Re: Tong-cho-la, a philosophical language

On Wed, Apr 16, 2003 at 12:30:45AM -0700, Joe Fatula wrote:
> As promised, here is a little sample of Tong-cho-la, which is (sort of) a
> philosophical language.
[snip]
> A word like "automobile" makes a lot of sense, seeing that it is capable of
> motion on its own.  Everything here would still apply, but we would put it
> in the "ta" category, indicating that it is mobile and animate on its own.
> But if there were some confusion between a car and some strange alien
> creature, newly discovered, that happened to possess wheels, we could add
> "metal" or "oil-eating" or perhaps "human-controlled" to the car's
> description.  But once everyone in the conversation knew that we meant a
> car, simply "ta" might be enough to refer to it.
I *like* this idea. Suddenly, I realize that my objection to taxonomic
languages is because once you've uttered a word the first time, subsequent
utterances of it seem like a restatement of the obvious. Our brains (or at
least mine) tend to mask on repetitive things that carry no new
information; that's why two very similar-sounding words are easily
confused -- the brain ignores what it incorrectly assumes as redundant
information, and so has trouble catching that one little difference.

Your approach is an elegant solution to this: specify the taxonomy once,
and then use what is effectively a more memorable shorthand for it
thereafter. Reminds me a lot of the "counters" (or whatever they are
called) in the Chinese languages. With this approach, you could even have
a easily-learned taxonomic language.

> Remember that human speech does not make the same distinctions all the
> time.  If I were talking to a farmer, we might refer to Holsteins and
> Jerseys, and we'd both know exactly what we were talking about.  But to
> someone else, I might have to specify that I was talking about
> Holstein-type dairy cows.  Why all the extra specification?  So that the
> listener knows what exactly we're talking about.  In the conversation
> with the farmer, it would be redundant information, so we just talk
> about the two types of cattle with the specifiers that distinguish them.
Right on. This is very insightful of you, Joe. Now you're giving me a
plausible way of constructing a taxonomic language that may actually work
IRL:

Suppose A,B,C,D,X,Y are taxonomic roots, then if we have two words ABXCD
and ABYCD, normally it would be very easy to confuse them, because
everytime you said them, the listeners have to listen carefully to every
syllable. However, if we introduce a shorthand specifier Z, we can use ZX
to refer to ABXCD and ZY to refer to ABYCD. It is effectively behaving
like a pronoun of some sort.

Then if ABXCE comes up in the conversation, we'd use ZXD to refer to ABXCD
and ZXE to refer to ABXCE. Basically, the Z constructions pick up the
*differences* between similar-sounding words and bring it to the fore
front of attention. This way, we eliminate the comprehension fatigue that
comes with a multitude of very similar-sounding words occurring near each
other.

> Or consider place names... I could tell my friend to meet me at the corner
> of 1st and Market, and he and I would meet at the same place.  But if I
> called up my cousin, I'd have to specify the city.  And if I talked to my
> friend in Mexico, I'd have to specify which region the city was in, as there
> are many by that name.  But if I said to my friend around here, "Meet me at
> the corner of 1st and Market in San Jose, California, in the US," he'd look
> at me kinda funny.  Or if I said, "Meet me at three in the afternoon on
> Tuesday, April 15th, 2003," I'd get the same reaction.  Why not just say,
> "Meet me at three this afternoon,"?  And that's exactly the way people do
> it.
[snip]

Exactly. So you *can* have a naturalistic taxonomic language; you just
have to provide a mechanism by which things which are clear from context
may be freely elided.

> By the way, Tong-cho-la is also supposed to be easy for most people in the
> world to pronounce.  It doesn't make any voiced/unvoiced distinctions, no r
> vs. l, only a small set of final consonants, five cardinal vowels with the
> outermost three making most distinctions, etc.
It looks very Chinese to me. European speakers may find it awkward. :-)

Nevertheless, I think you've hit at least one nail right on the head with
the elision mechanism you describe above.


T

--
A programmer is a device for turning computer programs into spaghetti. A
*good* programmer is a device for turning spaghetti into computer programs.

Replies