Tuesday, May 27, 2014

Detail 88: Verbs of motion

Often, languages differ in what exact things they encode in their verbs of motion: is it manner (run vs. walk vs. slouch vs. mosey vs. hobble) or direction/aim (ascend, descend, enter, exit, approach, circumambulate). We find that in English, Latin loans tend to incorporate direction and such information, whereas Germanic inherited words tend to just encode the manner of motion.

Typologically, these two seem to be the categories of most interest.

What if we'd have verbs also encode for environment of movement? In English, we have wade (for walking in water), and sort of trek (often for walking in mountains or wilderness although it seems to increasingly also encode something about the type of walk you're taking: a walk of considerable distance for enjoying nature). But let's code for things like walk in mud, walk in water, walk in a forest, walk in a hilly area,walk in a desert, ...and then the same for run, and for a few others.

In one way, it'd tell us quite a bit more about what the culture the language is supposed to present cares to code for.

Thursday, May 22, 2014

Dairwueh: Verbs for possession

Dairwueh almost conforms to the average European way of having a verb primarily used for such constructions. However, it is not a direct translation of 'to have'. In fact, there is not one verb but two, with certain distinctions.

The two main verbs are gadek and surdun. Gadek is used when the possession of a thing is seen as something beneficial. On the opposing side, surdun obviously stands for ungainful possession.

By adding a reflexive morpheme, gadek and surdun become intensive. The intensive of surdun basically means 'being overburdened by, succumbing to', whereas gadek in its intensive form tends to be used to denote strategic, decisive advantages given by the possessed thing.

Certain things that English could express using have is not expressed that way in Dairwueh. Diseases are suffered, pusteg, you are born, imbet, for your relatives (who are in the dative), younger relatives are born to the status of having you as a relative, placing you in the dative. The verbs used for expressing relatives also are used to state other things about relations. These complications require a small diversion:

eme imbewis nesepalivit resepanivik 
I     born.past saints.dat criminals.dat.and
I have saints and criminals for relatives/my relatives are saints and criminals
 ver imbewis loparvit
I born.past brother.dat - I have a(n older) brother
lopar imbewiŋ vevit brother born.pastIIIsg me.dat - I have a (younger) brother 
If one wants to express a multitude of relatives, if there is a single one older than the person whose relatives are being listed, that person is the subject, if all are younger, he's the indirect object.
ver kar imbewis sopind
I him-dat born-1sg_past cousin - I was born his cousin
kar vevit imbewiŋ sopind
 he me-dat ... - he was born my cousin
Generally, ungainful but unharmful possession is expressed using gadek. However, if there is a need to distinguish neutral status from beneficial status - which does happen on occasion, more so in the literary and legal languages than in colloquial varieties, a number of periphrastic constructions are used, such as "X is Y-dat", "Y holds X", "Y took X" for the neutral type.
-ci is obsolete and replaced by -ivit. -k assimilates the -t in -ivit. 'wemi' replaced by 'vevit', 'eme' replaced by 'ver', komi by kar, kon by ker

Wednesday, May 14, 2014

Tatediem morphology: Gender congruence on nouns, adjectives and verbs

1. Gender and Case Marking

Since the gender and case system of Tatediem is inspired by Bantu-style morphologies, it has taken me a lot of effort to break out of the squarely European-style patterns I am used to.

The gender marking on the noun is a prefix. For most genders and cases, this is only present for the definite form. There may be complicated tone sandhi effects caused by the prefix. This morphophonological detail will be dealt with in a later post.

Most prefixes are monosyllabic, along the lines of ku- or sar-. Two are longer, and there are also asyllabic prefixes, t- and l-. The "grammatical" category contains a finite set of rather functional words - conjunctions, a few special adverbs and such. As far as morphosyntax goes, these tend to behave somewhat like nouns, and they also may trigger congruence on verbs and such.

Table of nominal prefixes:
           sing           dual     plural     (mass)
masc    ne-|0         wan-   lan-      (re-)
fem      sar-|0        xan-    lan-      (re-)
neut 1  ku-|0         t-        tsi-       (kku-)
neut 2  ye-            ya-      l-         (l-)
neut 3  gah-         geme-  gge-       (ske-)
neut 4  sse-         setem-  t(t)ew-     (ske-)
gram.  re-            ra-    rax-     re-

(Or n3s: ye-, n1p: l-)

Finally, neuter 1 sometimes uses ye- as a gender marker on indefinite nouns. The partitive is not definite, but sometimes causes gender congruence. The 0-prefix on masculines, feminines and neuter 1s does cause tone sandhi.

Adjectives take a somewhat less detailed set of prefixes:

e- wa- a- de-
ra- ra- la- de-
ke- ke- ke- kku-
ye- l- l- l-
ye- gge- gge- gge-
ye- ye- ye- ye-
ur-   ur-  ur-  ur-

Adjective congruence disregards definiteness and is thus present even in indefinite noun-phrases.

Verbs use the nominal prefixes for subject congruence (and for owner congruence for the owner-subject-... paradigm), and the adjectival prefix series for object congruence (but for subject congruence in the owner-subject-... paradigm). A bunch of reductions and haplology is applied.

Indirect object marking only exists for masculine and feminine, and has its own series:
masc: -ek- -wak- -lak-
fem :-sak- -xak- -lak-
Mass nouns cannot be indirect objects.

Monday, May 12, 2014

Detail #87: A few participle-like things

In a participle-rich language, the following participle-types could imaginably exist:

1. A "made by"-participle
Let us say it's formed by -ekt. 'hammer-ekt tin', tin made by hammering it, 'burn-ekt brick', brick made by burning it, etc.

2. A "made by"-participle restricted to food-stuff.
grill-nepekt fish
deepfry-nepekt potatoes
dry-nepekt grapes

3. A "-ing"-participle restricted to food-stuff:
brew-neper man : the man that brews (beer)

4. A "made for"-participle:
endure-tert house
protect-tert mansion
confuse-tert puzzle

Can be extended to adjectives by verbing them:
strong-[verb derivation]-tert fabric: fabric designed for strength

All of these can be used as gerunds as well:
brew-neper in moderation complements all do-neper: brewing in moderation complements all cooking.

endure-tert is better than impress-tert
"being made for endurance is better than being made for impressing people"

Sunday, May 11, 2014

Detail #86: Interaction between the case system and the verb

Let us consider a language that uses the dative  to mark direction towards, and say, any one out of {genitive, accusative, partitive, locative} to mark location.

Let us also consider a verb system in place in the same language, where some kind of aspect distinction is made. For convenience, we call them imperfect and perfect. Whether it really is perfectivity, perfectness, telicity or something else is not that important, as long as it is something along those lines.

Now, let us imagine how this language will express location or direction: I find it pretty likely that direction and location merge in perfect contexts, or rather, direction is considered meaningless in the perfective, or at the very least sufficiently less meaningful to warrant a naked caseform marking it.

I go.perf town.dat = I am/was in town
I go.impf town.dat = I am heading towards town
I go.impf town.[the other case] = I walk in town

Thus, the dative alternates between locative and directional depending on the aspect of the verb.

Friday, May 2, 2014

Bryatesle: Verb Morphology

Bryatesle was somewhat inspired by Hebrew at the time, and therefore did not have any tenses. What it does have is an aspect system, although the marked aspects are more closely related to telicity than to perfect-imperfect or perfective-imperfective.

Both telic and atelic verbs showcase rather similar morphology:

ḑet, to run

1sg     et - diţat 
2sg     er - diţar
3sg     ei - diţai
3sg.n. er  - diţa
1pl     im - diţam
II pl  ine/diņe - diţane
III plh dies - diţanes
III pln ei/er - diţanei/diţa

The somewhat unrealistic morphophonemic dissimilation that causes some dental consonants to turn into postalveolar and vice versa in the vicinity of other similar sounds accounts for the change in the initial consonant in these words. The vowel changes are even less regular, but this example accounts for quite a large number of verbs. The left column holds the atelic forms, the right one the telic forms.

Bryatesle has three conjugations that all are pretty annoyingly huge, with a few verbs having an additional past tense form. These are sometimes coordinated with regular verbs to construct tense forms, but are far from mandatory.

The Imperative can be formed in a few different ways: attaching the suggestion marker morpheme to the 2nd person verb, for one, can attain such a function. The suggestion marker on the object can also have a similar effect. Finally, having a noun in the vocative case followed by a second or third person verb is often a way of constructing an optative or imperative sense, depending on the person of the verb. Finally, the object sometimes is in the exclamative in such sentences.

Other modal forms do not exist, but are attained by other types of markers, primarily determiners in noun phrases serve to mark various types of irrealis.