Grambank reveals the importance of genealogical constraints on linguistic diversity and highlights the impact of language loss

Abstract

While global patterns of human genetic diversity are increasingly well characterized, the diversity of human languages remains less systematically described. Here we outline the Grambank database. With over 400,000 data points and 2,400 languages, Grambank is the largest comparative grammatical database available. The comprehensiveness of Grambank allows us to quantify the relative effects of genealogical inheritance and geographic proximity on the structural diversity of the world's languages, evaluate constraints on linguistic diversity, and identify the world's most unusual languages. An analysis of the consequences of language loss reveals that the reduction in diversity will be strikingly uneven across the major linguistic regions of the world. Without sustained efforts to document and revitalize endangered languages, our linguistic window into human history, cognition and culture will be seriously fragmented.

Languages with more speakers tend to be harder to (machine-)learn

“…Another promising avenue for future work would therefore be to investigate what kind of grammatical structures tend to be harder to learn for LMs and if those features co-vary with speaker population size. A recently published new cross-linguistic database of grammatical features of unprecedented size 17 provides an ideal starting point for such endeavours.…”

“…Yet, depending on how you count, there are between 6,000 and 8,000 different languages and language varieties on the planet [13][14][15] that vary greatly in their structural properties. 16,17 A growing body of cross-linguistic research has begun to document that the natural and social environments in which languages are being used and learned drive this diversity [18][19][20][21] , that language structure is influenced by socio-demographic factors such as the estimated number of speakers 18,[21][22][23] and that that the long-held belief of a principle of "invariance of language complexity" 24 may be incorrect. 25 In this article, we investigate another long-held assumption that, to the best of our knowledge, has not been systematically tested yet: the assumption that all languages are equally difficult to learn.…”

Societies of strangers do not speak grammatically simpler languages (original submission)

“…Finally, the standardized variance-covariance matrix is transformed into the precision matrix. We estimate two spatial matrices for two sets of parameters: a) ϕ=1.25 and κ=1 ("local set") and b) ϕ=17 and κ=1 ("regional set") (see (67) and (28) for other examples of how this type of control for spatial autocorrelation is implemented). These parameters of the spatial covariance matrix were chosen to differentiate between two assumptions: under parameters corresponding to the local set, the diffusion of similar metric scores between languages are not likely across distances over 1,000 km, while with the regional set parameters the diffusion can take place over several thousands of kilometers (see Figure S1 in Supplementary Materials for details).…”

“…To test the association between grammatical structures and exotericity of the relevant societies, we introduce metrics that quantify two distinct dimensions of grammatical complexity that have been claimed to be reduced in exoteric societies: (1) the degree of phonologically fused grammatical markers ("fusion"), and (2) the number of obligatory grammatical marking ("informativity"). We obtain grammatical features included in both metrics from the large global dataset Grambank (27,28) (see Table S1 in Supplementary Materials for the list of Grambank features in both metrics).…”

The importance of worldwide linguistic and cultural diversity for climate change resilience

“…These global approaches strikingly ignore the actual complexity of human societal and cultural diversity, as exemplified by the myriads of local minorities, minority cultures, vernacular languages and subtly differentiated dialects, which, taken together, constitute a significant part of humankind (Díaz et al., 2019). Such a focus on uniformity may reflect limited knowledge on the extent of human linguistic and cultural diversity (Skirgård et al., 2023), underrepresentation of cultural and linguistic diversity in the science community (Couée, 2022; Graves Jr. et al., 2022a, 2022b; Konno et al., 2020; Lynch et al., 2021; Marks et al., 2023; Smith et al., 2023), or hidden assumptions that sheer technological, socio‐economical and political efficiency should prevail over the complexities of cultural diversity (Amel et al., 2017; Isbell et al., 2017; Ripple et al., 2022). Alternately, environmental studies that take into account the diversity of human cultures (Copete et al., 2023; Díaz et al., 2019; Mistry & Berardi, 2016; Reyes‐García et al., 2019) usually emphasise specific cases of emblematic Indigenous and Tribal Peoples in highly‐specialized environments.…”