Re: conlang survey

En réponse à Heather Rice <florarroz@...>:

> I think there were a few people on the list asking
> about a conlang survey, so I decided to repost it.
>
I saw on your report that you had forgotten Maggel, for which I had sent a
reply. So I'm gonna resend it, edited with the new things I know about this
language.

>
> Language name,
Maggel (pronounce [m@'gE:l]).

 creator's name,

Christophe Grandsire.

 realative date of
> creation (just any old number will do),
First created in 1992. Revived and in process of complete revampification in
2002.

 country and
> first language of creator,
the Netherlands (though my birth country is France), French.

 purpose of conlang
> (auxlang, conlang, loglang, . . . ).
>
Hoaxlang and chaoslang ;)) . Its purpose is to explore the limits of Maggelity
(term meaning: the state of being entirely unpredictable, coined by Muke Tever,
see
http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0207B&L=conlang&P=R13513,
and given an exact definition by Tristan McLeay, see
http://listserv.brown.edu/archives/cgi-bin/wa?A2=ind0207B&L=conlang&P=R13711)
and cross them!

Maggel is an 'a priosteriori' (no, it's *not* a typo ;))) ) conlang: namely a
priori, but taking influences
from all the worst all the natlangs of the world can show ;))) , and contains
quite a few pun-words :))) .

> Phonetics:  number of consonants,
Difficult to say, since I have been unable to pin down the *phonemic* inventory
of Maggel (it suffers too many mutations, lenitions and other gradations to
give a good idea of its inventory. And the spelling doesn't help either,
although it's part of the goal ;))) ). But a quick glance gives around 80
consonants, which seem to be all more or less recognised by the speakers.

 number of vowels,

Same problem. A quick look gives 21 vowels and 12 diphtongues, this inventory
can be doubled by lengthening. It also has syllabic consonants (around 8, the
inventory depends on idiolect).

> presence of nasalization,
Maggel has both nasal vowels (already included in the inventory above) and
nasalised vowels and diphtongues (due to the presence of a following nasal as
coda of the syllable). Although the spelling mix them, they are different,
since real nasal vowels don't have a nasal consonant following them, while
nasalised vowels and diphtongues are nasal only because of the presence of a
following nasal).

 tone and how many,

No tones, Maggel is a stressed language.

 where the
> accent generally falls.
>
Complex rule: the accent falls on the syllable of a word which is highest in
the following hierarchy (from the highest to the lowest):
- long diphtongues,
- long vowels,
- simple diphtongues,
- simple vowels or syllabic consonants,
and on the first of them if there are more than one fitting the criterion. IIRC
the stress tends to be on the first syllable unless a diphtongue, long vowel or
long diphtongue attracts it. Note also that Maggel features 'tensed' consonants
which create a secondary stress on the following syllable (they tend to appear
only as onsets), stress which becomes actually primary if this syllable
contains a syllabic peak among the ones of the highest hierarchy within the
word. Note also that while prefixes don't modify the stress position, infixes
and suffixes can.

> Morphemes:  presence of allomorphs,
So many!

 mutation,

Yes! But like any other process in the language, it's highly irregular and
cannot even be described correctly by modern linguistic tools ;))) .

> assimilation,
Also! There is regressive assimilation of voice (the voice of the last
consonant of a consonant cluster determines the voice of the whole cluster, but
unfortunately it's not something you can always guess from the spelling ;))) ),
and partial two-way assimilation of PoA of nasals with the surrounding
consonants (with regressive assimilation winning over progressive assimilation,
and limitation in how far nasals can assimilate. If it goes too far, a stop is
inserted instead - a bilabial stop for [m], a dental stop for any other nasal -
).

 prefixes,

Yes. I can for instance give the nominalising prefix used by adjectives (not
all, there are irregularities) to make nouns of things or people described by
the adjective. This prefix is normally |ma-| [m&] or |joa-| [Za], but
distributives use a prefix |s-| with various phonological outcomes. There is
also the |e(i)-| augment used to mark non-singular numbers of neuter nouns.

 suffixes,

Plenty!

 infixes,

Not yet, but I can't imagine Maggel doesn't have any ;))) .

> suprafixation,
What does it mean?

 dicontinuation,

Idem.

 exclusion,

Idem.

 total
> fusion,
Idem.

 subtraction,

Yes! (mainly in some forms of the adjective)

 reduplication.

Not yet, but probable.

Maggel also have different types of affliction, i.e. umlaut and ablaut. I can
think for instance of the a-umlaut of the dual of masculine nouns and the i-
umlaut of the plurals of those same nouns. Note that those umlauts are rather
strange in themselves. For instance, the i-umlaut result of |e| [E] is |ei|
[Ui]! (umlaut is quite transparent when written, but the pronounced result is
not quite so :)) )

  Is the conlang
> agglutinating, isolating or fusional?
>
Mainly fusional, with hints of isolation and agglutination, and a light touch
of polysynthesis ;))) .

> Nouns and such:  subclasses of nouns (common/proper,
> abstract, things that may not be expressed explicitly
> in affixes),
Three main classes of nouns: countable, uncountable and plural (not to confuse
with plural number. The terminology is confusing I know :)) ). Countable nouns
can be put at any of the four numbers, uncountable can be put only in the two
plural numbers, and refer to things that cannot be counted but can be measured
(like sugar or water). Plural nouns are nouns which are semantically countable
but grammatically uncountable. Result: they only have plural forms, and yet
have a singular meaning! (explicit plural is given with other means for those
nouns)

 presence of cases and how many and what
> kind,
Not really, though some common nouns in Maggel still have a non-subject form
(highly irregular) and most have a construct form (used to mark the possessed
noun in an absolute construction). Both forms can appear on the same word at
the same time.

 kind of possession (alienable, inalienable, no
> distinction, etc.)
I'm not completely sure yet, but I think Maggel doesn't make any difference
between alienable and inalienable possession. On the other hand, it makes
another distinction, namely between subjective and objective possession. The
normal marks of possession (the absolute construction consisting in adding a
noun after another noun, the second noun possessing the first, and the
possessive prefixes) only mark objective possession, i.e. if the possessed noun
can be converted into a verb, the possessor will be mandatorily the object of
this verb (and not the subject like most languages do!). Note that objective
possession corresponds in Maggel also to actual possession, since the verbal
expression for possession always have the possessed as subject and the
possessor as object. Subjective possession is rendered using prepositions,
mainly |o|, but others are possible.

 presence of gender,

Indeed! Maggel has pure grammatical gender, divided in masculine, feminine and
neuter. Note that this gender is *not* connected to natural gender. It means
that naturally male living creatures can be named with a masculine noun, but
it's not mandatory and is not even majoritary (it only corresponds to a
plurality of cases). Moreover, gender agreement of the pronouns is not
according to the natural gender of the antecedent, but according to the
grammatical gender of the noun naming it! So for instance, since the word for
captain (which I've not discovered yet, although I know its gender since it
comes from an expression meaning "ship's wife") is feminine, you must use
feminine pronouns to refer to a captain, even if this person is a man!!! It
means that, depending whether you refer to somebody by his/her profession,
status, or just name, you may use different pronouns, and that in the same
conversation!!!

 number,

Yes. Nouns in Maggel can be put in 4 different numbers: singular, dual, plural
indefinite and plural definite. The two different plurals don't have to do with
definition as it's usually meant, but to whether there is a grammatical
indication of amount (even imprecise) or not. If not, use the plural
indefinite. If so, use the plural definite. I still don't know whether this
indication must always be present grammatically, or if once it's been done the
plural definite must always be used, whether the indication is repeated or not.
Note that there are some exceptions too. For instance, I've recently discovered
that with numeral nouns (referring to high quantities), the plural indefinite
has to be used (while with the numerals themselves the plural definite is
normally used). Note also that some nouns, referring to things or people coming
in natural pairs, have two duals. The regular one refers to an unnatural pair
(like two right hands), while the second, very irregular dual refers to the
natural pair (a married couple for instance, or both my hands).

> articles,
Maggel has one article, |a| (|an| in front of a vowel), but I cannot call
it "definite article" since definition is not its main job. Actually,
the
article is an incredibly empty word whose presence is nearly entirely dictated
by the presence of other types of words. In only two cases the article has a
meaning:
- with a noun alone, without any adjective or other noun to complete it, the
article does mark definition,
- with some prepositions, the presence or absence of the article modifies the
meaning of the preposition (in which case the article doesn't mark definition
at all).
All other cases are dictated by the other words present around the noun. For
instance, in the absolute construction, the possessed noun *never* takes the
article, while the possessor *always* takes it (and exceptionally, the article
must be in the form |an| even in front of a consonant). Adjectives demand that
the article is added, whether the noun is definite or not, while prepositions
(except the ones of the case above) prevent it completely. Finally, numerals
use the article to modify their meaning, from cardinal/ordinal to
multiplicative adverb. Of course, all those influences are contradictory, and
the presence or absence of the article is often a complicated story. Still, the
absolute construction and the presence of a preposition are the strongest
influences, and they always decide of the presence of the article, whatever the
other words completing the noun.

 demostratives,

Probably, but I didn't work on them yet. I am nearly sure that they won't be
distinguished from personal pronouns and that they will include "out-of-the-
world" pronouns, i.e. pronouns referring to things that are outside of this
world (pretty practical to discuss about alternate realities, although I think
its basic use would be to discuss about things happening in dreams).

 adjectives,

Yes, and adjectives are a special part of speech, comparable neither to the
nouns nor to the verbs. They have a special behaviour, different from any other
parts of speech.

 quantatives.

Do you mean numbers? Quantitative pronouns like nobody, everyone or someone? In
the case of numbers, Maggel has them, and they are a different part of speech,
even different from adjectives (some numerals are adjectives and others nouns,
but they are derivations from the main numbers). Also, the types of numbers are
not the same as what you find in most natlangs. Numbers can be in the
declarative, combined, distributive and personal forms. I think there may be
other forms, but I've not discovered them yet :(( .
The declarative form is used when the numbers are used alone, for instance in
phone numbers or mathematics. They are *not* used to count something which is
understated. The combined numbers are used for that. The declarative form is
the basic form of the numerals, and all the other forms can be said to be
derived from it.
The combined form is used when the numeral completes a noun (whether this one
is stated or not), or as multiplicative adverbs. They correspond to the
cardinal numbers as well as the ordinal numbers, and their main characteristics
is the presence of the article in front of them (but the article disappears to
make them into multiplicative adverbs). They are cardinal when used in front of
a noun, and ordinal when used after it. Only in the cardinal use the noun can
be missing. In the ordinal use the noun, if missing, has to be replaced by some
semantically empty noun.
The distributive form gives the meaning "one by one", "two by two", etc..., and
contrary to other forms is completely adjectival. Distributive numerals behave
exactly like any adjective. They can be used to count plural nouns (those nouns
which have a plural form but a singular meaning).
The personal form is used only to count people, and can be used alone as well
as with a noun referring to persons. Contrary to other forms, the personal
numerals are full-fledged nouns, and thus can be translated
by "duet", "trio", "quartet", etc... Since they are mandatorily used to count
people, it's as if in English the phrase "two men" was ungrammatical, and you
are obliged to use the phrase "a duet of men". They don't exist for all numbers
though (I think they exist only until 16, but I'm not sure yet).

An important thing to know is that numerals as a separate part of speech exist
only until 7999. Nouns are used for higher numbers, and demand very different
constructions when completing other nouns.

  Are
> comparatives expressed by affix, word order or both?
None of them. Adjectives are not modified when they have a comparative or
superlative meaning (exception: adjectives ending in -VCVC lose their last
vowel). For comparison, it's the preposition used to introduce the thing
compared which marks which kind of comparison it is (e.g. the preposition |lu|:
on, marks comparison of superiority). For the superlative meaning, Maggel uses
an expression where "the biggest" is rendered as "big among the
big ones".

> Do pronouns express gender, number, declension?
All three.

  Are
> there indefinite pronouns,
Yes, although I don't know yet whether there will be separate pronouns (they
are a separate part of speech, comparable neither to nouns nor to verbs nor to
adjectives), or if they will be just nouns and/or adjectives (or maybe
verbs :))) ).

 possessed pronouns?

Do you mean possessive pronouns like "my", "mine"? "my" and the likes are not
pronouns in Maggel but prefixes. They are also used to mark the object in some
verbal constructions. I don't know yet whether there are any pronouns
like "mine", "yours", etc...

> Others?
Don't know yet, but count on Maggel to have quite strange things around ;))) .

  Are prepositions bound, unbound?

They are written unbound, but they are probably bound.

  How many
> prepositons (approximate).
Not a lot. A score of them or so. Most of them have multiple meanings, and
specialised meanings are rendered using prepositional phrases completed with
another noun (like "among" in the example above is actually |f ihbaodhoir|
['fTa:x], meaning "at the middle of". It's the first part of an absolute
construction using the noun |ibaodhoir| ['pa:x]: middle).

  Presence of clitics.

A lot! Apart from the article and the prepositions which can be argued to be
clitics, there are also true clitics, written bounded to the previous word, and
often corresponding to conjunctions. For instance, there is the clitic |-ba| [v
(U)], meaning "and", and used only between nouns and with numerals (with
adjectives, you have to use the clitic |-jn|, pronounced  [zd(I)], and
meaning "and/or").

  Is
> derivational morphology mostly by compounding words or
> by affix or both?
>
Mostly by affixation. There are no compound words as such, but there
are "eroded phrases", phrases that became used so often that they began to be
treated as one unity, and got eroded until they had only one accent and became
a single word. Those words usually have very strange plurals, reminiscent of
their former status.

> Verbs and such:
Note that the rest of this presentation is to be taken with a big question
mark, as I have not worked on it yet, and most of what I know about it has not
yet been put on paper.

> Are person, number, object expressed with the verb?
Not exactly. It seems that in some persons the subject pronoun (when present)
combines with the verb, but not in all (details still unknown). Also, with
analytic constructions (using a nominal form of the verb), when the object is a
pronoun it appears as a possessive prefix to the nominal form of the verb.

> Are there static verbs (to be)?
Yes! Like in Celtic languages, 'to be' is the most important verbs in Maggel
(and I'm not even sure putting 'verbs' in the plural was a typo ;))) ).

  Is the object
> incorporated into the person marker (making a
> phonetically different affix like in the Native
> American languages)?
No.

  Is transitivity marked for
> transitive, intransitive, bitransitive or other?
I don't know for all verb conjugations yet, but I don't think so.

  Is
> the person inclusive, exclusive, no distiction?
Maybe in the dual number, I don't know yet for sure.

  Kind
> of gender.
?

  Are past, present, future expressed?

Yes, although I don't know the details yet. I do know, though, that Maggel has
quite complicated conjugations.

> Recent, remote?
I think it does make this distinction both in the past and the future.

  Is mode express, what kind?

Yes. Grammatically, I'm sure of the presence of an indicative mood and a
subjunctive mood. Maybe an optative mood is there too...

  Is voice
> expressed?  What kind?
Yes, but no passive! Maggel has active voice (the default), two voices which
substract one core argument (reflexive and reciprocal) and various voices which
*add* a component! (applicatives, and probably causative). It doesn't have any
passive at all, although verbs often have an impersonal form which can be used
as a pseudo-passive. It has also something which is traditionally described
with the voices but doesn't seem to change the valency of the verb (as far as I
know. I may be wrong though as I'm under the impression that the subject in
such a construction must be changed somehow). It's called the "sensitive" and
seems to be used to indicate the the speaker (not the subject) has some feeling
associated to what he says. It seems to work a bit like an evidential as it
indicates: "be careful, what I'm saying is important to me", or "this is an
opinion, I'm not sure it's true", or "I (dis)like it", or even "I am involved
somehow in it". But although it indicates that the speaker is involved in it
through his/her feelings, it doesn't precise which kind of feelings they are.

  Manner?

Probably using phrases and adverbs.

  Aspect?

Probably continuous and perfect. Not sure for the rest.

  Please list
> what kinds of manner and aspect the conlang expresses
> in its verbs.
I don't know yet :((( .

Maggel makes another distinction: performative vs. describing. Most languages
have semantically performative verbs (verbs which perform an act just by
pronouncing them, like "I command you") separate from describing verbs
(with sometimes a bit of overlap). Maggel makes the distinction grammatical, so
that virtually every verb has a performative and a describing form.

  Presence of adverbs,

Maggel does have adverbs (again a separated part of speech). What it doesn't
have is adverbs derived from adjectives! Instead, it uses the nominal forms of
those adjectives in manner phrases.

 pro-drop.

No. There is the impersonal form of the verb which never takes a subject, but
since it's a special form it doesn't really count.

  Can
> nouns, adjectives, adverbs be changed to verbs and
> vice versa?
>
Not that easily. Derivation in Maggel is usually quite complex and irregular
(if you think French derivational morphology is irregular, think again ;)))) ).
But there are possibilities.

> Presence of adjective, adverbial clauses and relative
> pronouns.
>
Quite unsure about the presence of relative pronouns, but subclauses do exist
in all kinds of ways. I just don't know yet how they are implemented :(( .

> Sentences:
> Does the conlang have an ergative or accusative
> system?
Ergative? Accusative? Maggel kicks them and sends them to orbit around
Pluto!!! :))) Seriously, Maggel seems to appear quite accusative, but it's
difficult to be sure with its strange syntax.

  Word order and is it free or strict?

Sentence word order is relatively fixed: mainly VSO, though VOS and OVS do
appear depending on the conjugation, clause status, etc... Most clitics and
clitic-like words, like the article or the prepositions, also have a fixed
position in the noun phrase. The place of the combined numeral is important to
determine its meaning. On the other hand, adjectives can appear as well before
as after the noun they complete.

Maggel also has some strange ordering habits. For instance, nothing can go
between the noun and its article or the noun and its preposition, so when the
adjective goes in front of a noun, it is put in front of the article or the
preposition!

  Are
> adjectives, adverbs and prepositions before or after
> the modified word?
I already answered that actually ;))) . Adverbs are relatively free, like
adjectives. They tend to come after the word they complete though. Maggel is
quite head-first, but likes to break its own constraints ;)))) .

  Is the word order changed in a
> question?
No. But the verbal conjugation is heavily modified, which in turn can change
word order in some cases (but that's a result of the conjugation used, not of
the fact that the sentence has become a question).

  How many (approximately) conjugations are
> there?
>
You mean classes of verbs? Well, difficult to say in a language where basically
every word is an exception, and only exceptional words are regular ;))) . But
I'd say that you can broadly fit all verbs in three classes: unseparable
verbs, separable verbs (verbs comparable to the German or Dutch verbs with
separatable prefix, which can appear as one of two pieces depending on the
environment. But unlike in Dutch and German, there doesn't seem to be any logic
nor reason behind the cutting of those verbs) and the verb 'to be', which
deserves its own category (since it even makes distinctions other verbs
don't!).

> Other:
> What is the number base for the numeral system (10?
> 12?)?
Mainly 20, with shades of 5, 8 (or maybe 16), 11, and even substractive
elements! The highest two numeral nouns seem show a base 10, but seem to be
borrowings. Actually, those are strange borrowings. For the first one, its
shape seems to have been borrowed from some noun, but its meaning from another!
It is |muorhiadf| ['mIr\_j&d]: million. The second seems to be a
backformation from the example million-billion, but with the continental
meaning of the prefix. So we have |bmuorhiadf| ['bIr\_j&d]: 1000 billion (or
billion in the French sense).

  Presence of idioms,

A lot! 'Ships's wife' for 'captain' was just one example :))) .

 irregular forms of nouns
> and verbs.
Are you kidding? This language *is* irregularity itself!!! :)))

  Is the language syntax very predictable,
> or are there many exceptions?
What do you think? ;)))))

  How much literature has
> been produced and what kind (I'm not talking about
> translations, but stuff you wrote yourself).
Nothing yet, because I'm still in the process of working on the grammar.

  Is there
> a history and dictionary of the conlang?
Not yet. There is this nice idea from Hanuman Chang that this language could be
the language of the Picts if those had survived until now, but right now it's
no more than a nice idea ;))) .

  Script
> invented?
Maggel is written using a subset of the Latin alphabet, written in a form of
Uncial style and using strange and numerous ligatures. Like in the Arabic
script, some letters automatically connect to the next one, while others don't,
which is a strange feature for an alphabet derived from the Latin alphabet.
Here on the web I use a Latin transliteration which is easy to make since
Maggel uses only 17 of the 26 letters of the alphabet. I just have to take the
Latin equivalent of its Uncial letters. In the order of the Maggel alphabet,
the transliteration has:
a m r n o u i g l e t s f d b h j
Note that the direct transliteration hides some shifts of use of some letters.
For instance, |t| *never* has the value [t], but always refers to some
fricative (usually [T] or [D], but often also [C] or [j\], and even [K] or
[K\], depending on the letters around).

  Other conlangs produced by the creator of
> this one.
>
Astou
Azak
Moten
Reman
Notya
Tj'a-ts'a~n
Chasmäöcho
O
Narbonósc
Itakian

And a few other projects which are better left unnamed ;))) .

> If you could summarize your conlang in a sentence,
> what would you write?
>
What is the difference between a rhinoceros?
- It doesn't know how to fly either.

I think it sums it up pretty well ;)))) .

> On my servey, I knew I couldn't possibly cover
> everything that conlangs will be, so I included a long
> notes section.  If you want to provide any other
> information about the conlang, just keep typing!
>
Hehe, what could I write more... You could write your book only about Maggel
and just scrap the surface of it ;))) . Maggel takes irregularity as a way of
life, surprises even its author on the way it works things out, is a master in
presenting a familiar face and then, at the moment you expect it last, strike
with a completely unpredictable feature! And of course it goes the other way
round: at the moment you think you're completely lost, it shows suddenly a
familiar face! The wonder and annoyance of Maggel is that it's never *totally*
exotic. If it was, the brain would promptly give up trying to understand. But
with Maggel it's always under the impression that there must be some secret
principle that will dissipate the clouds of irregularity and show the
regularity hidden deep inside the language, and keeps on trying to find it,
never stopping, despite the hopelessness of such a search ;)))) . Written,
Maggel is the equivalent of what you get when you fall asleep on your keyboard
and begin snoring. Spoken too ;)))) . But like the Tao, the Maggel which can be
explained is not the true Maggel. So throw away all my explanations. Maggel
just *is*.

> Thanks a lot all!
>
You're welcome!

Christophe.

http://rainbow.conlang.free.fr

Take your life as a movie: do not let anybody else play the leading role.