are many physical attractors that bring things into a set point. Gravity brings a pendulum back to its resting place (which Tyler demonstrated with a handcrafted pendulum) and sinks pull water down their hole. Tyler argues it therefore makes sense to conceive of attractors in time as well. To do this, he took a proposition like “Mayuri is dead”, arguing that there is a set of all times where that proposition is either true or false. You can then think of a time attractor like a basin, pulling in all collections of times that lead to that proposition being true. If you are on one of the branches of time in this attractor, then that event is inevitable. Tyler argued that this makes sense of fictions like Steins;Gate, but also gives us a way of thinking about inevitability that isn’t deterministic. Events could be different on a different branch of time, but some things are inevitable once you get pulled into an attractor.
Time travel is objectively one of the coolest things you can study, and Tyler managed to deliver some technical material with clarity and humour, making this a great talk! We had a number of fascinating questions, including what attractors would look like in models of time. Tyler described a typical attractor like a basin, leading a load of branches into an inevitable event. However, he also said that you could imagine a leak in the basin, where some branches appear to head towards an inevitable event, but suddenly diverge and escape the basin. So, there would be many ways to conceive of these models.
Thank you to Tyler for the talk! Be sure to check out his paper (2025) on this topic if you want a more detailed explanation of his thinking (or take Metaphysics). As always, leave a comment below if you have any questions or reflections on this talk.
Bibliography:
Brunet, T. (2025) “Kripke Attractors”, Erkenntnis, June.
Lewis, D. (1976) ‘The Paradoxes of Time Travel’, American Philosophical Quarterly, Vol.13, No.2, pp.145-152.