Data from: Technological complexity and combinatorial invention in small-scale societies
Data files
Sep 29, 2025 version files 21.06 KB
-
README.md
2.64 KB
-
tool_data.xlsx
18.42 KB
Abstract
Technology plays a central role in all human lifestyles, from foraging to industrialized economies, so understanding what causes technological complexity to vary over time and space is central to understanding the diversity of human adaptive strategies. While technology provides adaptive solutions to physical problems, it also comes at a cost, leading to trade-offs. Here, we focus on the optimization of technological complexity in small-scale societies, reflecting this balance between costs and benefits. Specifically, we study the relationship between toolkit richness T, defined as the number of distinct tools in a toolkit, and tool-part richness $P$, the number of unique components within a toolkit. Using data from ethnographic studies, we show that toolkit richness scales sublinearly with tool-part richness as T ~ P0.7. This result reveals diminishing returns, where each new part has a decreasing impact on toolkit richness due to material and economic limits. Here we present the dataset.
https://doi.org/10.5061/dryad.jwstqjqm8
Description of the data and file structure
We compiled data on toolkits from previously published studies that examined the technology used by 127 small-scale societies in terms of the number of distinct tools and tool parts in their subsistence toolkits. Our coverage is global, but uneven, which we control for statistically. Two of the studies recorded toolkit data from 35 hunter-gatherer groups and 45 small-scale food-producing groups, including populations from Africa, Asia, the Americas, and Oceania. One of the studies includes 21 populations from the Northwest Coast region of North America, and the other regional study includes 17 Australian Aboriginal groups. In the few cases of overlap in these datasets when the same group was recorded in different studies, we used the data from the most recent source. All data used in this paper are provided in the Supplementary Materials together with the R code.
We chose these four datasets specifically as they are consistent in their use of Oswalt’s method of counting tools and tool parts. Oswalt devised two measures of toolkit structure, the number of subsistents and technounits, to make replicable, quantitative comparisons of toolkits used by different groups. We refer to subsistents in this study simply as tools. Examples of subsistants or tools include a wooden club, spear, bow, fish lure, pitfall, and snare. Oswalt defined a technounit as an ``integrated, physically distinct, and unique structural configuration that contributes to the form of a finished artifact'' \cite{Oswalt1976}. In other words, technounits are the unique parts of a tool. For example, a single tool, such as a spear, can be made of three parts—a shaft, a barbed point, and a binder—or if the spear also has an attachment line and float, it will have five parts.
Column headings in the tool_data.xlsx are as follows:
- ID# is the identification number
- Groups is the group name
- Lifestyle is the categorical description of the society: Mobile forager, Sedentary forager, and Subsistence farmer.
- Lat is latitude in degrees
- Long is longitude in degrees
- Types is the number of unique tool types in the toolkit
- Parts is the number of unique tool parts in the toolkit
- Complex is the number of parts per type, i.e., Types/Parts
Note that the longitude and latitude coordinates given are the centroids of historical geographic ranges and are not the physical locations of any population settlements.