Is Python the Esperanto of programming languages?

Steven Taschuk staschuk at telusplanet.net
Fri Mar 21 20:24:16 CET 2003


Quoth Carl Banks:
  [... -s agreement marker on English verbs ...]
> Sure, there's a marginal benefit to it.  The morpheme has a tad of
> meaning left, when the number of the subject can't be determined.
> (E.g. "The sheep go to the barn.")  The ending is a little redundant,
> but mostly superfluous.

I think we agree that the information "the subject of this verb is
third person and singular" is usually but not always available
elsewhere.  (I ignore the tense-marking aspect of the morpheme.)

I see in your other recent post the claim that, in real language
processing (as opposed to after-the-fact thought), the information
from the subject takes precedence over that of the inflection of
the verb if they contradict one another (which they do if the
morpheme is omitted when agreement calls for it).  This seems
plausible, so I'll so stipulate.

I don't, however, think that these add up to the morpheme "saying
nothing".  In my personal theory of meaning (idiosemantics?), it
always means "third person singular", and its absence always means
"not third person singular"; this explains why
   1.   The man goes to the store.      is good,
   2.  *The man go to the store.        is bad, and
   3.   The sheep goes to the barn.     is not ambiguous as to the
                                        number of sheep.
It does not, on its own, explain why (2) is merely bad instead of
ambiguous to the point of incomprehensibility; for that we can use
your idea that the information from the subject dominates.

I don't see what is explained by calling the morpheme meaningless
in (1).

  [...]
> If a language could be invented that completely lacked redundancy, and
> was simple enough to be internalizable, then I don't think humans
> would have any problems communicating with it.

Ah.  I understand better now your point.

There are scenarios of language use in which noise is a major
concern; military speech over radio comes to mind.  (Both the
so-called phonetic alphabets, which are by design highly
redundant, and the use of formulaic utterances such as "say again,
all after X".)  Admittedly such speech is not representative of
typical language use, and has little if any influence on language
development in the broader community.

However, there's always *some* noise.  I think it plausible that
there's enough noise normally (not so much in transmission as in
utterance and interpretation) to require some degree of redundancy
for error detection and correction.  How much, I don't know; it
might well be less than any existing natural language actually has.

-- 
Steven Taschuk                  staschuk at telusplanet.net
"Telekinesis would be worth patenting."  -- James Gleick





More information about the Python-list mailing list