Access
This article is part of Nature's premium content.
Published online 21 May 2008 | 453, 446-448 (2008) | doi:10.1038/453446a
News Feature
Language: The language barrier
Some researchers think that the evolution of languages can be understood by treating them like genomes - but many linguists don't want to hear about it. Emma Marris reports.
Consider a word as it tumbles through history: khun, a Nepali word for blood. In the early twentieth century, the word fell all too often from the lips of the Gurkhas, a Nepalese brigade in the British Army, in songs describing the horror of the First World War.
To read this story in full you will need to login or make a payment (see right).
Comments
Reader comments are usually moderated after posting. If you find something offensive or inappropriate, you can speed this process by clicking 'Report this comment' (or, if that doesn't work for you, email redesign@nature.com). For more controversial topics, we reserve the right to moderate before comments are published.
As an evolutionary biologist dealing everyday with the subleties of language (I get my money from conservation policy and action), I find myself somewhere in no-man's land. Language evolution is basically and intrinsically Lamarckian -the will to say something in a certain way is what makes the identity of a speaker. Not that environmental pressures are negligible, but linguistics can only provide intriguing analogies with natural selection. The point is, I would say, whether any insight into these analogies can be truly enlightening. Linguists should become more acquainted with phylogenetics, and tree-builders would profit from perusing, for example, the Ethnologue compilation of languages of the world.
When words (genes?) change overtime it is because context (genomic-linguistic?) changes. It is hard to imagine words changing arbitralily without the contextual environment being changed. In this respect the claim of natural selection (for single words or single genes) is a tautology at best. besides, natural selection is an outcome not a mechanism as claimed. The mechanisms reside in genome changes over time, re-shuffling and re-organization driven by several different mechanisms, leading to heterogeneity and in the long term to different species. On the other hand, evolution of languages is driven by different mechanisms altogether. Hence, the analogy has its limits.
I will agree that the analogy has its limits, but I think that it none-the-less is a very powerful one. Sounds are like amino acids, words are like protiens, sentences are like biological processes, syntaxes are like the protien-handling protocols in cells. Just as the genes of the organisms in a population need be be almost stable in the short term to ensure that the traits of the species are properly transmitted from one generation to the next, so too must language be almost stable in the short term to ensure that cultural and technological data in properly transmitted from one generation to the next. Innovations can lead to new genes/words, while functions that become unused can lead to genes/words being lost. In addition, genetic and linguisting drift can occur over multiple generations within a population. Care must be used to ensure that realistic assumptions and sub-analogies are being made when trying to treat language families like biological clades, but overall I think that this avenue of research is very promising.
Genetics - and largely genomics, proteomics, metabolomics, lipidomics- are, as is linguistics, knowledge systems (and each induces the appearance of metasystems). As they address intellect, they must share traits making them understandable. Linguistics is the oldest among them, so that what it does to make languages understandable as structures has high chances to be found in other systems. Namely, I find that “morphologically†an exon could be considered a noun, SNP variations as phonetic peculiarities of dialects, transciptase as an auxiliary verb; genetic regulatory networks as simple phrases. I find also that “syntacticallyâ€, activators, chromatin-modifying enzymes, and basal transcription factors function like reflexive verbs; exons as subjects or direct objects. The codons seem equivalent to letters (phonetically) and nucleic bases + sugars (ribose, desoxyribose) to frequencies in their acoustic spectrum. More phonetics, vocabulary aren’t yet here. ETC. A letter to the editor is not the place for a treatise on parallels between linguistics (and stylistics, theory of literature) and “-omicsâ€. Starting it probably suffices; many will do the rest (huge). Horia Georgescu, MD, PhD, DEEA 78100 France