Non-spurious correlations between genetic and linguistic diversities in the context of human evolution
Abstract
This thesis concerns human diversity, arguing that it represents not just some form of noise,
which must be filtered out in order to reach a deeper explanatory level, but the engine of
human and language evolution, metaphorically put, the best gift Nature has made to us. This
diversity must be understood in the context of (and must shape) human evolution, of which
the Recent Out-of-Africa with Replacement model (ROA) is currently regarded, especially
outside palaeoanthropology, as a true theory. It is argued, using data from
palaeoanthropology, human population genetics, ancient DNA studies and primatology, that
this model must be, at least, amended, and most probably, rejected, and its alternatives must
be based on the concept of reticulation.
The relationships between the genetic and linguistic diversities is complex, including interindividual
genetic and behavioural differences (behaviour genetics) and inter-population
differences due to common demographic, geographic and historic factors (spurious
correlations), used to study (pre)historical processes. It is proposed that there also exist nonspurious
correlations between genetic and linguistic diversities, due to genetic variants which
can bias the process of language change, so that the probabilities of alternative linguistic
states are altered. The particular hypothesis (formulated with Prof. D. R. Ladd) of a causal
relationship between two human genes and one linguistic typological feature is supported by
the statistical analysis of a vast database of 983 genetic variants and 26 linguistic features in
49 Old World populations, controlling for geography and known linguistic history.
The general theory of non-spurious correlations between genetic and linguistic diversities is
developed and its consequences and predictions analyzed. It will very probably profoundly
impact our understanding of human diversity and will offer a firm footing for theories of
language evolution and change. More specifically, through such a mechanism, gradual,
accretionary models of language evolution are a natural consequence of post-ROA human
evolutionary models.
The unravellings of causal effects of inter-population genetic differences on linguistic states,
mediated by complex processes of cultural evolution (biased iterated learning), will represent
a major advance in our understanding of the relationship between cultural and genetic
diversities, and will allow a better appreciation of this most fundamental and supremely
valuable characteristic of humanity - its intrinsic diversity.