One area I didn't get into (my post was wordy enough), is the notion of the unit of selection and the substrate upon which language evolution operates. I imagine that the forces of language evolution operate on individual words as well as phrases and other linguistic constructs, similar to how biological evolution can operate at the level of genes, biochemical pathways, individuals, and societies.
As for substrate, language cannot exist without minds that understand and speak it, so the evolution of language is played out on a mental substrate. Given that there is fairly strong evidence for innate neural language modules in our brains, language evolution thus likely has played a role in our biological evolution, physically shaping the development and structure of our brains.
And human language hasn't been around very long in evolutionary time, originating as recently as 40,000 years ago. So this language-based shaping of the human neurobiology and evolution is happening literaly as we speak (sorry, couldn't resist).
Language itself is a substrate for the evolution of memes. It's interesting to think about what sort of role language evolution might play in meme evolution, and the interplay in the evolution of memes, languages, minds, and societies.
- Can we identify memes that are able to persist through long-term changes in a language?
- How do memes influence language evolution and biological evolution of individuals/societies in which the meme persists?
- What properties of memes allow them to span different languages?
This speculation is all well and good, but there is some genetic evidence emerging for a link between language evolution and biological evolution. The FOXP2 gene in humans may have played a key role in the emergence of language in human beings. See these links for more on this fascinating story:
Here a Scientific American article related to this thread: A Way with Words: Do languages help mold the way we think? (25 Mar 2002).
ReplyDeleteAnother area affected by language evolution is the communication of scientific knowledge. The same forces that shape language "in the wild" are also at play in the refined world of science, and this creates problems for conveying specific meanings.
Language evolutionary forces act to constrain and confound the flow of knowledge in the scientific world. This is serious stuff. Especially as increasing amounts of scientific data are computationally accessible -- such as with the human genome project. Building robust, interoperable computer-based systems that can make use of scientific knowledge requires a precision of language that is not typically available in the way humans naturally use language.
Yet scientists like to use natural language to convey their findings. Getting a computer to handle natural language in all the ways humans can is an unsolved problem. So is there any hope?
The main hope for salvation here is the widespread adoption of common ontologies and controlled vocabularies, such as those maintained by the Open Biomedical Ontologies project.
Such project will help build a sort of periodic table of concepts for use in communicating scientific information. But this doesn't in itself get at the problem of teaching computers (and other humans) what these concepts really mean, and how to reason over such encoded data. That is a goal of the semantic web. More on this later.
Here's a recent finding on the language-perception link:
ReplyDeleteWords help deterimine what we see.
The basic finding is that we're more likely to perceive color distinctions via our left brain (or right visual field) than our right brain. This is presumably because our left brain contains the language center, and if we can label the distinction, we're more likely to notice it. It's almost as if the left brain is primed to notice the distinction, simply because it can label it.
Some interesting follow up questions:
1. Do people who process language with their right-brain show the expected effect in their left visual field? (I'm pretty sure there is a small fraction of people who have language processing on their right brain, possibly due to injury on their left side or for unknown reasons.)
2. Is there a similar affect with conceptual distinctions as well as physical distinctions? I think so. We've all had the experience of learning a new word and then suddently noticing it used in various things we read after learning the word. But what about pure concepts that can't be represented by a single word? I bet there is a similar effect, but it may not have a bias towards one hemisphere or another since processing of concepts occurs throughout the cerebral cortex, distributed on both sides of the brain (I'm not a cognitive scientist, so I'm not sure about this, but I believe it's the case).
To me, this finding is an example of how good our brains are at pattern recognition. Probably the most extreme example of this is our ability to recognize faces and to see faces-like shapes in odd places (like the surface of Mars or certain strings of characters ;-).
This pattern recognition occurs in our audio as well as video realm of perception. A good example of this is when listening to music backwards, we think we hear various messages, often demonic or otherwise of a dark nature. This happens when there was no intent on encoding any message by the artist. There's lots on the web about this. See here for example.