Rate of language evolution is affected by population size

Abstract

The effect of population size on patterns and rates of language evolution is controversial. Do languages with larger speaker populations change faster due to a greater capacity for innovation, or do smaller populations change faster due to more efficient diffusion of innovations? Do smaller populations suffer greater loss of language elements through founder effects or drift, or do languages with more speakers lose features due to a process of simplification? Revealing the influence of population size on the tempo and mode of language evolution not only will clarify underlying mechanisms of language change but also has practical implications for the way that language data are used to reconstruct the history of human cultures. Here, we provide, to our knowledge, the first empirical, statistically robust test of the influence of population size on rates of language evolution, controlling for the evolutionary history of the populations and formally comparing the fit of different models of language evolution. We compare rates of gain and loss of cognate words for basic vocabulary in Polynesian languages, an ideal test case with a well-defined history. We demonstrate that larger populations have higher rates of gain of new words whereas smaller populations have higher rates of word loss. These results show that demographic factors can influence rates of language evolution and that rates of gain and loss are affected differently. These findings are strikingly consistent with general predictions of evolutionary models. language evolution | sister-pair comparison | Austronesian | lexical change | Poisson regression P opulation size can play a crucial role in the evolution of languages and cultures (1). However, opinions differ on both the possible mechanisms and the expected patterns (2-7). It has been suggested that larger populations will generate more innovations and are less prone to random loss of cultural elements (8-10), but may have less efficient diffusion of innovations than smaller populations (4). Alternatively, languages spoken by small isolated populations of speakers may have lower rates of loss if they maintain tighter cultural norms that improve transmission and resist change (11). Rates of change might be accelerated by founder effects when a new population is started from a small number of colonists, which could result in loss of elements from the ancestral language (11-13). Population size might also influence language complexity if small populations can develop greater linguistic complexity (11), whereas large, widespread languages that are often learned by adults may become simplified (14). Conversely, it has been suggested that the average rate of word turnover is essentially the same in all languages (15-17), or that it is determined primarily by other factors such as language contact (6, 18).Uncovering systematic patterns of rates of language change may reveal underlying mechanisms of language evolution (13,19). In particular, investigating rates of language change can demonstrate whether langua...

Human population dynamics in Europe over the Last Glacial Maximum

“…In addition to the destiny of other species, human population size also influences our own cultural and genetic evolution. Large pools of interacting individuals can create and maintain adaptive skills, as well as phonological variation, more effectively than small populations, and they are also capable of faster cumulative cultural evolution (3)(4)(5). A decrease in population size may even result in a loss of complex cultural traits (6).…”

The relationship between population size and rates of language evolution

“…Bromham et al (2015) applied this technique to 20 Polynesian languages and found a correlation between speaker population size and rates of linguistic change. Specifically, their results conformed to the expectations of population genetics, as large populations gained words faster while small populations lost words faster.…”

“…Specifically, their results conformed to the expectations of population genetics, as large populations gained words faster while small populations lost words faster. However, Bromham et al (2015) only investigated a restricted dataset of 20 Polynesian languages, a small subgroup of the Austronesian language family (Hammarström et al 2014). They argued that Polynesian languages were an ideal study group as they are well documented, and spoken on isolated islands that were sequentially founded as humans expanded across the Pacific Ocean (Gray et al 2009).…”

“…They argued that Polynesian languages were an ideal study group as they are well documented, and spoken on isolated islands that were sequentially founded as humans expanded across the Pacific Ocean (Gray et al 2009). Their study was also limited due to its small sample size, so it is unclear if the conclusions of Bromham et al (2015) are specific to a feature of Polynesian cultures and languages, or can be generalised to all language families.…”

“…In Welsh et al (in press), the statistical robust comparative linguistic method developed by Bromham et al (2015) is applied to three major language familiesAustronesian, Indo-European and Bantu-to test the generality of the correlation between population size on patterns and rates of language evolution. Together, these families span four continents, have population sizes ranging from 100 million (Nauna, Austronesia) to 258 million (Hindi, Indo-European) and display different social structures and geographical distributions.…”