The uses and abuses of tree thinking in cultural evolution

The uses and abuses of tree thinking in cultural evolution. Cara L. Evans, Simon J. Greenhill, Joseph Watts, Johann-Mattis List, Carlos A. Botero, Russell D. Gray and Kathryn R. Kirby. Philosophical Transactions of the Royal Society B: Biological Sciences, July 5 2021, Volume 376Issue 1828, online May 17 2021, https://doi.org/10.1098/rstb.2020.0056

Abstract: Modern phylogenetic methods are increasingly being used to address questions about macro-level patterns in cultural evolution. These methods can illuminate the unobservable histories of cultural traits and identify the evolutionary drivers of trait change over time, but their application is not without pitfalls. Here, we outline the current scope of research in cultural tree thinking, highlighting a toolkit of best practices to navigate and avoid the pitfalls and ‘abuses' associated with their application. We emphasize two principles that support the appropriate application of phylogenetic methodologies in cross-cultural research: researchers should (1) draw on multiple lines of evidence when deciding if and which types of phylogenetic methods and models are suitable for their cross-cultural data, and (2) carefully consider how different cultural traits might have different evolutionary histories across space and time. When used appropriately phylogenetic methods can provide powerful insights into the processes of evolutionary change that have shaped the broad patterns of human history.

1. Introduction

Theories of cultural evolution are built on the observation that cultural features undergo innovation, modification and transmission. Over time, these processes have generated remarkable variation in human cultures. Humans speak around 7000 distinct languages, affiliate with hundreds of religions, employ a range of kinship systems, engage in an array of subsistence practices and adhere to a bewildering number of social conventions [1]. Phylogenetic methods provide a powerful approach to studying macro-evolutionary patterns of innovation, modification and transmission [2–4]. Their application to human culture has helped reinvigorate cross-cultural comparative research but has also been subject to criticism—both valid and misguided.

Phylogenies, also known as evolutionary trees, represent the common ancestry of populations and the splitting events that have occurred over the course of their history. Phylogenetic methods encompass a broad family of mathematical approaches that can be used to construct, analyse and incorporate phylogenies (figure 1). Originally developed to study the evolution of biological organisms, these methods offer a general toolkit with the potential to provide answers to a range of cultural evolutionary questions.

Figure 1. Phylogenetic methods that can be used to study cultural macro-evolution. Black arrows indicate that the preceding methodological steps are directly incorporated in later methods: (a) tree construction [5] is required for all subsequent steps; (b) testing for phylogenetic signal (e.g. [6–8]) forms an integral part of phylogenetic regression (e.g. [9–11]), which in turn forms the basis of phylogenetic path analysis which can identify causal relationships; (c) ancestral state reconstruction (e.g. [12]), estimated in conjunction with rates of trait change and transformation (e.g. [13,14]), is required for models of trait correlation [15–17] and diversification ([18,19]; but see [20]). Red arrows indicate that suitable tests of phylogenetic signal (i.e. that the trait data fit sufficiently to the history inferred by the tree) should be conducted by the researcher before using methods detailed in (c); (see also §2). Shading: grey shading indicates methods that both assume and require inferred historical relationships between the cultural units (tree taxa) to sufficiently reflect the history of the trait; green shading denotes methods that detect and quantify tree-like structure in cross-cultural data; blue shading denotes methods that detect and control for tree-like data structure among societies, but do not require it.

An important distinction in cultural phylogenetics research is between methods of building trees (i.e. reconstructing the histories of cultural units based on assumptions of vertical transmission of cultural features (traits); figure 1a) and methods that use previously constructed trees in models that investigate the evolution and distribution of other cultural traits (figure 1b-c). A further important division in tree thinking occurs between those methods and questions that simply detect and control for tree-like structure when examining variation in cross-cultural data (e.g. What does the distribution of traits among societies tell us about the history of those societies and/or traits? Does horizontal or vertical transmission better explain the observed distribution of traits?figure 1b), and those methods that require that the modelled data are tree-like (i.e. methods that ask: What was the ancestral form of a cultural feature?figure 1c).

Phylogenetic methods offer exciting possibilities for a wide range of questions, only some of which explicitly require tree-like data. For data that are sufficiently tree-like, one of the strongest appeals of phylogenetic methods is that they offer the possibility to illuminate the unobservable past. Phylogenetic methods can reconstruct the ancestry of a vertically transmitted trait from the evolutionary signatures detected in its present-day distribution, even when archaeological records are entirely unavailable. However, despite this exciting potential, debate continues over how best to integrate cultural heterogeneity, disentangle the signatures of vertical transmission, horizontal diffusion and local socio-ecological drivers, and demonstrate that a cultural trait exhibits enough tree-like structure to justify using methods that reconstruct its evolutionary past.

Here, we review the application of phylogenetic methods in cross-cultural research. We focus specifically on the questions researchers should ask in order to avoid common methodological pitfalls when (i) deciding about the units of the underlying cultural data, (ii) constructing trees and (iii) assuming tree-like transmission of other cultural features. Throughout, we outline a series of best practices and highlight emerging methods that promise to advance our understanding of macro-evolutionary patterns of mechanism and causation in culture.