For me, the advantage of computational methods is precision and scale: First, we can describe language with a level of detail that no human could keep in their memory. Second, we can observe this precision at a scale that allows us to see how language functions as a complex system. My particular focus is the interaction between how language emerges within individuals and how it varies across populations. I’m working on a joint model of these two problems that can be tested at scale in realistic computational simulations.