Re: (opinion) self-seggregation considered harmful

On Tuesday, July 6, 2004, at 11:35 , And Rosta wrote:

>> This is just an personal remark abou self-seggregative conlangs: except
>> computers, who cares about words being unambiguously separable? This
>> prevents puns, and for these of you whose L1 is French, you certainly
>> know how semantically productive puns can be!
>
> People who care about self-segregation would be those who value the
> potential for unambiguity over the potential for *perfect* puns.
> (Even a self-segregating language can allow for near-puns and other
> paronomasia.)
Yep - to imagine that self-segregating morphemes and unambiguity are going
to put an end to puns and other word-play underestimates IMHO the
inventiveness of the human mind.

Indeed, in languages were puns are not so easy to make as they are in
French or English, puns and word-play are admired. Puns and word-play
figure, e.g. in Medieval Latin in what we would call serious writing.

In English, on the other hand, where puns are easily made, they are
usually met with groans rather than laughter - and not generally
considered particularly clever.  The challenge in English is not to
produce a pin (a relatively easy thing to do) but to produce a pun which
will be considered clever or imaginative.

On the other hand, I suspect the creation of near-puns and other word-play
in languages like Lojban or Classical Yiklamu would be considered smart
per_se.

>  In my linguistic life I far more often find myself
> straining to formulate a locution unambiguously than straining
> to pun.
Well, yes. IME misunderstandings cause far more problems (sometimes
serious ones) than puns cause merriment.

But to simply dismiss self-segregation as something odd-balls do to keep
computers happy is to take a very blinkered view and not to be aware of
the vast variety of language types there are. How far self-segregation is
desirable depends, surely, on the language in question.

All natlangs have a good deal of in-built redundancy so that meaning can
be disambiguated with self-segregation. But once such redundancy is
reduced or eliminated (as, for example, I guess it inevitably is in
loglangs), then other mechanisms like self-segregation become needed.

Dutton's Speedwords is designed, as the name suggests, to maximize brevity.
  I can assure you from personal experience that one of the problems I
encountered when learning the language in the 1950s was figuring out where
the morpheme boundary were (this was decades before I or most of the
general public knew anything about computers!) - quite simply, it impeded
comprehension. That is why, right from the start, self-segregation has
been one of the aims of my various 'briefscript' and/or BrSc experiments.
It will improve the language for *humans*.

Ray
===============================================
http://home.freeuk.com/ray.brown
ray.brown@freeuk.com    (home)
raymond.brown@kingston-college.ac.uk   (work)
===============================================
"A mind which thinks at its own expense will always
interfere with language."         J.G. Hamann, 1760