PVV Reviews

My first exposure to the concept of emergence — the phenomenon of a given system containing features that cannot straightforwardly be attributed to its constituent parts — as an object of intellectual interest came in the form of Douglas Hofstadter’s classic Godel, Escher, Bach, which I read around seven years ago. Hofstadter’s book does not study the concept of emergence, but I think it’s ultimately about how intelligence and meaning emerge from seemingly unintelligent and meaningless lower-level processes. In subsequent years, I mainly came across the philosophical concept of emergence in the context of discussions on consciousness. Recently, however, I’ve come to understand it’s a broader concept that can stand alone as a subject of philosophical inquiry. There are a few places I could have started (for example, the Stanford Encyclopedia of Philosophy entry on emergence is written by Tim O’Connor), but Mark Bedau’s name came up a few different times as an influential philosopher in this area, so I decided to give his 1997 paper Weak Emergence a shot.

A common example of emergence (given by both Bedau and O’Connor in his SEP article) is a tornado, which is obviously made up of numerous components (dust, debris, wind, or if you want to go smaller, molecules, atoms, etc.) but which as a macro state seems to be more than the sum of its parts and can act autonomously from them (to steal from O’Connor, consider the fact that you don’t need a degree in particle physics to understand how tornadoes work, even though particles ultimately comprise the tornado). Bedau’s paper is concerned with what he calls “weak emergence”, which occurs when the macro state of a given system that has a “micro-dynamic” governing how the system’s micro states evolve over time can be derived from the micro-dynamic and the external conditions of that system via simulation.

This definition took me a few reads to get my arms around, but the idea is essentially that we can say a system (say, a tornado) is weakly emergent if and only if we can take (1) the physical laws governing the evolution of the system’s micro-states over time (so, the rules of particle physics or quantum mechanics or whatever that tell us exactly what the parts making up the tornado, such as the molecules and atoms and the like, will look like from one moment to the next), and (2) any conditions external to the system (so any outside influences that might interact with the system but aren’t themselves constitutive of the system), and input the information from (1) and (2) into something (say a computer) that can simulate what will happen over time given that specific input. Then, the results of that simulation give us the macro-state.

That explanation was a bit wordy but I think makes it a bit clearer what Bedau wants to call “weak emergence.” It’s also important to distinguish this from “strong emergence”, which is just a system which cannot be reduced to just its micro-states. In other words, while there may be micro-states (e.g., particles and the like) that make up the system, the system has emergent qualities that cannot be explained purely by reference to how the micro-states evolve over time, making it truly more than the sum of its parts (and, on some accounts, involves “downward causation” on its lower-level, constitutive parts). Bedau, and many other philosophers and scientists, reject the entire notion of strong emergence, likening it to magic and criticizing it as unscientific.

Bedau’s paper then goes through a few examples of weak emergence , such as Conway’s Game of Life, explains how his definition can map on to the study of complex systems, and argues that the concept of weak emergence is scientifically useful since it gives us a framework and vocabulary to study complex systems despite it ostensibly not being an especially interesting phenomenon (since it’s obvious to everyone that systems are just made up of constitutive parts).

Weak Emergence was interesting and framed the issue well enough, but it was not immediately clear to me why it’s considered an important paper in the history of the philosophy of emergence. Some further reading on this topic, however, explained to me that Bedau’s real contribution here was the idea of simulation as a necessary condition to studying the weak emergence of a system (and importantly, the evolution of the micro-states need only be able to be simulated in principle to qualify — certain systems may be so complex that we lack the current computing technology to simulate their evolution, but we would still be able to call the system weakly emergent from its micro-states).

Bedau’s paper is really focused on whether the concept of weak emergence is scientifically interesting/useful, but it’s helped spark my own interest in the topic of emergence more generally, since it seems to bear on a lot of interesting questions. For example, is phenomenal consciousness (i.e., subjective, experiential, qualitative feelings) weakly emergent such that it can be fully described by the neurochemistry of the brain? I’m also just intuitively skeptical of the claim that systems can ultimately be described in terms of the pieces that compose them + the laws of nature. Opponents of strong emergence agree that there are different levels of description that are more or less appropriate for a given system — for example, even if a tornado is just weakly emergent from its parts, it would not be particularly helpful if the weatherman tried to describe its flight path at the molecular level) — but I’m sympathetic to the idea that there are systems that cannot be described at all in terms of the micro-level interaction of their constituent parts. There’s also something appealing to me about the idea of systems exerting downward causation on their micro-states, as I don’t really see why we privilege the “bottom up” perspective of systems over the “top down”.

Ultimately, I’m guessing I’ll be persuaded against strong emergence by arguments that it requires us to put credence in the existence of phenomena that are beyond our current best understanding of physical reality. But for now it’s fun to believe!