Literature

This question has led to some fascinating discussions within the linguistics community and we are indebted to Emma Marris for her recent report which flags up some areas of fundamental disagreement. Representing the mainstream profession are the "historical linguists".

"Historical linguists have been reconstructing languages since the 1780s. Their tool is called the comparative method and it relies on extensive knowledge of the language group at hand, along with a broad grasp of, and intuitive feel for, the ways in which languages change. A linguist might notice that the way a vowel is spoken has shifted in two languages when compared with an ancient one, and infer that the shift happened before the two languages split. This will help to place the split relative to other splits but gives no information about when it happened. Hence the comparative method produces trees, but no dates."

The newcomers are evolutionary biologists working on language histories, using the tools and techniques of molecular phylogenetics. Their approach is to draw an analogy between language evolution and biological evolution: "Like biological species, languages slowly change and sometimes split over time."

"The advent of molecular genetics provided a new depth to the analogy. Just as the four nucleotides of DNA can produce a staggering variety of creatures, the alphabets of the world's languages can generate an infinite number of sentences. These alphabets, the words they make, and the sounds and grammar rules that frame them are passed down from parent to child in a process that, at least superficially, resembles the inheritance of DNA.
Even some complications are the same. Just as species can shade off into a maddening continuum of subspecies, populations and hybrids, languages dissolve into an untidy collection of dialects and intermediate forms. And the rampant borrowing of words between languages resembles, graphically at least, the promiscuous horizontal gene transfer that microbes engage in."

The report portrays the tensions in terms of physics envy, with the old school struggling with the statistics and the new boys bringing quantification to the discipline for the first time.

"It is putting it mildly to say that many historical linguists find the evolutionary biologists working on language histories to be bungling interlopers who have no idea how to handle linguistic data. It is also an understatement to say that some of these interlopers feel that their critics are hidebound traditionalists working on a hopelessly unverifiable system of hunches, received wisdom and personal taste. And that's just the mood between the historical linguists and the newcomers."

The problem for the historical linguists is that the evolutionary modelers appear to have no real feel for the languages they are analyzing. They know that all languages have a deep structure: words, grammar and syntax. Languages are used for communication between intelligent agents about all sorts of things: past, present and future events, concepts and abstract ideas.

"Ultimately, many linguists felt that this type of analysis oversimplified their cherished subject more than they could bear. Linguists love the little details that give a language personality: to them, the identifying sounds or peculiar borrowed words are nuances that tell the tale of a tongue. The new breed brushes over these details in pursuit of generalities, trends and statistical rules."

The problem is that languages are fully teleological, whereas the tools of molecular phylogeny do not acknowledge teleology in genomes. Those tools really grate on those who are experienced in linguistics.

"Why get excited about it when it is still so preliminary?" says Johanna Nichols, a historical linguist at the University of California, Berkeley. "We are not impressed by a computational or mathematical paper per se. We have to see that it blends well with what is known by historical linguistics and really adds to our knowledge. Then we will be excited."

There are many parallels here between this controversy in linguistics and the situation in biology. Not all biologists go along with molecular phylogenetics, as is evident here. These biologists are a diverse group, but they include ID scientists. They point out that if genomes are designed, then it is not a good idea to study them with tools that start with the presumption that there is no design or teleology in nature.

The language barrier
Emma Marris
Nature 453, (21 May 2008) | doi:10.1038/453446a

Extract: A new approach
In the past five to ten years, more and more non-linguists such as Pagel have used the computational tools with which they model evolution to take a crack at languages. And one can see why. Like biological species, languages slowly change and sometimes split over time. Darwin's Galapagos finches evolved either large beaks or small; Latin amor became French amour and Italian amore. Darwin himself noted the 'curious parallel' between the evolution of languages and species in The Descent of Man, and Selection in Relation to Sex.