sci.lang discussion of Esperanto pt 1 of 4

[Logical Language Group <lojbab@GREBYN.COM>]

Complexity of Esperanto Syntax  Pt 1 of 4

Over the last couple of months on sci.lang (and possibly at least in part
on soc.culture.esperanto), there has been an interesting discussion
on the linguistic properties of Esperanto.  Now that it has finally ended,
I have edited the discussion together into a single stream (in order to
make it short enough to post).

This discussion shows the type of questions I think people who are
seriously trying to develop conlangs need to discuss, if they want to
have a complete design, rather than the old-hat discussions of phonology
and morphology that seem to dominate everyones description of their pet
project.  Esperanto is clearly shown, despite its 'simplicity', to be
an extremely complex language for people to actually use (and the
Esperantists agreed!).

We've tried with Lojban to define a lot of the things that were left up
to custom in the design of Esperanto (rather than explicitly defined),
hopefully resulting in a more satisfying result, at least for linguists.

I didn't have time to do a comparison with Lojban.  Perhaps someone else
in our community would wish to tackle this???

The summary is divided into 4 parts since some people do not receive
traffic over 16K.

lojbab
----
lojbab                                                      lojbab@grebyn.com
Bob LeChevalier, President, The Logical Language Group, Inc.
2904 Beau Lane, Fairfax VA 22031-1303 USA                        703-385-0273

==========================================================
MR:  From: markrose@spss.com (Mark Rosenfelder)
DH:  From: donh@netcom.com (Donald J. Harlow)
KM:  From: miner@kuhub.cc.ukans.edu
PJ:  From: pcj1@cunixf.cc.columbia.edu (Pierre Jelenc)
ID:  From: iad@cogsci.ed.ac.uk (Ivan A Derzhanski)
NY:  From: nyoung@desire.wright.edu
EGE: From: etg10@cl.cam.ac.uk (Edmund Grimley-Evans)
SLB: From: slb22@cunixb.cc.columbia.edu (Seth "the Lesser")

MR1:
Esperanto syntax is about as idiosyncratic and complex as that of any
natural language.
(later modified to read "the average natural language")

 DH1:
 Having studied Esperanto, Latin, and (to some degree) French,
 Russian and German, I would seriously question Mark's contention
 that Esperanto's syntax is in any way as idiosyncratic and complex
 as the syntaxes of any of the other four languages in that group.

  MR2:
  Well, this is merely an unsupported assertion... as was mine, of course.
  Would you care to build a case for your position?

   DH2:
   I should here paraphrase an old proverb:  "Post in haste, repent at
   leisure."  Actually, I would agree with Mark that Esperanto's syntax is
   "probably" as complex as that of any "average natural language"
   (assuming that we all know what a "natural language" is, and what is an
   "average" one).

    MR4:
    Oh, good.  I won't even object to the "probably", given the informality
    of our discussion, and the lack of in-depth syntactic analyses of
    Esperanto.


   DH2: (cont.)
   I responded quickly, and took "complex" to mean
   "complicated," which of course it does not.  I would argue that
   Esperanto's syntax is not as "idiosyncratic and complicated" as those of
   other languages, but "complexity" is quite another matter.

   As to building a case -- sorry, I have no rigorous way of doing so.  I
   can only quote my own experience, which is "hands on" rather than
   theoretical.  During the period 1957-1960 I studied Latin in high
   school, and in 1960 won an award at a statewide foreign-language field
   day as the best Latin student in the state of Oregon.  During six months
   in 1959- 1960 I was an exchange student in Denmark, and was required to
   learn and use Danish in my studies and daily life.  At various times I
   studied French (one year), Russian (one year) and German (one semester)
   in college, and have spent some 28 years in the company of a native
   Spanish-speaker (my wife) and various in-laws.  I can read (to some
   extent), understand, write and speak (to a much lesser extent) any of
   these languages, but only with extreme difficulty.  I studied Esperanto
   on my own, out of Cresswell & Hartley's "Teach Yourself Esperanto", in
   the summer and fall of 1959, and in October of that year heard the
   language spoken for the first time, at a meeting of the Randers
   Esperanto-Forening in Denmark; the speaker was a Japanese visitor, and I
   was astonished to discover that I could understand everything he said,
   including one very bad pun.  I have never had a similar experience with
   any other language I've studied.

   Whether this can be attributed to a less complicated and idiosyncratic
   syntax is a question that others would have to answer.  Perhaps there is
   a budding linguist out there who is looking for a dissertation topic?

    MR4:
    Just one quibble:  syntax is precisely what you can throw out or mangle
    and still have a fair shot at understanding.  I'd agree that Esperanto
    is easy to learn (for English speakers, at least), but not necessarily
    because of its syntax.

    By the way, what Esperanto I have I learned from _Teach Yourself_ book
    too.


  MR2:
  By the way, do you consider English syntax simple?

   DH2:
   Not that I've ever noticed.

    MR4:
    Good; I would have jumped on you if you said it was.  :)


  MR2:
  To approach this question scientifically we'd have to do some solid
  syntactic investigation, such has been done with English over the last
  thirty years.  I don't know how much of this kind of work has been done
  with Esperanto.

  I'll attempt to provide a bit of informal support for my claim by noting
  some areas where the same kinds of complexities found in natural
  languages exist in Esperanto.  Some of the complexity lies in the mere
  fact that these constructions exist; but I emphasize that most of it
  will lurk in the detailed rules that tend to surround these
  constructions, and which would require further study to elaborate.

   DH2:
   Some reactions to a few of Mark's specific comments:

    MR4:
    Thanks for the reactions, which however I won't pursue.  I was trying to
    suggest the range of questions a syntactician would be interested in; I
    still think a detailed study of Esperanto syntax and pragmatics would be
    quite worth doing.


  MR2:
  * Case usage.  There are arbitrary differences between verbs, as seen in
      Ili diris AL SHI kelkajn vortojn.   'They spoke several words to her'
  vs. Demandu do LIN vi mem.              'So ask him yourself'

   DH2:
   Personally, I always say "Demandu do AL LI vi mem."  It is also
   legitimate to say "Ili diris SHIN", but not if there is a direct object.
   In actual usage, the -N ending can replace the preposition AL as long as
   there isn't a direct object to confuse the matter (this basically
   follows from rule 13); but since, with words such as "diris" or "sendis"
   there is almost always a direct object, you almost never see this usage.
   (This "priority" rule, incidentally, is an interesting example of a
   syntactic rule not, so far as I know, explicitly described anywhere.)

   ID1:
   Could there be no arbitrary differences between verbs?

    MR5:
    Sure there could; but they add to the complexity of the language.

     ID5:
     Then which differences are arbitrary and which aren't?  Could the
     complexity of the language be lowered by postulating that all speech
     verbs must use the same marking for the speaker, the same marking for
     the addressee and the same marking for the content of the communication?

      MR6:
      Ceteris paribus, sure.  Too bad Zamenhof didn't read Fillmore...


   ID1:
   Why does one expect the two highlighted arguments above to bear the
   same case marking?

    MR5:
    Well, French manages it: _Ils lui disaient quelque chose_; _Je lui ai
    pose la question_; _Demandez-lui vous-meme._

     ID5:
     So French marks the addressee of `tell' ("dire") and `ask' ("demander")
     in the same way.  Is this a desirable thing to have in a planned
     language?

      MR6:
      I was thinking about the complexity of Esperanto syntax, not about what
      is desirable.  But since you ask:  sure, I'd think it would be
      desirable.  Not that I think it matters what features exist in a planned
      language.  It's not the lack of some feature or another that keeps
      Esperanto from taking off.


   ID1:
   Do they have the same thematic role, and if so, what is it called?

    MR5:
    c. Are you assuming that thematic roles are the same in all languages?

     ID5:
     I don't see why I should assume the contrary.

      MR6:
      Nor do I; but I wouldn't exclude it either.  How much analysis of deep
      case has been done in non-European languages?


  MR2:
  * Preposition usage.  Many of the footnotes in the _Fundamenta
  Krestomatio_ relate to choice of preposition.  Again, many are simply
  arbitrary:  why do you get married _kun_ and not _al_ somebody else?

   DH2:
   Actually, you can get married _al_ somebody, depending on what word you
   use for "get married":

   Mi edzighis _al_ mia edzino (I got married to [became husband to] my wife)
   Mi geedzighis _kun_ mia edzino (I got married with [jointly] my wife)

   Personally, I prefer the latter.  It emphasizes the "jointness" of the
   act of marriage.

   ID1:
   This is essentially the same question as the first one.


  MR2:(cont.)
  * Collocations.  It's often quite a chore to know what verbs go with
  what nouns in a language, and I don't know why Esperanto would be an
  exception.  For instance, you 'take' a course in English, but 'follow'
  it in French.

   DH2:
   In Esperanto, "Mi STUDAS la kurson", which seems reasonable.

   ID1:
   And in Bulgarian the student takes the exam, if the professor will give
   it to him, whereas in Russian it is the other way around.  But this has
   nothing to do with syntactic complexity, as far as I can see.

    MR5:
    What words must be used with what other words does seem like syntax to me.

     ID5:
     What categories of words must be used with what other categories of
     words is syntax, but here we're dealing with phraseology, to which
     syntax is insensitive.  For all syntactic purposes that I can think of
     "take a pencil" and "take an exam" are identical.


   ID1:(cont.)
   In any case, I would expect a planned language to restrict metaphor to a
   necessary minimum.

    MR5:
    Which leads to the questions-- Why do you think that would be desirable?

     ID5:
     Because the only two alternatives are to allow any speaker to come up
     with any metaphor that seems handy to him, which (given the speakers'
     different cultural backgrounds) would make the result unintelligible, or
     to do what natural languages do, that is, to postulate a (necessarily
     idiosyncratic) set of licensed metaphoric expressions for learners to
     struggle with.


    MR5:(cont.)
    Why do you think it would even be possible?

     ID5:
     Well, since that necessary minimum can't be strictly defined, what I
     formulated was a tendency rather than a state of affairs to achieve, and
     Lojban, for example, follows that very tendency.

      MR6:
      I was asking these questions thinking of Lakoff's ideas on the
      centrality of metaphor to language.  I also wonder whether one can
      dictate the amount of metaphor in a living language...  I suspect
      speakers will use it extensively no matter what the language academy
      says.


  MR2:
  * Transformations.  Like any natural language, Esperanto allows clauses
  to be embedded in other clauses in various ways:  via conjunction,
  relative clauses, participles, etc.  It would be interesting to know
  what kinds of clefting and raising are permitted-- or not permitted-- in
  the language.  What are the rules for pronoun movement, or negative
  movement?

  * Impersonal constructions.  One can say either
     Gravas manghi multe da freshaj fruktoj
     Estas grave manghi...          'It's important to eat lots of fresh fruit.'
  These are a nice source of complications in English; how about Esperanto?
  For instance, is the "Estas..." form ever prohibited?  ever obligatory?

   DH2:
   No and no.  Which one you use is a question of personal style
   (personally, I'd tend to alternate them; that way, nobody gets bored).