Newsgroups: sci.lang,soc.culture.esperanto
From: philip@storcomp.demon.co.uk (Phil Hunt)
Path: cantaloupe.srv.cs.cmu.edu!das-news2.harvard.edu!news2.near.net!news.mathworks.com!udel!gatech!swrinde!cs.utexas.edu!news.sprintlink.net!pipex!peernews.demon.co.uk!storcomp.demon.co.uk!philip
Subject: Re: One point against Esperanto
Distribution: inet
References: <794976261snz@storcomp.demon.co.uk> <DJOHNSON.95Mar14125238@tartarus.ucsd.edu>
Reply-To: philip@storcomp.demon.co.uk
X-Newsreader: Demon Internet Simple News v1.27
Lines: 215
X-Posting-Host: storcomp.demon.co.uk
Date: Thu, 16 Mar 1995 16:42:49 +0000
Message-ID: <795372169snz@storcomp.demon.co.uk>
Sender: usenet@demon.co.uk

In article <DJOHNSON.95Mar14125238@tartarus.ucsd.edu>
           djohnson@tartarus.ucsd.edu "Darin Johnson" writes:

> Well, given this task, I would hope you learn from the mistakes of the
> scores of other interlanguages.  A good reference is the "Cambridge
> Encyclopedia of Language", from the Cambridge University Press.  It
> has a whole chapter on this, and in particular, one page on features
> for the ideal artificial language. 

I've read it.

> Personally, I'd like to see someone (or a team) that
> knows at least 3 out of Latin, Japanese, Chinese, Swedish, Xhosa,
> Hindi, and Arabic attempt this. 

I know a bit of Latin. I expect Swedish is similar to other Germanic
langs, so I don't think that would help me. I have a slight knowledge
of Japanese and Chinese, and know little of the rest.

me sav suma Latin. me pense, ka sveriga lang est equaleta de alia
deutsha-groupa langs, raisonop me no sav, ka it aid me. me savet
nihona lang et Putongua, et me no sav la alions.

> Roots and grammars wouldn't have to
> be taken from all of them, but at least the designers wouldn't be
> working with a set of blinders.

I agree, if you know how lots of langs handle a particular feature, it's
easier to design a good way of doing it when creating an artificial
lang.

me est concorda, ka si on sav, quae multa langs fac sum-on, on facilae pos
creat bona metod de facation it, qua temp on creat artifacta lang.

> I did not, that it's *Euro*lang, which instantly removes the arguments
> that you've got a European language and are trying to pass it off as
> something more.  Thus, if you explicity say you are euro-centric, you
> forestall the inevitable criticism :-)

Eurolang (EL) is based on Romance vocabulary. It should be easy to learn
for people who know a Romance lang. For people who don't know a
Romance lang, it is harder to learn, because they won't recognise any
of the words. But it is still easier than a natural lang, because it
is regular, and bases its vocabulary on a small number of roots.

Eurolang est based super latin-groupa word-list. persons, qui sav 
latin-groupa lang, probablae pos facilae lern it. bi persons, qui no
sav latin-groupa lang, it est plus opfacila lernation, raison los no
vid-savov anya words. but tamen EL est plus facila relat opartifacta
lang, raison it est regula, et it base its word-list super opgranda
numero de radixes.
 
> Ie, in Esperanto, although widely used outside of Europe, has many
> distinctive European roots that need overcoming by other speakers.

AFAIK, virtually all E-o roots are European. 

> Ie, the sounds could have been made a lot more universal ("hx" isn't
> even a widely used European sound).  Esperanto has "s", "sh", "ch",
> "gh", and "jh", 

EL has all of these, although the "jh" sound isn't used much.
 
> > > Well, you said you don't like the accusative nor the agreement of
> > > numbers and cases:
> 
> Well, that was one of the Esperanto criticisms.  IMHO, the number
> agreement was unecessary, but the accusative would have been fine
> except that it was the only thing that had a special noun case.  Ie,
> why does a direct object take a special suffix, but no other noun
> cases?  What about a language where none of the nouns need special
> modification and are identified via position and/or particles, or one
> where noun uses are always identified by modification?  Mixing the
> two, although used in natural languages, seems odd.

I think Farsi does this.
 
> > > What about nouns and pronouns?
> > 
> > Nouns form plural by adding -s. (Or -es if the noun already ends in -s) 
> 
> Lack of a plural would be good in a universal language, but in
> a European one it's more familiar.

I agree. EL grammar is quite similar to that of English, German, French,
Spanish, Italian. It's even more similar to E-o; the main difference is
that EL doesn't have the -n accusative ending or adjective agreement. 

Since this is also the case for Mandarin and Japanese, EL should be
easier to learn than E-o for these people.
 
> > > But certainly you're not going to ignore 100+ million people are you?
> > 
> > No. But more people speak Romance+English than speak German. German 
> > speakers are no worse off learning a language based on Romance roots 
> > (since it is unlikely it would be politically acceptable for German to 
> > the EU's common language), and Romance and English speakers are better off.
> 
> Well, this seems to be a common trend in universal languages.  That
> is, at some point, vocabulary must be learned well, and for some group
> of people, there will be few familiar root words.  What this tends to
> lead to is grabbing the root words from a small set with the
> assumption that other groups will just have to learn them. 

I hope that an educated person with a Romance background will be able
to read EL after having studied it for a day. That is why it uses 
Romance roots (and to some extent grammar).

An even more ambitious goal would be for someone to understand it *without
having learnt it at all*. I know this is possible, since I can read
Gode's Interlingua without having studied it. This is certainly not the
case with E-o.

OTOH, Interlingua, although easy to read, is hard to write because its
grammar is complex (certainly more complex than E-o).

> The
> drawback is that the language gets stereotyped, and much much worse,
> the language not only uses root words from the base languages, it
> borrows heavily on the grammer, spelling, and pronounciation rather
> than keeping those necessary parts more universal. 

I have tried to minimise these problems.

> The faulty logic is that German speakers are no worse off.  This isn't
> shown.  Instead, it is likely that by making things simpler for
> Romance and English speakers, things can be made harder for other
                                                   ^^^^^^
> speakers (even if unintentional).  This is more than just picking root
> words.

I assume here you mean "harder than Esperanto". Possibly this is the
case. If 10% of EL roots are familiar to a German, and 15% of E-o roots
are, then that German speaker would probably find E-o easier. (I don't 
know what actual proportions of E-o roots come from German. Does anyone 
have these figures?)

However, if EL has less roots than E-o, then the total number of roots
that need to be learnt might be less for EL than for E-o.
 
> If that is your intention, then it is my opinion that this can
> backfire and you end up with few people getting beyond the beginner
> stage, a lot of pidgin variants, and a lot of so called speakers being
> unable to communicate.  As a real example, there was something called
> BASIC English, which was a simplified form.  The upshot was that it
> was much easier to read this language, than to write in it, since it
> was easier to read a restricted vocabulary than to actually restrict
> oneself when writing.  There was too much tendency to toss in words
> that weren't really in the language.)]

I imagine BASIC English would be very hard to write. OTOH, a program
like a spelling checker could be used, to ensure that a word was in 
BASIC, and if not, suggest an alternative word/phrase.
 
> > The purpose of this is to take the effort out of learning it.
> 
> No, the mistake is that it takes some effort out for some people.  It
> isn't necessarily easier for the rest of the people, and may in fact
> be harder.  This is the mistake made by many earlier artificial
> languages.

(Again you don't say harder or easier than what, so I'll have to 
make some assumptions.)

EL will have a small vocabulary, (Hopefully smaller than E-o), so 
people who don't know its vocabulary will still find it easier than
E-o or a natural lang.
 
> > That is why EL uses "qu", "x", "c" in words like "aquo", "export", "carn": 
> > so that the words look as familiar as possible, and people can guess they
> > mean "water", "export", "meat".
> 
> But this is another case of simplifying the language for some while
> complicating it for others.  For instance, why "x" when "ks" is
> simpler, more logical, and a more universal way of getting the sound.
> Now you have the possibility of more than one way to spell a word!

Yes, but it practise this isn't a problem, since /ks/ will nearly always
be spelt "x", unless the /k/ and the /s/ belong to separate morphemes.
The argument applies to /kw/ --> "qu".

> And few Europeans will have difficulty with "eksport". 

I agree. But "export" is even easier.

> As for "aquo",
> you set yourself up for big troubles.  You're going to be hearing a
> lot of "akvo", since "w" isn't that common of a sound.

Possibly I could get rid of the /w/ sound from the lang, and reassign
"qu" to /kv/.

> > It is designed to be as easy to read as possible, and a
> > 1-to-1 correspondence from pronunciation to spelling is secondary to
> > that. (The pronunciation can however be inferred from the spelling).
> 
> I think this is a wrong goal to take.  It is this very sort of thing
> that will make learning the language more difficult for non-Romance
> or English speakers.

The only difficulty with pron-->spelling are:
  
   sound    spelling
   -----    --------
1. /k/      c, k
2. /kw/     qu, cw, kw
3. /ks/     x, cs, ks

In (1) the 1st alternative will be correct 90% of the time since "k"
is not used much except in "international" words like "kilogram", and
in words where it is derived from a word spelt with "k".

In (2) and (3) the 1st alternative will be true 100% of the time.
The only exception is when the sounds belong to different morphemes,
such as:

  crak    "a crack"
  craks   "cracks"

How many people try to misspell "cracks" as "crax"? Not many I would
guess.

> In fact, it will make it more difficult for
> Romance speakers where the spelling isn't in the native language.

Yes. But only slightly. It is not much different from E-o where AFAIK /ts/ 
can be spelt "ts" or "c".

-- 
Phil Hunt...philip@storcomp.demon.co.uk
"on no pos fac omelet, opcum brekigation ovums"
