Is Python the Esperanto of programming languages?

Carl Banks imbosol-1048251251 at aerojockey.com
Fri Mar 21 15:30:52 CET 2003


Erik Max Francis wrote:
> Carl Banks wrote:
> 
>> I say redundancy is not required at all for a language, and most
>> languages have very little of it.  Sometimes what looks like
>> redundancy isn't really redundancy, but rather superfluousness.
> 
> That sounds like a distinction without a difference.  It isn't
> redundancy because you choose not to call it redundancy, you'd rather
> call it superfluousness.

When I say redundant, it means there are two things saying the same
thing.  When I say superfluous, I mean there is one thing saying
something and one thing saying nothing.

I say, when a language requires agreement, usually one of the things
being agreed (linguists call them morphemes) has a lot more force than
the other one.  This makes the other one useless, more or less.
Sometimes, more rarely, I'd say, the morphemes have about equal force,
in which case there is redundancy.


>> For
>> example, in English, if I say, "The man go to the store," no native
>> listener is going to wonder what I meant.  Leaving the final sibilant
>> off the word "go" sounds bad, but it doesn't affect the meaning of the
>> sentence.  It's completely superfluous.
> 
> Actually, it isn't.  "The man go to the store" is unclear precisely
> because it's incorrect English.  Does he mean "goes," or "went," or
> "will go," or "would go," or "must go"?

I say it's not unclear.  A listener who hears "The man go" understands
it to mean "The man goes" and nothing else.  I say a listener will NOT
take it to mean any of the other possibilities you give up there,
unless the conversation allows him time to think about why the
sentence sounded bad.  (In other words, our language processing unit
flags this as sounding bad but meaning clear; our intellect asks
whether the speaker might have meant something else.)

The -s ending on verbs is devoid of meaning in English.  It's
superfluous; it doesn't add redundancy.

I can give you more examples.  Suppose a Spanish speaker says "Buenas
dias."  I'm not a native Spanish speaker, but I highly doubt any
Spanish speaker will understand this any other way than a bad-sounding
"Buenos dias."  The ending used with the adjective "bueno" is
completely superfluous; it has no meaning; changing the ending will
make an sentence sound bad, but not affect its meaning.

French verb endings agreeing in number and person.  I'm not quite as
sure about this one.  The subject pronouns are required in formal
speech, rendering the endings rather useless.  I imagine a native
French listener will understand "vous avons" to mean (a very bad
sounding) "you are," but it's likely it will be uncertain enough that
the listener will ask whether the speaker meant "vous avez" or "nous
avons."  (Especially since "nous" and "vous" sould alike.)  Maybe a
native French speaker can share insight on how a listener would react
to subject-verb disagreement.

Obviously, the difference between redundant and superfluous is not
clear-cut; thing vary from one extreme to the other; I would guess
French subject-verb agreement is in the middle.


>> Sometimes it is redundancy, though.  For example, someone hearing the
>> phrase "this heads" might wonder if the speaker meant one or many
>> heads.
> 
> Same thing is going on here.  Lack of proper conjugation (in a language
> which requires it) inherently involves the introduction of ambiguity.

Agreement does not make things inherently ambiguous.  "Buenas dias"
isn't ambiguous at all, even though it doesn't agree.  It is only when
the two morphemes have roughly equal force that ambiguity arises.  In
this case, the word "this" indicates singular about as strongly as the
ending "-s" indicates a plural.  That was not the case in the above
examples.


-- 
CARL BANKS




More information about the Python-list mailing list