Natural Language (Score 2)

In all natural languages, syntax is so difficult and complex that despite hundreds of years of effort, we are still incapable of creating even an adequate rigorous guideline for a single language. Yet your average computer language can be described completely, rigorously, and absolutely in somewhere between a few dozen and few hundred pages of text. This is because computer languages need to be utterly unambiguous- a sharp contrast to human languages, which are riddled with ambiguity and imprecision. This need for precision has led us to create computer languages that contain nothing but the barest minimum of syntax. Even a syntactically rich language like Perl has a fairly limited set of rules and an extremely small lexicon. For this reason, computer languages appear fairly natural-language independent. Instead, they encode the most basic logical rules needed to encapsulate a series of orders- rules that appear to be fairly universal. In some sense, a computer language appears to be what you'd get if you boiled down a natural language to it's absolute bare bones. In that sense, it's independent from any one language and instead incorporates the most basic element of all of them. But the richness- the things that make natural languages so interesting, and of course so different- that doesn't compute. A more interesting question might be to discover if anyone has in fact "translated" the lexicon of an English computer language into another natural language. I imagine this would be a fairly trivial task- a language like C has perhaps a few dozen keywords and no more than a few hundred functions in the basic libraries. The apparent simplicity of this kind of semantic transformation underscores the basic nature of computer languages. Similar tasks between natural languages are extremely complex, even for human beings.