Limited evidence that fitness interdependence produces historical origin myths

Abstract

This commentary points out some theoretical lacuna in the argument and then evaluates, in a preliminary way, its main comparative empirical hypotheses. It finds very limited support for the observable implications of the evolutionary theory. By contrast, the historical remoteness of foundational myths is closely associated with how long a society has been ruled by a centralized state, pointing at the important role of political history.

This sweeping and intriguing article seeks to explain why so many human groups developed foundational myths that proclaim their origins in the deep historical past. The main argument is that such historical tales are developed by political entrepreneurs who seek to profit from the human tendency, evolved through millennia of evolutionary pressures, to equate shared history with shared interest in each other's survival (or “fitness interdependence”). Mythical narratives of shared historical origins should be more prevalent, the authors argue, (1) in larger groups – because there are fewer opportunities to observe individual fitness and thus which individuals would be advantageous to form a coalition with, (2) in groups that offer a more diverse range of possible coalition partners and thus more competition among political entrepreneurs, (3) in groups that are exposed to more intense survival pressures especially through warfare, (4) in groups that are not internally differentiated into dominant and marginalized segments, and (5) in groups that offer distinctive markers of difference that allow an easy identification of coalition partners.

There are some unresolved issues in this thought-provoking argument. The first is that it runs into the same problems that plague the theory of elite manipulation, swiftly criticized by the authors: Why would individuals believe in promises of group solidarity and the evolutionary advantages that it could potentially offer? In other words, why do individuals follow the siren songs of entrepreneurs who invent historical myths, if there is no causal link to objective and effective fitness-interdependence? Nowhere in the argument do individuals evaluate the objective fitness advantages of varying historical myths and the associated coalitions. We thus do not know why they embrace such myths at all and if they do, which ones they chose.

Second and relatedly, there are plenty of examples of historical myths that are not tales of successful co-operation and survival, but tales of victimhood, defeat, and injustice, as witnessed by Zionism or Polish nationalism. If the coalition offered by these tales has not proven, historically, to increase the survival chances of its adherents, why would individuals continue to embrace them and the corresponding group identities? Third, these examples perhaps point to the more general limits of evolutionary arguments in explaining the varieties of human experiences across the world. While the article zooms in on group size as a crucial explanatory factor for this variety, distinguishing between face-to-face societies (which don't need historical myths) and all other societies, much important and consequential variation in historical origin stories remains unexplored – between ethnic and civic versions, between imperial stories of superior civilizational origins and the familiar nationalist tales of the modern age, between genealogical charters and non-genealogical ideas of historical continuity, and so forth.

Fourth and perhaps most importantly, the article does not offer any empirical evidence for the observable implications of the argument. In what follows, I will put their five hypotheses, summarized above, to a preliminary test using available data. How do we operationalize the existence of historical myths? Since these are ubiquitous, as the authors note, in societies organized into larger than face-to-face groups, a binary coding would not make much sense. I thus interpret their argument as follows: In societies that are especially prone to produce historical myths, the origin stories should be further removed from the present. Kaufmann (2015) has coded what he described as the “foundation years,” according to historical legends, of the majority ethnic group of each contemporary country. The data were generated with the help of experts in the ethnic and national histories of peoples from around the world. Despite its many limitations, let us use these data. The earliest imagined foundation year is 750 years BC (Greece), the last 1975 (Papua New Guinea). The average is the year 1419.

The size of the population in 2005, which I will use to test H1 above, comes from Wimmer, Cederman, and Min (2009). I use data on the linguistic diversity of countries in the 1960s, as recorded by Soviet linguistics, to evaluate the effects of the range of coalitional opportunities (as according to H2, diversity offers opportunities for assimilation, for fusion and fission of related languages, etc.; other fractionalization indices produce similar results). Data on the number of wars fought between 1816 and 2005 come from Wimmer and Min (2006) and will be used to test H3. The degree to which a population is divided into dominant and marginalized segments is measured with the average population share of the ethnic groups not represented in national-level government between 1945 (or the year of independence) and 2005 (for testing H4; data again from Wimmer et al., 2009). Finally, to test H5 I use again data from Kaufmann to code in a binary way if a self-determination movement used ethnic diacritic (such as language or religion) to delineate the nation (as was the case in Germany or Israel, but not the United States or France).

I add one important control variable without which the analysis would be unconvincing for any historically oriented social scientist: The degree and duration of past statehood in a society. Obviously and as mentioned by the authors, state elites are often those who craft narratives of historical origin and continuity. How far back in history there was a state may thus very well influence how far back historical narratives locate national origins. I use Bockstette, Chanda, and Putterman's (2002) well-known state-antiquity index, which combines state age with degrees territorial control and levels of centralization into a single index.

Results of a linear regression model are shown in Table 1.

Table 1. Linear regression on the assumed origin year of national majorities

Standard errors in parentheses; *p < 0.10; **p < 0.05; ***p < 0.01.

The results provide no evidence for four of the five hypotheses. Only ethnic nationalism tends to project history back further, for about 290 years. This may very well be an endogenous relationship, however, as ethnic nationalists not only search for deep historical origins, but also look more systematically than civic nationalists for diacritic that distinguishes themselves from their neighbors. The antiquity of statehood, however, is a powerful predictor of how far back foundational histories go: One standard deviation increase in the index is associated with a 226 years older historical origin. It is therefore wise not to dismiss the specifics of political history in search for universal, evolutionary explanations.

In conclusion, I don't find much support for the hypothesized mechanisms producing a need for deep historical origin stories. And I miss a conclusive argument about why these myths should take root in the population at large, thus leaving one of the main puzzles in the study of nationalism unresolved (for my own, modernist answer, see Wimmer 2018).