Prehistoric Languages and Prehistoric Minds, Part 1
New research provides a deeper understanding of our remote history.
Posted Nov 20, 2016
Most people are interested in language change. Some of us may be interested in the origins of a particular word. For example, we may want to know why we use pig for the living creature, but instead use pork for its flesh, whereas Spaniards use cerdo for everything (this is related to the Norman Conquest of England, in the early Middle Ages). Or we may be intrigued about the fact that German and English are not very similar in spite of being regarded as “Germanic languages,” to the extent that we cannot easily understand “English” texts from the X century, whereas Spaniards can easily read “Spanish” from the same period (again, this is related to the Norman Conquest). Very probably, most people have heard about the myth of the Tower of Babel, according to which human beings spoke the same language until God punished them, confounding their speech so that different languages emerged.
The diversity of language have been the subject of deep inquiry over the last decades. According to mainstream view in linguistics, this variation is constrained by the architecture of the human brain. Interestingly, differences between languages, even if limited, result from external influences. Languages lexicalize and grammaticalize aspects of the environment (physical or cultural) that are important for the people speaking them. For instance, when you live in a hierarchical society your language usually has many honorifics (that is, polite forms of the addressee). If you live in a technologically advanced country, you probably speak a language with a dozen terms for colors.
We know much about prehistoric societies, but this knowledge derives mostly from archaeological remains. It is clear that knowing the shape of the languages these people spoke would enable us to refine our view of prehistoric cultures and our remote history. A famous example concerns Indoeuropean languages. One important piece of evidence supporting the view that the first Indoeuropean peoples came from the Russian steppe, and not from Northern Poland, is the fact that linguists have been unable to reconstruct a Proto-Indoeuropean word for beech, a tree that does not grow in Southern Ukraine. The word was coined later in some daughter languages.
Importantly, we know pretty less about the prehistoric mind, essentially because thought does not fossilize (language either!) and we need to infer mental processes from the archaeological record (forms of behaviour, strategies for tool assembly, etc.). Nonetheless, language is intimately related to thought. Although thought cannot be conflated with language because some kinds of thought are non-linguistic by nature, important ways of thinking, like inductive reasoning, depend on (and are made possible only because of) certain linguistic structures, like conditional sentences (if A then B).
Linguists have discussed for decades whether the language you speak affects the way you think (the Sapir-Whorf hypothesis), or whether we all think similarly and use different languages to convey our thoughts (the Fodorian hypothesis on the Language of Thought). I will write about this interesting topic in a future post, but today I wish focus on prehistoric languages. Specifically, I am intrigued about the possibility that knowing more about the features of prehistoric languages may help better understand the features of the prehistoric mind that learned and used them. The famous linguist Noam Chomsky once admitted to being intrigued by “the possibility that by studying language we may discover abstract principles that govern its structure and use, principles that are universal by biological necessity and not mere historical accident, that derive from mental characteristics of the species.” Well, the same can be said about our remote past.
What were prehistoric languages like? And how can we know that they were like that? These are tough questions. Two centuries ago many scholars believed that ancient languages had been more complex than present-day languages. In the prologue to their famous dictionary of the German language, Jakob and Wilhelm Grimm wrote: “The further back in time you can climb, the more beautiful and more perfect he finds the form of language, [while] the closer he comes to its present form, the more painful it is to him to find the power and adroitness of language in decline and decay.” Some decades later, many people thought instead that prehistoric languages had been simpler than present-day languages. In the 20th century, most linguists agreed that prehistoric languages probably did not differ from present-day human languages because they were the product of a similar brain: the brain of anatomically modern humans.
Traditionally, ancient languages of which we have no written evidence have been explored using a process of comparative reconstruction. By applying the general principles of language change to pairs of words with similar sounds and the same meanings in related languages (called cognates), we can infer how these words would have sounded some hundreds or even thousands of years ago. But the method is not perfect. Because language change is not constant nor directional, and because languages can borrow any type of linguistic items from other languages, related or not (from sounds, to morphemes, to syntactic structures), confident reconstructions are not available for languages spoken 8,000-10,000 years before present. What can we say then about the remaining 190,000 years of our history?
Recent attempts to illuminate this huge period of our history involve disparate areas of research, such as ecology, sociolinguistics, theory of grammar, or deaf studies. For instance, ongoing research on the ecology of languages has revealed that language structure can be influenced by the physical environment. Accordingly, tonal languages (like Chinese or Thai) are found in tropical regions only, plausibly because of the negative effect of dry climates on the vocal cords.
In a related sense, in present-day human groups linguistic complexity has been found to correlate with aspects of social complexity. Complexity in language is tough to define, but here we are simply referring to morphological complexity (in the plain sense of having long words made of many morphemes) or syntactic complexity (in the sense of having an elaborated grammar; for instance if you mark the subject of the sentence by putting it before the verb, but also allowing it to agree with the verb and inflecting it, like in Latin). That said, simpler and more regular morphologies have been found in languages with higher percentages of non-native speakers and are spoken by communities involved in regular cross-cultural exchanges with other groups. After all, complex languages are more difficult to learn for adults, although for kids it is great to have more cues about important aspects of grammar (such as which part of the sentence is the subject). Because prehistoric languages were spoken by small communities of hunter-gatherers, they might have been more complex than many modern languages.
At the same time, we have found that an increase in intergroup communication triggers syntactic complexity. Accordingly, if you live in isolation from speakers of other languages, your native language may not exploit all the subtle grammatical intricacies that your linguistic brain allows for. After all, if a simpler language works, why should you use a more complex one? This could be the case of the renowned Pirahã language spoken in the Amazonian forest.
On a related note, grammaticalization theory has recently emerged as a robust tool for pushing linguistic reconstruction back to the past and formulating hypotheses about how languages changed in prehistory and may have gained complexity over time. Grammaticalization is the process by which linguistic elements (words, phonetic items, discourse strategies) change into grammar components or by which syntactic constructions become more grammatical in time. For instance, in English the auxiliary "will" evolved from a lexical verb meaning “to wish”. Ultimately, the whole structure of grammar can evolve from lexical items like nouns and verbs, as suggested by Bernd Heine and Tania Kuteva.
Likewise, what happens in isolated deaf communities could illustrate what happened even earlier in our history, when first anatomically (and cognitively) modern human emerged. We have been speaking (oral) languages for thousands of years. However, sign languages are very young and some of them are still at a very early stage of development, like the homesigns that have emerged in isolated communities with a high prevalence of genetic hearing loss. Although these languages are functional (they enable signers to communicate effectively), they initially lack some key design features of human languages, like duality of patterning (in other words, signs are holistic), which seem to develop in response to environmental triggers (time, input, size of the community, degree of interaction, distribution of signers, etc.) and ultimately, from iterated learning and cultural evolution.
Finally, I wish to add a cautious note about the putative effect of gene diversity on language diversity. Studies suggesting a direct link between aspects of language and certain alleles of the genes are very controversial. Nonetheless, it might be possible that certain gene alleles, provided that they bias language acquisition or processing, may affect language change more indirectly through iterated cultural transmission, and ultimately, language structure. We need to test this possibility carefully. I think this is a good moment for trying, because we already have large databases of genomic diversity (like the 1000 Genomes Project) and of linguistic diversity (like the World Atlas of Language Structures).
Overall, the evidence reviewed above suggests that some aspects of languages can be an adaptation to ecological, social, or even technological niches. This means that if we have a confident knowledge of the milieu in which human beings lived, we can speculate quite confidently about the languages they spoke.
The history of languages based on comparative reconstruction, grammaticalization theories, the ecology of languages, social dynamics, cultural evolution, and the emergent properties of complex systems is helping us gain a more focused view of the languages spoken in the remote past. Nonetheless, in all cases we are assuming that the cognitive abilities involved were the same in the past as today. But changes in our genome, brain, and behaviour are thought to have occurred even during our recent human history. And I am not referring here to changes within the hominin lineage (I will write as well about the language faculties of other hominin species), but within the modern human lineage. Should we expect that these changes involving the hardware of language also affected language change in the past and the nature of prehistoric languages? I think so. But this will be the topic of the second part of this series on prehistoric languages… and minds.
Benítez-Burraco, A. (2016). A biolinguistic approach to sign languages. In M. Marschark and P. E. Spencer (Eds.), The Oxford Handbook of Deaf Studies: Language and Language Development (pp. 247-263). Oxford: Oxford University Press
Bolender, J. (2007). Prehistoric cognition by description: a Russellian approach to the upper Paleolithic. Biol. Philos. 22, 383-399
Dixon, R. M. W. (1997) The Rise and Fall of Languages. Cambridge: Cambridge University Press.
Heine, B. and Kuteva, T. (2010). The genesis of Grammar. Oxford: Oxford University Press
Lupyan, G. and Dale, R. (2010). Language structure is partly determined by social structure. PLoS One 5, e8559.
Lupyan, G. and Dale, R. (2016). Why are there different languages? The role of adaptation in linguistic diversity. Trends Cogn. Sci. 20: 649-660
Wray, A. and Grace, G. W. (2007). The consequences of talking to strangers: Evolutionary corollaries of socio-cultural influences on linguistic form. Lingua 117, 543-578
Be sure to read the following responses to this post by our bloggers: