Taking a break from my Carnap streak for at least a day, I intend to give an overview with some comments on Michael Tomasello's The Cultural Origins of Human Cognition in the context of a seminar on the evolution of cognition. First, however, a warning: I am not really an expert on any of this material. I am just writing about what struck me having read the book.

Tomasello's book stands on several pillars:

1. Human cognition is significantly different from NHP cognition.
2. A modular approach to human cognition has several problems, not the least of which is the time-scale on which the modules must have evolved.
3. If humans had evolved an ability to see conspecifics as intentional and mental agents like themselves, then cultural learning/evolution could account for the distinctly human aspects of cognition.
4. Human historical and ontogenetic patterns provide evidence that the suggested ability is a better explanation than modular claims, due to the sequential and cumulative nature of human development.

While Tomasello musters some substantial evidence, especially for claim 1, there are several lacking areas. First, the part of the plausibility argument of 2/4 based on probability of distinct mutations happening seems somewhat strange. The probability of a mutation at any given point is so small, that whether one happened leading to Tomasello's suggested ability, or a small slew happened at once leading to the development of modules doesn't seem to be as big a deal to me as Tomasello would seem to claim.

The story for 3 is somewhat sketchy in terms of evidence, but Tomasello does make it sound plausible besides that. The story goes something like this:

1. Children develop the ability to see other humans as like themselves.
2. Joint attentional scenes allow children to begin seeing other humans as intentional agents.
3. This then allows for the development of basic language. Seeing others as intentional agents allows a child to determine the intent of words used in things such as naming games. Additionally, it allows for general imitation learning as opposed to the emulation learning of NHPs.
4. Children abstract due to social and individual pressures from simple language to verb island constructions, and later to more abstract grammar.
5. Greater social interaction and the development of seeing other humans as mental agents, having beliefs and desires like the self allows children to think discursively, develop simultaneous representability, self-regulation, and meta-cognition.

This is all quite interesting, but from the perspective of the course, it is somewhat disappointing. Tomasello focuses most of his efforts on point 3, whereas the seminar is most interested on what the nature of cognition is and how it might have evolved. Hopefully what we will be reading for the rest of the quarter will have a greater relevance to that topic.