Thread by @jonasnoelle

Cool work on complexity and simplicity in language evolution across species by @Limor_Raviv and @cedricboeckx. They start with an interesting discrepancy between animals and humans in how social complexity shapes the complexity of their communication systems #Protolang7

An important distinction we need to make is whether we are talking about grammar or simple signal variation, and what 'simple' or 'complex' actually refers too. The mirror pattern we see might relate directly to how we distinguish these concepts.

In animal communication research, the social complexity hypothesis contrasts on the surface quite directly with the linguistic nich hypothesis by @glupyan et al, suggesting a seemingly disciplinary conflict

However, terms like simple and complex are used quite differently when talking about grammar-like rules vs signal variation, e.g., holistic constructions in lang can be very complex but animal communication systems that have some syntactic structure are considered complex

Where does this differentiated use of simple/complex come from? Perhaps it relates to memory constraints vs productivity depending on whether we are looking at a language or animal communication system. Also the terms are socially connotated (simple can both be inferior/superior)

But with respect to grammatical structure, both fields actually converge (increase in structure = more complex). In both fields the system that has a reduced inventory/signal variation is regarded simpler. But the fields diverge with their conclusions re group size effects!

In animals increase in signal variability is good for individuation, whereas e.g. sign languages evolving in smaller communities (ABSL) show greater signal variabiliy, in contras to those evolving in bigger communities (ISL) - reduced signal variation here is good for converge

Why is more variability considered more beneficial for for humans than for animals? This could relate to the problem of detecting meaning and grammar in animal structures, and the fact that in this field combinatoriality and compositionality are often use synonymously.

We should therefore move away from terms like complex and simple, and instead should formulate their predictions using very descriptive language. See their upcoming paper for more details!