Lambda the Ultimate

Languages tend to go in isolation for periods and then mix together in unpredictable fashions. I'm of the opinion that language can arise spontaneously, as our brain can create and manipulate language as the need arises. The opposite hypothesis I would have to Sapir-Whorf would be that people tend to influence language much more than language influences them (at least with language in a non-PL sense). (Okrent gives the example of deaf people negotiating sign language, if neither of the people can initially understand one another). Not sure I could believe that there was one language at some point in time, that suddenly got splintered (Tower of Babel).

Anyhow, as with all words, they do not necessarily convey exactly divisible concepts. The idea of invented (vs. natural) languages is touched on in an interview with Okrent:

All languages are in some sense invented in that people create them, but generally people do this as a group without being conscious of what they are doing. The invented languages I’m talking about are those where someone sat down and built a brand-new language from scratch.

Well, as a person of faith, I too have trouble interpreting the pre-history stories in Genesis literally. I think that trying to look to scripture for answers to scientific questions or as a foolproof record of history is a big mistake. I don't think the original link is credible.

That's an interesting counterpoint to the Sapir-Whorf hypothesis; one that makes sense to me. Look at the evolution of language since the dawn of the internet. The use of smileys is fairly mainstream, for example.

I think a hypothesis of a common language is interesting (it is mentioned in the above book a couple of times). But, from a scientific standpoint, an hypothesis would have to be judged on its utility and testability. And, other than being a useful moral story about how language can separate us, I'm not sure it brings us much further than noting that many languages share common root words. Of course, the explanation may be as simple as economic - people tend to share and interact with one another over seemingly great times and distances. And languages have a tendency to die out or blend with others. So the hypothesis could just as easily be that these other cultures had their own languages but found it more profitable to standardize on languages brought in by the greeks.

(Edited, since I want to avoid religious entanglements. :)

Edo Nyland is a complete nutball and you're absolutely ridiculous for making that comment.

...language design also tends to invite people with grand theories of everything. You have to have a certain level of naivete and idealism to think that something as large as language can be improved upon. But the book does make the point that these people do expend a good amount of energy pursuing and implementing their thoughts. So, even if they fail, we can usually learn from their failures.

.. but then it's almost the point of the book under discussion, people who invent conlangs are "mad dreamers".

As Larry Wall said in "man perl".."The three principal virtues of a programmer are Laziness, Impatience, and Hubris."

And what greater Hubris can there be to try reinvent the very language in which we madly dream?

Alas, Program Language Design folk are no less mad, no less dreamy.. the difference is we can at least find something, even if it is only a computer, to speak our new tongue.... and then futilely try convince the world of our Rightness.

Given how many programming languages exists and how few are in wide use... alas, LtU cannot afford to sneer at the conlangers. The hard stats are that we are madder (more languages) and dreamier (with less uptake) then they.

But oh my, it's a fascinating game... digging deep into the heart of the meaning of meaning.

John, I happen to be a programmer and a conlanger, one who frequents a forum that Edo Nyland used to hang around. The man is a nutball. Luckily he was gone long before I showed up, but you can still see his posts. He's truly insane and believes that there's this grand conspiracy of Basque monks. He's not a conlanger of any normal sort, most conlangers make conlangs for the fun and aesthetic of the act, not for some idealistic dream of blah blah blah. It's just the nutters who are the loudest and thus most visible.

I don't have much to contribute to this website in terms of theoretical knowledge of programming languages, which is why I didn't register with a real name. I have been following it because it's been informative and I like functional languages and I like to keep up with the bigger picture in general. However I wanted to inform the regular contrubutors here; ones pontificating uninformedly in responce to your comment, how that, and the very mention of him here, are damaging to the credibility of this weblog.

So: guys, you look like you know a lot about programming languages, but when it comes to linguistics... Nyland is about as sane as the Time Cube guy. And about as coherent. And he is an in-joke among enthusiasts of constructed langauges.

EDIT: relevant thread about the book.
Apparently she didn't bother informing people whose languages she included examples of, which would have been nice of her. (it's fair use)

Indeed, to construct "analog" or descriptive software we would use the symbol manipulation languages such as Lisp or Prolog. My post is motivated by the fact that this is practically a lost art. Or more correctly an art that has been identified and studied but is not used as far as I know. Description and analogy are the very essence of language. The math, logic. or PL theory that make it happen are tools and not the real purpose of this agenda.

To clarify this a little, there are fields that deal with representation. These fields do use symbol manipulation but don't seem to recognize analogy as an issue. This is to say that they view their field as a form of conventional programming. Symbols have precise meaning with a "formal" spin.

Analogy and Cybernetics view representation as descriptive in terms of a system-state abstraction. The product is semantic but not formal. The semantics of a system-state abstraction is well understood mathematically and is closely related to abstract algebra, or field theory in the continuous case.

We might summarize by saying that a formal view attempts to reduce everything to mathematics. The Cybernetic view is to describe in terms of a mathematically consistent model, a semantics.