|2018/04/27||Erik Hoel||Emergence and causal structure in science
|Columbia University||Araya Inc, Toranomon 15 Mori Building 2F 2-8-10 Toranomon, Minato-ku Tokyo 105-0001 2|
Debates about emergence and reduction have raged for millennia. Generally most agree that science should both in practice and in theory attempt to reduce systems down to their most atomic constituents (the micro-level) to understand and explain them. However, at the same time, science spans a vast array of different spatiotemporal levels, and the mechanisms and explanations of scientists are rarely given in terms of their fundamental microphysical processes. Reconciling these two facts, recent research drawing on the application of information theory to causal analysis has revealed that the causal structure of the higher levels of a system can indeed sometimes bear more information, do more causal work, and make for better causal models. This allows us to formalize the age-old idea of emergence mathematically as a form of noise reduction in causal relationships, and has consequences for how scientists model, understand, and experiment on systems at a micro and macro-level.