When are two tools better than one? Mortars, millingslabs, and the California acorn economy

• We present a formal model of tool manufacture, use, and discard.
• We model trade-offs in the use of generalized vs. specialized tools.
• Generalized tools are favored when subsistence tasks are evenly distributed.
• Increases in task predictability and payoff certainty favor more specialized tools.
• California plant processing tools are modeled relative to acorn intensification.

The question investigated by this study is: how much behavioral specialization is necessary before tool specialization is worthwhile? The toolkits of hunter–gatherers vary considerably over space and through time from simple and multifunctional, to complex and specialized. The decision to use one tool over another can be modeled as a fairly straightforward consideration of costs and benefits, but the problem becomes more complex when individual tools are employed in multiple tasks. We introduce a formal model that helps explain when and why multi-use, or flexible tools, might outperform specific-use, or specialized tools, or vice versa. This model is used to help understand the adoption of mortars when acorns became a staple food in prehistoric California. The model suggests specialized tools win out when tasks they are designed for are performed often enough, or occur with enough certainty, to make their added cost worthwhile.