Is a historic (phylogenetic) tree a fair representation of the history of languages?
In a recent linguistics tutorial, we were talking about the formation of Pidgins and Creoles, languages that develop where different languages are forced to interact. The TA used a stereotypical story to try and illustrate how these languages develop, and I’ll recount it now as it helps to explain things.
++The story of Pidgins and Creoles++
Imagine an island where well to-do foreigners show up, set up banana plantations, and get the native locals to pick bananas for them. Neither people speaks the others’ language, and neither is particularly interested in learning the others’ language. The adults just want to conduct buisness as quickly as possible, and so develop a basic code for communicating.
The ‘Language’ that develops generally has little to no grammar. The words are usually taken from one language and then strung together in an arbitrary fashion. For example: “Three bushel banana, ten dollar.” Indicating a willingness to pay ten dollars for three bushels. The important thing here is that this is just a series of important words without grammar, they generally don’t get pluralized or otherwise modified. This is a Pidgin.
Later, the children born into this environment are brought up with most of the communication going on around them in this pidgin. Which means that they aren’t exposed to any grammar. The amazing thing is that without any exposure to a developed grammar they still create one of their own, using the words available but creating grammatical structures from scratch. This has been used as very strong evidence of the fact that every new generation has the potential to create a new language if there isn’t available data.
++Connected to what we’re talking about++
So what if languages are created from scratch? The concern discussed though was that, if languages are created from scratch each generation, how can we say that one language has decended directly from another? One can certainly claim that the majority of a language is similar to another one, but it seems like it would be more accurate to say that language X is 80% language Y, and 20% languages W and Z.
Unfortunately for anyone looking for parallels between linguistic and genetic evolution, this doesn’t seem like genetic evolution. Genetic evolution has a bunch of information passed from one individual to it’s offspring. Language is grown based on the available data, each generation.
So is this just a superficial difference? or perhaps it’s worth keeping in mind? I’ll try to remember to bring this up when we get around to linguistic evolution, and I’ll see if more information comes up in linguistics.