Morpheme index project

Hello, :-)

I'm interested in building an index for the morphemes
of a projected isolating IAL. In it, they would be listed
without having them assigned to the actual words that would
represent them in the language yet (because one of the most
difficult parts for IAL designers is to reach at a consensus
about its external appearance: phonology, source for
vocabulary, etc.), but just defining them, e.g. "progessive
aspect marker", and assigning them an indexing 4-digit
hexadecimal number that would serve to identify them clearly,
easily and shortly when designing the grammar and to have
them neatly organized and classified. The indexing would be
based upon semantic-functional grounds, thus grouping together
e.g. all the morphemes for the semantic field "colour", all the
morphemes for aspect markers, etc.

Please note that I'm not talking about a "philosophical"
classification nor about linking form to classification
(what's more, on the whole contrary, I'm in favour of
having closely related morphemes clearly differentiated by
means of unconfusably different words, so as to avoid
misunderstandings such as those caused by the similarity
of e.g. Esperanto's pronouns "mi" and "ni").

The morphemes in a language MUST be classified, at least
for what concerns their function (or range of functions);
that's true even for languages like Chinese and Vietnamese
(you can't place and use a word like "shi4" in the same
places and ways as you place and use a word like "wo3",
which proves they belong to different word classes). And so,
I think it's much better to have them classified into a useful
index, at least for the sake of keeping a neat record of
which morphemes we have already. Thus, the index would list
e.g. first functional morphemes, grouping all of each class
together, then lexical morphemes, grouped according to the
semantic field they belong in a manner similar to a thesaurus.
Otherwise, the result would be an impractical listing like
that of Lojban's sumti, which forces you to read and study it
from head to top before you can make yourself an idea of e.g.
which concepts it considers basic for semantic fields like,
say, colours, animals or kinship terms.

The main purpose of the index is to help in designing
the IAL: it will allow to easily develop the grammar
in abstract schemes (the index should also list codes
for groups of morphemes, e.g. a code would stand for
"any morpheme of the set of determiners", so as to
allow for general sentence formulations); to check that
grammar using computer programs; to analyze the morpheme
frequency and distribution and test different
word-morpheme assignments in order to match them in
"phonetically harmonic" sentences (I mean, e.g. to avoid
word-morpheme assignments that lead to frequent
alliterations and tongue twisters); to start building
dictionaries without a previous assignment of actual
words to morphemes, etc.

Examples of possible entries in the morpheme index:

0000 (any morpheme)
0001 predicating particle
0002 existence declarating particle
...
...
08b7 locative marker
...
...
...
1d51 determiner: an indefinite one
...
1d54 determiner: that one there
1d56 determiner: the one previously mentioned
...
...
...
35a2 adjectivizer
...
...
6f41 colour: white
6f42 colour: black
...
...
823c birds: duck family
...
...

So, with the help of the morpheme index, we could now
give sample sentences of the projected IAL as this:

- 1d56 823c 0001 35a2 6f41
(The mentioned duck is white.)

- 08b7 1d54 0002 1d51 823c 35a2 6f42
(There is a black duck there.)

Thus not having to include a long description there for
each of the morphemes nor needing the actual words that
in the end would carry their meaning in the IAL.

Then, at the end of the process of designing the IAL,
when the grammar is fully developed and a consensus about
phonology and the source for vocabulary finally reached,
morphemes would be assigned to actual words, e.g.:

i    0001 predicating particle
e    0002 existence declarating particle
ce   08b7 locative marker
nu   1d51 determinative: an indefinite one
ta   1d54 determinative: that one there
ga   1d56 determinative: the one previously mentioned
ki   35a2 adjectivizer
bil  6f41 colour: white
hoj  6f42 colour: black
wak  823c birds: duck family

The previous sample sentences would then look as:

- Ga wak i ki bil. (The mentioned duck is white.)

- Ce ta e nu wak ki hoj. (There is a black duck there.)

Any help will be kindly appreciated. E.g. proposals for
a general subdivision of the chart, proposals of morphemes
(with their definitions; e.g. "anaphoric pronoun", "colour
white"), of functional-semantic groupings (e.g. "set of
pronouns", "set of colours"), proposals to assign those to
a certain range of codes (e.g. "set of determiners, possible
code range: 1d00 to 1d29"), etc.

Thanks a lot, ;-)

Cheers,
Javier

Reply