Loans in Hunter-Gatherer Basic Vocabulary

Humans lived by hunting and gathering until the beginnings of agriculture around 10,000 years ago. Many have maintained that kind of subsistence since that time, and up to the present, in a few areas. It has been suggested that the dynamics of hunter-gatherer languages are distinctively different in a number of ways from that of farmers, for instance that farmer languages spread widely and form large language families but hunter-gatherer languages do not. Rather, they stay in place and borrow heavily from neighboring languages, to the extent that some have argued that language families cannot be detected or reconstructed using methods developed for families spoken by agriculturalists.

These hypotheses are beset with problems. In this presentation we look at one aspect of the issue: the rates and types of loanwords in hunter-gatherer languages. We draw on research by the Hunter-gatherer Language Dynamics project in regions of North America, South America and Australia where the indigenous populations were either exclusively hunter-gatherers until recently (Australia) or where there were many groups who lived exclusively or predominantly by foraging (the Americas). The comparison utilizes analysis of vocabularies from 100 languages, which were analyzed by standard comparative linguistic methods into inherited items and loans.

Loan rates in the data are low, which an average of 6.75% comparable to that of non-hunter-gatherers (5.15% in our sample). There are two exceptional small clusters: (1) very low loan numbers in the Amazonian sample languages and some North American languages; (2) high loan numbers of 40-50% in a few Australian languages.

Resistance to borrowing vocabulary has been discussed widely in relation to areas of ‘linguistic exogamy’ in the Vaupes of Amazonia but the ‘linguistic purism’ extends beyond this. Among the highest loan numbers in Australia are among Eastern Ngumpin speakers like the Gurindji. A possible explanation of the latter is a history of these languages moving into this area and adopting features from neighboring languages as they established themselves. We discuss whether it is possible to propose a sociolinguistic theory which will encompass both low and high borrowing cases. Such a framework would not be hunter-gatherer specific but factors such small population size, low density and high mobility have an impact in how the fundamental properties of the model play out among hunter-gatherers. Certainly no theory which attributes a single type of loanword profile to all hunter-gatherers will work.