Print

Print


I'd like to make some points about "chance frequency": 

>From: Constructed Languages List [mailto:[log in to unmask]] On Behalf Of Jeffrey Brown
>Sent: Wednesday, July 15, 2015 1:25 PM
>To: [log in to unmask]
>Subject: Re: Alien language research

>What I object to are absolute statements. (never, always, all, none). What if we agree to adopt Greenberg's terminology from his original "Universals of Grammar" article:

>(1) "With overwhelmingly more than chance frequency, rocks sink in water."
>(2) "The constituent order of object-subject-verb occurs with overwhelmingly less than chance frequency."

What is chance frequency? Do we just use 50%? That would mean something like, things which happen around half the time require no explanation; things which happen overwhelmingly more or overwhelmingly less than half the time need one. Yet if about half the people I meet are men, about half women, that certainly seems to call for a biological explanation (fortunately, we have one).

It seems like what's being hinted at here is more like statistical significance. But significance is a poor tool too; a result can be highly significant but actually differ from chance a miniscule amount. Still, you at least have to pose a "null" hypothesis to define what "chance" would be, which is a step in the right direction.

>To me, it seems unlikely that such deviations from chance frequency can be accounted for only by historical and cultural factors. I believe that the deviations show that there must be some sort of neurological factors at work. Whether the syntax we perceive is "hard wired" or whether it is emergent from "lower level" factors that are neurological is not the issue--
>*something* is "hard wired." The question is, what is it?

Your belief that it's unlikely has to be based (loosely speaking - I'm not saying everyone has to do the math) on a model which gives a probability. For example we can claim that historical and cultural factors are, above all else, highly mutable - cultures take time to change, sure, but given time and a space of possibilities to explore, they will produce something very much like an even spread over that space. That kind of a model might get us what we want here...

The problem is, though, that an "even" spread over a space is highly subjective. Essentially, to describe such an even sort of chaos is equivalent to choosing a language to describe languages with. One person might start with " S ‒> aS" and add rules, as Patrik Austin suggested. Another person might make up something more specifically attuned to the job. When two languages are fully capable of describing the same things, where they differ will be in how long or short the description turns out to be; and one language cannot be better than the other, it will just differ in which things it describes more succinctly.

For example, someone might specifically use the claimed universal features A through Y in their description of languages - "This language's grammar can count past two" etc. Anyone who does this, and agrees culture is essentially random or highly mutable, will conclude that some additional constraint is at work.

Someone whose idea of a "random" language - that is, their language for describing languages - matches well with observed languages, will decide that the drifting changeability of culture is all the explanation needed.

So, I've inadvertently described Bayesianism (well, a Bayesian would go on to describe how to update the description language when it doesn't match reality) but my point was just that focusing on those specific questions is what makes it look like more explanation is needed. And keep in mind, those features were specifically designed as things no human language does. We can't expect "chance frequency" when the terms we're working with were specifically selected for deviating from it. Anything can be proven that way!

>And this is indeed a relevant issue for those who construct languages. An auxlanger would be wise to avoid those syntactical features that occur with "overwhelmingly less than chance frequency" lest the reason for their low occurrence has a neurological basis. And for conlangers, especially those who create "experimental languages," it may be fascinating to see what it is like to try to communicate in a conlang that breaks the "rules."

>Jeffrey