Thursday, April 30, 2015

Detail 155: Spicing up predicative possession

Let's assume a construction along the lines of English - i.e. a dedicated verb for predicative possession. Let's further assume it behaves very simply in the present tense. ('Present tense' could be changed to something else - imperfect(ive|oid|ish|oroid) aspect, indicative mood, definite object, non-specific forms of 'having', etc). Using English as a meta-language for how this would work we get for the basic present tense the following quite vanilla construction:
I have (a) car
However, to spice this up each noun has an associated preposition*. In other TAM/etc configurations, this prepositions surfaces. This serves to increase the amount of redundancy in the language, and is basically a congruence marker - and since it codes for two things [1) a type of noun class, 2) verb TAM] it might be rather an effective strategy with regards to increasing congruence. Now, there's several places this preposition could surface. The first, and obvious one, is with the possessed noun.
I had with a car
John had to an idea
You had by an apartment in the town, right?
Of course, we could have case:preposition:etc congruence behave oddly for these if we wanted to distinguish them from situations where we for some reason want to say that in collusion with a car, we had another object. Word order could of  course also help figure out which noun is the possessum in case several prepositional phrases occur if we want that to be a strategy here.

Next up is the idea of the preposition being independent of any NPs - either like a German separable prefix or like some kind of general adverbial. Not all that fascinating, but you can probably see it happening:
Erin had a boat by
John had a good life on
Prefixing it to the verb is the next option, giving us all kinds of byhad, onhad, withhad, tohad etc.
Finally, we could make the subject be affected by the preposition of the object! This is somewhat natural - the spatial way we 'have' things is somewhat affected by the nature of the object. (Of course, for some combinations of object and subject, the combination itself may affect the manner of possession - maybe the language takes account of some such combinations, maybe not.) In this case, the language almost gets a peculiarly located split-alignment for possession. At this point we then get:
by Erin had a boat
on John had a good life
to John had an idea
with me had a car
Historically, the classes of possession-types may generalize into expressions for manners of having things, so that by changing the preposition, we change the implied relation of the possessor and possessum. Such semantic differences being explicit only in some TAMs is not unknown - the Georgian verb paradigm is a pretty clear example of markers that do distinguish meaning only appearing in part of the paradigm.

Now we're being a bit boring though - this distinction only happens with one particular verb (or cluster of verbs derived from it, if we consider the preposition a derivational device). So, how about we extend it to a bunch of verbs that are somewhat 'possessive-like'. We could include verbs that include physically moving things (carry, drag along, pull?, push?), verbs of allegiance and such, and maybe some other things as well. Suddenly we've extended the system a bit - and obviously, this system could permit for derivation in the same way it previously did. Of course, the derivation option is not what we're primarily interested in - that's just a door we open for derived languages.

* Of course s/pre/post if you want to, but I warn you, you'll run into the word postdicative and the phrase postsent tense if you do that to this post.

Wednesday, April 29, 2015

Challenge: Do "Fluid-Inverse" Languages Exist? What Could We Do With Such a Typological Thing?

Inverse alignment is of course interesting by itself. We have verbs that are marked for whether the subject is higher or lower on some hierarchy (usually something like 1 > 2 > 3 ( 3 prox > 3 obv) > animates of various types > inanimates of various types).

Now, this mentions nothing of the situation with intransitive verbs! One could imagine a bunch of things going on there. (But at the moment, I have not had any good ideas, and hope to post a few tomorrow, maybe.)

[+Inverse, +Intransitive] could indicate non-volitionality - much like absolutive indicates non-volitionality in Fluid-S languages.

However, could we have it to indicate something else? If so, what'd be a reasonable thing for it to indicate?

Also, anyone who knows of such a language is welcome to inform me of it.

An Observation about Ćwarmin

Ćwarmin already has gotten one of these "problematic" things that tend to pop up in my conlangs. I am not much for the painstaking meticulous diachronic derivation of a conlang - although it certainly would be good for consistency's sake.

First look at the case and definiteness systems. The combination is full of intentional gaps. That is all fine and dandy, but I later came up with the idea that the definiteness system is the result of a previous proximative-obviative system that has been re-analysed. Makes sense that far, doesn't it. But ... morphologically it doesn't seem to make much sense - I don't really see any way a proximative or obviative marker would fail to combine with certain cases the way they've done in Ćwarmin. Nor does there really seem to be all that much logic to where the failures occur.

Well, I imagine there may be some way of salvaging it - I am not going to abandon it nor change these details - it's just that I basically have to say that "the hypothesis that Ćwarmin previously has been proximative-obviative is somewhat iffy given the facts, and we must hypothesize that several of its relatives instead have developed typical proximative-obviative systems independently while Ćwarmin itself has failed to entirely line up with such a system, and instead either developed some non-canonical variety or something slightly different from it."

On the other hand, flaws like these are appealing in their own way - they're a part of the artwork much like the various imperfections in the piece of wood that has been shaped into a small statue might contribute to its beauty.

Friday, April 24, 2015

Detail #154: Some Experiments with Inverse Alignment

Inverse alignment is a system wherein nouns are ordered on a hierarchy of animacy  or somesuch (often also person, such that i.e. 1 > 2 > proximative > obviative or similar, as in Ojibwe). Objects and subjects are not explicitly marked as such by congruence on the verb nor by case marking. Instead, verbs have a marker that can invert the order, so that the lower-ranked noun is subject instead of the higher-ranked one. Unlike a passive, this does not reduce transitivity.

However, could we have some neat other things going on? How about involving the indirect object in some way? It's quite natural that the indirect object will be higher ranked than the direct object; might we simply have two suffixes, "-INV1" and "-INV2", where INV1 switches subject and object, and INV2 switches indirect object and object.

Now, of course, this ends up somewhat incomplete. We can now generate the meanings that follow in English. Note that the examples are dumb, but it's pretty hard to come up with an example where three nouns that all could be objects or recipients or subjects occur. Let's assume a culture where kingdoms are legal entities distinct from kings, and where slavery is permitted so as to make sense of it all.

assumed hierarchy: the woman > a man > a kingdom (note, the = prox, a = obv in this example, and not the usual def. vs. indef)
the woman gave a man a kingdom (no inversion)
the kingdom gave a man a woman (INV1)
the woman gave a kingdom a man (INV2)
the kingdom gave the woman a man or the kingdom (INV1+INV2)

I'm making some odd assumptions here: 1) inverse switches the highest and lowest. This seems unnatural - it is more likely for the highest and next highest to switch places. Let's fix that. Now INV1 switches the two "highest", and INV2 switches the two lowest.
the woman gave a man a kingdom (no inversion)
a man gave the woman a kingdom (INV1)
the woman gave a kingdom a man (INV2) 

We still lack a few possible meanings. I don't like the idea of repeating operations like these - seems too much like some kind of algebra. An INV3 that switches the end-point arguments would probably be a bit better than repeated operations. We'd still lack one possible meaning:
the woman gave a man a kingdom (no inversion)
a kingdom gave a man the woman (INV3)
a man gave the woman a kingdom (INV1)
the woman gave a kingdom a man (INV2)
?a kingdom gave the woman a man (lacking)
?a man gave a kingdom the woman
Some languages do not really distinguish direct and indirect objects, so we could possible accept conflating some of them. Having all six possible permutations as independent morphemes seems somewhat boring. Combining the INV-suffixes could work - and it would be interesting to do some weird stuff where the combinations are not entirely straightforward.


Tuesday, April 21, 2015

Ćwarmin: Personal Pronouns - a final version

So, the Ćwarmin personal pronouns after some pruning and further design.

Notes:
The distant possessive subject is a case that is exclusive to the personal pronouns. 

Ćwarmin has gone through a rather interesting change in its typology: hints from cognate languages suggest that their shared proto-language had a proximate-obviative system. Remnants of this exist in Ćwarmin, but behave more like a switch-reference system, if only for just a handful of verbs. As the proximate-obviative system disintegrated, it was replaced by a specific-definite-indefinite three-way contrast. The definite signifies a referent known to both speaker and listener, the specific is only known to the speaker.

This also made its way into the pronoun system - in the singular, the third person distinguishes referents known to the speaker only from referents known to both. In part, this carries over from pronouns for proximate and obviative nouns. The distinction was lost in the plural, but survives in the nominative and accusative for the paucal. Now, on the other hand, the definiteness morphology is starting to appear on first person pronouns as a way of coding for clusivity in some dialects - definite first person plural is inclusive, specific first person plural exclusive. However, the morphology presented below is a more usual, classical Ćwarmin set of pronouns.

When stressed, the pronoun takes the bolded form. When unstressed, it agrees in vowel harmony with the following word, except over intonation phrase breaks - in which case it agrees in vowel harmony with the preceding word.

The third person singular definite is also the demonstrative used for a nearby object, so basically 'this' (with a slight tinge of 'that'). The morpheme i/u- can combine with IIIpl to form 'these' (with a slight tinge of 'those'), thus "i sit" or "u sut" gives the nominative, "i sić" or "u suć" the accusative, etc. It is seldom combined with the paucal. Ək/ak is the demonstrative used for a non-nearby object, so basically "that". It can combine likewise with the IIIpl for plural demonstratives, but the IIIpl without ək/ak basically serves as such a demonstrative in its own right as well.
For singular as well as plural nouns, i/u serves as the demonstrative determiner ('this noun'), whereas ək/ak serve as the demonstrative for singulars, sit/sut as the demonstrative for plurals.

Nota Bene: blogger hates letting tables go unmolested. It is very possible that new lines are deleted, and so on. I've been trying to get it to work properly for a while now and it just seems blogger can't stand not altering it. Thus, if, say, san and śen again are merged, you can tell them apart by one being bolded, the other not. The same happens in a bunch of other places throughout this table. There apparently is no way of circumventing this without adding actual extra rows of cells.

IsgIIsgIIIdefIIIspecIpcIIpcIIIdefpcIIIspecpcIplIIplIIIpl
nom



sanśənbec
bac
i
u
əkakdaldelranrəntawoktejəkmewok
mejək
marmercer
cor
sit
sut
acc



ataśətəśbacaś
becəś

təś
taś
daljaś
dəljeś
ranaś
 
reneś
waśjeśwoś
jəś
meś
maś
cereś
coroś
sić
suć
dist.
poss.
 subj



atak
ətək
bacan
becən 
inin unun teś
toś
dalun
dəlin
ranun
renin
--manak
mənək
cenek
conok
-
gen

anak
ənek
bacak
betək
ite
uta
teś
toś
daltudəltiranu
reni
tawun
tejin
-manak
mənək
cenek
conok
sitek
sutok
dat

aranś
ərənś
bacanś
bəcənś
in
un
tən
tan
donś
dənś
rənśronśwokśjikś-mankś
mənkś
cenkś
conkś
siteś
sutoś

gen.
abl



xaranś
źerənś
bacaś
 bəcəś 
ir
ar
ter
tar
doroś (no front form!)ronoś (no front form!) wośjeś-markś
mərkś
cerkś
corkś
sikeś
sukoś
inst



źerəp
xarap
barap
berəp
tap | təptap təpdaŕap
dəŕep
ranap
rənəp
tawap
tejəp
-manap
mənəp
canap
cənəp
sitəp
sutap
com-
to



xarkuś
źerkiś
backuś
 beckiś
itiśutuśtetiś
totuś
dalmaku dəlməkirammaku remməkitawaku
tejəki
-maruś
məriś
coruś
cereś

sitiś
sutuś
com-with

santuc
sentic
batuc
becic
itic
utuc
tetictotucdalmac
dəlmec
rammac
rəmməc
tawac
tejəc
-maruc
meric
coruc
ceric
setec
sotoc
neg


źerə
xaro
bicə
buco
istə
usto
tistə
tusto
daltus
dəltis
rono
rənə
tenə
tono
-marto
mərtə
carto
certə
settə
sotto

Saturday, April 18, 2015

Dairwueh: Diminutives

Dairwueh too has morphological diminutives. There are several ways they are formed, and they almost all involve non-concatenative morphology. Many diminutives can be somewhat ambiguous: they may both encode 'a small X' as well as a more specific derived meaning. Examples of the latter are sea:lake, skill:trick. The more specific meaning is given below if such a meaning exists for the pair given.

1. R-infixation
Hiatuses inside words can be replaced by -r-. Generally the previous syllable's onset is made heavier as well if possible:
baud: barud, farm, little farm
apaeg: appareg, servant, kid servant
nius: nnerus,  sea, lake
If no hiatus exists, an -r- infix instead goes after the onset of the second syllable, except if there is a cluster there, so
bakuse: bikruse: tree, bush
kartun: -no -r infixed form-! celebration
2. Vowel shifts
e → i, a  → e, u → ii, i → ii, i → e, o → u, a → u. Which shift takes place depends a bit on surrounding vowels, on the declension of the noun. In some words, they're quite freely applied, mainly to the first syllable, but sometimes to additional ones. In words with vowel shifts, either r-infixation or a suffix are also applied. The suffix generally is -er- (masc), -i- (fem), -en- (neut).
stanar : stenar hut, little hut
kartun: kertun, kertin celebration, smaller or less formal celebration
3.  p- and t-fricativization and lengthening
aper : affer cloud, wind
vutin : vussin  sail, cloth (humorous)
kapan : keffan book, letter
This is mainly applied on the onset of the second syllable. 
4. di- prefixation 
rasma : diresma a (largeish) fire, a small fire
dunvali : dirinvali kingdom, tributary kingdom (the change to -rinvali is due to dissimilation)
muste : dimuste skill, trick
Diminutives are generally used to showcase informality. They can be quite freely formed along the lines given above. Ones that have specific meanings may be avoided in favour of newly formed ones that lack such connotations, but one cannot rely on, say 'keffan' signifying a letter all the time - sometimes it is indeed just a booklet.

Friday, April 17, 2015

Detail #153: Passives, English-style Perfect and Person

Imagine a language in which some type of past tense or perfect aspect or such is formed along the lines of the English perfect - but only for first and second person. The third person, however, forms the passive along the same lines. An active third person perfect (or whatever) is formed by demoting the (semantic) patient to some kind of oblique status, and have the agent as subject.

This could be cooler if there further was some animacy split or something: animate third persons don't have the passivization happen like this, inanimates do.

Wednesday, April 15, 2015

A Challenge: Split Ergativity Grammaticalization Path (with added twist)

Normally, it seems grammaticalization that leads to split ergativity tends to produce a split where past|perfect|perfective|etc... stuff has an ergative alignment. What would be a reasonable grammaticalization path for the inverse situation to obtain, i.e. nominative-accusative for past|perfect|perfective|etc..., ergative-accusative in non-past|imperfect|imperfective|im-etc situations?

Days of the Week and the Barxáw Calendar

The Barxáw week does not have a seven day cycle; it does not even have a reasonably short cycle, although it has some subcycles. The repeating bit goes like the schematic below, underlines inserted where the non-cyclical element goes:
___, wík bir, tàx bir, ŋùx bir (but also ___), fát bir, tabí bir, nèm bir, ___
Four of these form a cycle. The initial day has different names in each subcycle: (bán) bán bir, wík bán (bir), tàx bán (bir), sòq bàn (bir) - the form essentially being "(one) one day, two one day, three one day, last one day". The last day of the subcycle is bán sòq (bir), wík sòq (bir), tàx sòq (bir), sòq (bir) or sòq sòq (bir) - the form essentially being "one last day, two last day, three last day, last last day". The fourth ŋùx bir is generally by lùjun ìsta bir ('big house day'), due to religious celebrations that usually occur on this day.

Days take no counters, and usually are not marked with prepositions or anything, except in the following construction: to specify which X bir in this cycle one is referring to, the first number before the initial day of that week can be presented after ðo. So, day three week three is tàx bir ðo tàx.

A cycle of four 'weeks' is called a 'bùcə', and the week itself is a 'dàqáw'. 
The cycles are numbered - there's years of 11 of them, interspersed with years of 12 (but the average length is 371, roughly the length of the year on the relevant planet). Different regions have different naming conventions for the 'months' which is why numbers are widely used in writing. The first month is generally the first month after harvest. 

Further, there are a number of specific additional names for particular months' lùjun ìsta bir and also for a few particular sòq sòq bir, as well as the first bán bán bir of the year. Local names vary.

Finally, the last lùjun ìsta bir of the year is sòq lùjun ìsta (bir), which is the largest celebration of the year.

Tuesday, April 14, 2015

Detail #152: Transitivity and Inanimates

There is a universal that says that object marking that appears on inanimates is more likely than not also to appear on animate objects. The idea that occurred to me for this post originally violated that universal a fair bit, but in an interesting manner.

Consider transitivity and how one can reduce transitivity by demoting an object to oblique status. Somehow, inanimates easily could be considered less 'genuine' participants in an action, and therefore, for instance, mark them by some preposition. We're still really talking about objects though, and what this would amount to would be inanimates marked by prepositions, and animates marked by at most a case suffix. 

This seems a bit unexpected. The opposite - animates being marked by a preposition, inanimates not at all - is attested. 

However, what if animates triggered a verbal agreement thing, which inanimates did not trigger? In that case, the preposition thing seems like less of a stretch - especially if the animate objects also had some case marking. 

So, what happens to other things? Passives - promote animate objects to subjects, leave inanimate oblique objects as oblique objects of an impersonal verb. Antipassives - promote subjects to intransitive subjects, demote animate objects to obliques, leave inanimate objects as obliques.


Monday, April 13, 2015

Dairwueh: Imperatives, Participles and Gerunds

Previously, the finite verb has been detailed in this post. In it, I referred to a future post detailing the non-finite verbs of Dairwueh as well as its imperative.

For verbs that are bisyllabic or longer, the imperative is identical to the IIIsg form. For monosyllabic verb stems, it generally has a short syllabic core. Thus taan: tan ('say'), don: don ('look). A few verbs have a vowel change involved - reeb: rib (tear), skoon: sken (run), faarg:ferg ('shut up'). In the singular, thus, some verbs conflate the imperative and the IIIsg, while some other verbs conflate the IIIsg and IIIsg2,but distinguish the imperative from the IIIsg. In the plural, the IIIsg form gets an -u suffix, so tanu, donu, reebu, skoonu, faargu.

The plural is sometimes also used to address the 1st person plural, and can then take an explicit subject pronoun.

Dairwueh has a small set of participles. A passive irrealis participle, negative and affirmative present and past active participles, and negative passive participles.

active
affirmative
present -un
past -ar
irrealis -umuš
negative
present -šun
past -eyš
passive
affirmative -šəŋ
negative e-___-šor
irrealis
 e-___-šis
 They tend to show less case morphology than other nouns, and conflate, for instance, accusative and nominative quite widely. The locative-instrumental can also be omitted quite freely. In fact, the accusative and locative-instrumental are only used as distinct case forms on them when they are used as designations of professions, e.g. partetun = baker, 'baking'. For use as a gerund, they normally take neuter morphology, and thus affix -e in the nominative, -e in the accusative, and so on.
The irrealis is basically an active form of '-able', i.e. someone who can do a thing. Passive irrealis likewise mainly signifies '-able'. 

Case congruence is somewhat simplified compared to other adjectives - essentially, when used as adjectival attributes they distinguish nominative from 'everything else' (except with neuters, where it's nominative-accusative from everything else). When used as complements, they behave as described above - conflating cases quite strongly (but dative and genitive are in fact distinguished).

The genitive is partially worth mentioning separately, as a participle as an adverb often is marked by that case if it pertains to the opinion of the speaker or if the speaker wants to attract attention to it.

The affirmative passive participle also combines with adjectives to form adjectives meaning 'who has been forced or turned into X (without voluntariness)'. Further, it is used to form ordinals from the cardinal numbers.

The present negative active participle also can go on 'integral parts' to mark their being missing, i.e.
sarišun 'handless' (or rather 'one-handed', lacking both hands would be sartašun)
kolšun 'legless' (as above, with koldašun for entirely legless)
pulšun 'hairless'
-eyš is only used in a similar construction for just a few particular nouns; nartereyš is a word signifying a man that now has reached adolescence. 'previously beardless' is its literal translation.



Sunday, April 12, 2015

A Rant about Lojban (not mine)

Although this rant is on a forum and I don't know how clever linking forum posts is, I must say I found much to recommend in Sumelic's rant about Lojban. Certainly Lojban was not designed to be a good auxiliary language (although apparently some seem to want it to be one), and thus its flaws in those regards are acceptable. However, some who try to make good international auxiliary languages seem to replicate some of its flaws. The logical language movement seems to fail to realize that communicative needs are shaped by the real world we live in, an evolutionary algorithms are pretty good at identifying what types of optimizations are beneficial. Essentially, linguistic evolution creates useful shorthands (which you can see an example of missing from Lojban in the point named 'one last thing').

Detail #151: Differential Object Marking and Adjectives as Verbs

Let us consider a language where adjectives mainly are intransitive verbs, and can be coordinated with another verb instead of being placed in a position where they are 'close' to their head noun:

bar-e veed nem-dag
big-* man run-3sg
veed nem-dag u bar-dag
man run-3sg COORD big-3sg
* is a marker that goes on any adjective that immediately precedes its noun or any other adjective that precedes its noun and so on. An 'unpredicate' marker to some extent.

But let's add a little complication. Let's have two object cases - first, a classical accusative, and for now we'll let -as do as its marker. For the other, we take the genitive, which in this language is -asi, just so we have something to work with.

bar-e veed kaud-dag stour-as
big-* man  fight-3SG bear-ACC
veed kaud-dag stour-as u bar-dag
man  fight-3SG bear-ACC and big-3SG
 both signify that it's a large man fighting the bear
Now, however, we might want to focus on the other fighter! And here's where the genitive case is used on objects.
veed kaud-dag stour-asi u bar-dag
man fight-3SG bear-GEN and big-3SG
the man fights the big bear
Generally, such separated adjective-verbs would go first, as emphasized adjectives in a sense. Thus, this serves as a change to connect such adjectives to the object (which is probably the second most likely constituent for which one would want to do such a construction). Of course, the adjective-verbs could have TAM markers that show that these verbs are results of the action:
veed lam-dag tist-asi u karn-em-dag
man color-3SG house-GEN red-INCEPTIVE-3SG
the man painted the house and it turned red
'the man painted the house red'

Ćwarmin: Deriving Nouns from Verbs

Ćwarmin has a few strategies for deriving nouns from verbs, see a previous post mentioning four such affixes.

Instruments can be derived from verbs by the suffix -(e)kye/-(o)kvo, which is appended to the verb root. However, this is not particularly productive any longer, although a bunch of such nouns are common, i.e.
sabokvo (from sab-, sabam, sabjul) - a duvet or blanket or cloth (from v. cover)
ripekye (from rip-, ripen, ripəm, ripjig) - scissors, (the verb meaning 'bite')
However, a more popular formation currently is simply to take the augmentative or diminutive prefixes har-/hər- or sir-/sur- and prefix them to the verb root.
sirsab (from sab-) - a lid
harsab (from sab-) - a roof
hərrip - (from rip-) - leghold traps
 Generally, things that are the size of two palms or thereabout or smaller take sir/sur-, other things take har/hər-. -okvo/-ekye can also be affixed to a noun, and then indicates a thing or person equipped with the thing marked by the suffix.
er ś(e)n-okvo - onehanded
one hand-instr
Numbers can also take this marking. It then signifies 'a thing that suffices for NUM amount', so i.e.

sikrekye = enough for three, a thing that suffices for three
dustokvo = enough for ten, a thing that suffices for ten
These can be used as adjectives or as adverbs, and even as nouns in their own right. As adjectives or adverbs, they usually take no case marking whatsoever (or alternatively,

For agents, the suffixes -ed/-ad, often in combination with a diminutive or augmentative serves the same role. However, it can be affixed to nouns as well, and then signifies someone who carries out a typical verb with regards to the object, i.e.
mauŋ-ed
food-er
'foodie' or 'eater', depending on context
 A set of more typical derivations would include
okkaulad - resistance fighter
itred - drunkard
rakad - reader
ragad - speaker
hacjad - thinker
The various distinctions that can be expressed by different infinitive suffixes are not carried over, but for some verbs, the -j- in the -jul or -jig infinitive suffixes may for some reason carry over.

With adjectives, -ed/-ad signifies someone distinguished by the quality the adjective marks, i.e. 'an X one', i.e.
samar-ad
fast-AGENT
a fast one
On numbers, it can also signify a set of such a cardinality:
sikred = trio, troika
mered = duo, pair
ered = solo, unity
ambad = a double quartet
Saying that a woman is sikred or mered and so on is also possible, and then indicates that she has given birth to that number of children.

Patients are formed from the passive -aśp|-əśp, with a diminutive suffix, usually the gar/yər- for inanimates, and others for animates, or sir-/sur- for animates. These can serve both as nouns and adjectives. A fossilized set of nouns survive with the former method of affixing -ad|-ed to -aśp|-eśp, giving -aśfad|-eśfed.

Saturday, April 11, 2015

Ćwarmin: Diminutives and Augmentatives

Ćwarmin has very little prefixing morphology, but some does exist. The diminutives and augmentatives are formed this way. Ćwarmin uses diminutives and augmentatives extensively, and quite weird idiomatic constructions appear.

gar-/yər-, wor-/ver-, har-/hər-, ka:-, ron- : augmentative, "big"
sir-/sur-, cir-/cur-, mir-/mur-, dži-, lis-: diminutive, "small"
Note that sir- is much preferred for diminutives. Sometimes, this ignores vowel harmony - curkaparsirkapar, 'small shelter'. However, it sometimes enforces front harmony onto lexical roots: sirkəpər: a village, 'small shelter'. More uncommon is vowel harmony switching to back harmony, but it does happen.

Lis- and har-/hər- are the 'weakest' of these, i.e. the deviation from average size or intensity is not usually very large when those are used. However, they are also the most 'emotionally positive' ones, and prefixing lis- to a female name or har-/ər- to a male name is done as a form of endearment. (A gender difference is that men usually do not use har-/ər- about other people, whereas females use both lis- and har-/ər- quite commonly.) For children, both dži- and lis- are common; ron- is sometimes used in a playful sense with children.

Combining prefixes is common:


verhərgimin : the largest moderately large fish, a rather large fish
curworkar : big small who? 'whatshisface', or 'a person whose existence is under doubt'.
cirverter : big small what? 'a thing whose existence is under doubt', or 'whatchamacallit'
garkaaćwurga = very huge storm
 curronguska = tiny big chief, a demeaning term for a person who thinks he is more important than he is
(dži)ronguska = little big chief (to a kid, indicating he's being bossy) 
gar-/yər-, wor-/ver- also serve to intensify verbs and adjectives, and mir-/mur- serve to weaken them. However, these should not be mistaken for comparatives or superlatives. Often, the intensifying prefix may go on a particularly salient noun or adjective in the same noun phrase even if another noun or adjective is the actually intense thing, i.e.
samar ver-širməs
fast big-ship
very fast ship, fast big ship
wor-samar širmes
very fast ship 
Intonation serves to distinguish the two; in the case that the prefix really intensifies a preceding word, the first syllable generally ascends, with the next syllable dropping a fair bit. If it intensifies the word on which it is prefixed, the first syllable of the root generally ascends a bit.

Finally, the plural indefinite in combination with the augmentative indicates 'X and things associated with X'.

Wednesday, April 8, 2015

Detail #150: Word-level diacritics

Diacritics that encode prosodical information and tell us about the intonational patterns over larger units than syllables is a thing I have never seen in a conlang or a real language for that matter.

Could be fairly interesting.

Tuesday, April 7, 2015

Some Ćwarmin Interrogative and Indefinite Pronouns (and some clitics)

Since new readers probably drop by every now and then, I guess the time to remind everyone of Ye Merrye Conlangre's treatment of indefinite pronoun typology has come around again.

And then we get the time for looking at Ćwarmin's interrogative and indefinite pronouns.

nomkarter
nom complkar(ca)ter(cə)
acckaracter
acc complkarcoterce
genkartośterəś
datkarna

terne
ablkarar
terer
instrkarapterəp
com-tokorkuterki
com-withkortuterki
negativekurstateśtə
interrogative pronouns lack a reciprocally possessed case
Kar corresponds to 'who', 'ter' to what. As for specific meanings, 'terəp' sometimes means something like 'how', whereas 'terce' means something like 'like what' or 'how (of manner of appearance)'. Terce is sometimes used as a complement to subjects as well. The locative forms are formed along the description given here, i.e. from the dat, abl and accusative forms.

'Tercə' is exclusively used when asking what something consists of - in other positions that syntactically correspond to the nominative complement case, 'ter' is used.

-(e|a)st is a fairly common suffix that denotes 'even, too, also, and'. However, with the interrogative pronouns above, it means 'someone, anyone' and corresponds to 1,2,3,4,5,8,9 in the typology linked to. The addition of -(e|a)st does however further conflate acc compl with acc, nom with nom compl (for kar), and genitive with accusative for both.

Direct negation and indirect negation is obtained by inserting a suffix after kar/ter, viz. '-mo|-mə-', thus giving:
karmo, karmoca, karmoc, karmoc, karmoś, karmona, karmor, karmowap, karmoku, karmotu, karmosta
termə, termə, termə, termə, terəməś, terməne, terməər, terməəp, terməki, terməki, terməśtə
 -ma|-mə also are used on adjectives to convey negation of a quality, and can attach to verbs to negate them as well.

-ada|-ədə is another common clitic that indicates 'however, but, so, albeit, ...'. It goes after the first word of a clause or the verb. However, it can also go on an interrogative pronoun and then indicates an indefinite pronoun of type 3,5,8,9 in the typology of the link. It is suffixed like -(e|a)st.

-sak|-sək further indicates 2,3 and is otherwise used to mark 'maybe, potentially, hopefully'.

Monday, April 6, 2015

Possession in Tatediem

Pronominal possession has already been covered. More general nominal possession is based on this, and therefore, reading about pronominal possession before reading this is recommended.

Generally speaking, when possession is marked on the possessum, the possessum is always marked for definiteness. The possessor is placed to the right of the possessum, and the object prefix corresponding to the possessor's gender and number is inserted between the definite gender prefix and the noun root in the possessum. The possessor is unmarked.

There is no mandatory distinction between reflexive possession and other third person pronominal possession. Tatediem, however, has reflexive object and reflexive indirect object prefixes for the verb, and sometimes the presence of the reflexive indirect object marker serves to mark that the object is reflexively possessed, in order to disambiguate. However, additional ambiguities are created that way.

The dialects that use -páhí or -gìan do not generally differentiate reflexive possession from regular third person possession. 

As for predicative possession, -páhí and cognates are used in almost all dialects to express 'X has Y'. However, some dialects permit two ways of expressing it: subj-obj-páhí and subj-dative-páhí - in the latter, the subject is the possessed thing, in the previous it is the object (mainly, dialects permit subj-obj-páhí). The verb -gìan that is present in most dialects only permits a construction with possessor as (indirect, if possible) object, and the possessum as subject.

In dialects that have both, -páhí is more likely to be used to express "x has y", whereas -gìan is used to express "y is x's". However, both verbs can be used for both meanings, and dialects that only have one of them naturally uses that one for both meanings. Word order adjustment - such as fronting one of the nouns -, and prosody serve to distinguish the two uses.

Sunday, April 5, 2015

Possession in Bryatesle

Unlike Ćwarmin (and Dairwueh), Bryatesle primarily has possessive head marking. Still, the possessor also tends to be in some non-nominative case (however, the more animate and definite the possessor is, the more likely it is to appear in the nominative). There are a variety of strategies used in Bryatesle that might be worth looking at.

1) Intransitive verbs turn pseudo-transitive, possessor into pseudo-subject
The possessor turns into a pseudo-subject with which the verb agrees, the possessum retains subject marking (but does not trigger congruence on the verb). The possessum is head-marked for possession. There's a slight probability for explicit object marking on such possessums, despite their essentially being intransitive subjects. The possessor may be marked with the secondary subject marking.

2) Possessor to the left of the possessum
The possessum has the possessum-marking, the case marking on the possessor is somewhat arbitrary. Proper nouns tend to be nominative, some nouns have a preferrence for some case, but this 'preferrence' goes both ways - both the possessor and possessee nouns may affect which case will be used. Some amount of displacement for the possessor is also possible - the fewer the arguments in the clause, the more likely such a displacement is. Such displacement is generally at most a word or two in some direction - or to sentence-initial position. The possessor should not interact syntactically with anything else, such as adpositions or anything.

3) Possessum carries possessum-marking, no obvious possessor
In a sentence with multiple arguments, there is a hierarchy:
subject > indirect object > instrumental = direct object > locatives > other obliques
If a noun that is instrumental in the sentence has possessum marking, it can be owned by an indirect object or a subject (even a pro-dropped subject is eligible). Likewise, locations can be owned by the direct object, the instrumental etc. A subject can be owned by a direct object, however, or by a subject or object of the previous clause. The subject, and direct and indirect objects of a matrix clause are the only constituents that can be owners of any argument of a subordinate clause. Naturally, the likelihood that the owner is high up in the hierarchy is also rather large - and gendered pronouns may be added to help disambiguate. These generally are in the dative case. The presence of such a pronoun generally removes the chance that the possessor is the subject as well. Definiteness (and demonstrative adjectives) increases the likelihood that a noun is the possessor. Secondary subjectness and reciprocal objectness may also affect the likelihood that some noun is the possessor (i.e. 2nd subjectness increases the likelihood for a non-subject to be the owenr, reciprocal object marking increases the likelihood that a noun that syntactically is in the same position is the owner, i.e. a noun that is also coordinated with its possessum.)

Thus, reflexive possession - although with a slight ambiguity as to which "higher up" noun is the possessor - is the default assumption in Bryatesle. Non-reflexive possession generally is obtained by extra pronouns.

4) First and Second Person Pronouns
The first and second person have possessive forms of their pronouns.
1 sg nere
2 sg tere
1 pl vele
2 pl xene
The -e part marks congruence for nominative masculine singular marker, and is replaced with the adjectival congruence markers for the relevant case. The possessive pronoun goes after its head noun. Third person forms masc: mezer, fem: ezer, neut: syn appear in some dialects a few centuries after the form of Bryatesle that I have primarily described. These do not trigger the presence of head marking. However, use of them is quite infrequent, as other strategies are preferred, especially whenever a first or second person subject is involved.

Saturday, April 4, 2015

Possession in Ćwarmin

Since I've recently been thinking a it about reflexive possession, I figured I might as well present a survey of the way my main conlangs deal with it. This will also be a short overview of possession strategies in general in these languages. The series of posts will end with a post that compares and contrasts their constructions. First out is Ćwarmin.

Ćwarmin

Ćwarmin possession is generally marked by use of the genitive on the possessor. The possessum can appear in any degree of definiteness (depending, of course, on the available degrees of definiteness for the case in which the possessum stands):
indef:
źaŋk-ututa girśim
fisherman-GEN.DEF net
the fisherman'snet
A(ny) net of the fisherman's

def:
źaŋk-ututa girś-iti
fisherman-GEN.DEF net-SG.NOM.DEF
the fisherman'snet
The fisherman's net, that net of the fisherman's

spec:
źaŋk-ututa girś-itək
fisherman-GEN.DEF net-NOM.SG.SPEC
the fisherman'snet
A specific net of the fisherman's
For plural forms, definite tends to be 'exhaustive' - all the X of Y; specific tends to signal exhaustive with regards to some limited set - i.e. 'all the X:s that belong to Y among these X:s'.

Indefinites as possessors are possible, but there is no distinct genitive form for them. There is some variation to what case an indefinite possessor goes in - nominative, accusative and dative all are fairly common, general ablative a bit less common but common enough to be understood by most speakers.

As for pronominal possession, most languages related to Ćwarmin have a set of possessive suffixes. However, these have been lost in Ćwarmin, and only survive in these few nouns:
sidestigə - 'my child' (a vocative phrase uttered by clergymen)
niźilgə - 'my love'
sarbatuŋra - our obligation (a phrase uttered in many liturgies)
midreviŋrə - our praise (again, uttered in many liturgies)
Of these, only nigilgə and sarbatuŋra are attested in other cases than nominative - nigilgə as nigilgəmcə (nom.compl) and sarbatuŋra as sarbatuŋrawuc (acc), sabratuŋrutćo (acc. compl), sarbatuŋrumca (nom. compl).

The pronouns have genitive forms, which are used to express possession. However, with direct and indirect objects that are possessed by the subject, the reflexive object case is used. Historically, it originated as a third person possessive suffix that was later reinterpreted, and now has nothing with person to do - only with reflexive possession. However, there are a few exceptions.

Across the conjunctions 'i', 'e', 'u' (and), it generally binds to the noun to the left of the conjunction unless it too is marked with the reflexive possession case, thus:


Egen e kamu-sunEgen and wife-rflxposs
Egen and his wife (Egen is a fairly common name)
This construction is possible regardless of the case of the leftmost noun (except of course reflexive possession), but it is most usual with the nominative, accusative and dative. Another construction that also occurs differs by its lack of a conjunction - Egen kamu-sun - with subjects and sometimes objects. Then it marks a more comitative-like meaning: Egen, with his wife|accompanied by his wife|...

In a few dialects - all of which retain the possessive suffixes - the reflexive possessive also can be used as subject of embedded clauses; in those, however, it often also is more clearly a possessive suffix and does combine to some extent with actual cases, although it combines with fewer case markers than other possessive suffixes.

As it happens, the third person genitive pronouns do not differentiate whether the referent is definite or specific (unlike how the pronouns do in several other cases), which means that for arguments that are not subject, object or indirect object, there is very little in ways of distinguishing different third-person possessors.

Friday, April 3, 2015

Detail #149: An Orthographic Thingy

Imagine a conlang with an orthography comparable in depth to that of English. The kicker? The language also has Turkic-style vowel harmony.

Wednesday, April 1, 2015

Some Considerations: Reflexive Possessives and Scope (and Lexical Effects)

Consider reflexive possessive pronouns. These exist in Scandinavian and Slavic languages, and very probably elsewhere as well. English does not have them, c.f.

a man1 should teach his1 son a trade.
John1 bought his2 car yesterday.
John1 bought his1 car yesterday.
However, in English, intonation can differentiate the two, and addition of 'own' also serves to disambiguate. However, some languages have separate words:
En man borde lära sin son ett yrke.
John köpte hans bil igår.
John köpte sin bil igår.
However, in Swedish, sin is restricted to non-subjects. You cannot say
Erik och sin fru for iväg till Amerika igår.
Erik and rflx's wife went away to America yesterday.
Erik and his wife left for America yesterday.
And scope prevents you from saying something like
*Hilma hoppas att sitt hus ska stiga i värde.
Hilma hopes  that rflx's house shall rise in value.
Hilma hopes that her house shall rise in value.
Instead, the regular possessive is used:
Hilma hoppas att hennes hus ska stiga i värde. Erik och hans fru for iväg till Amerika igår.
An idea that struck me is that maybe certain nouns in a language with restrictions such as those in Swedish still might have some laxity permitted for some noun-pairs? Husband and wife, parent and child, friend and friend, and various other social pairs, especially? Maybe such noun pairs even transcend the subclause boundary showcased above?

Or maybe there is a hierarchy, where some nouns even can possess from subordinate positions in the clause - i.e. if parents were higher than children in this hierarchy:
rflx's children honour the father
his1 children honour the father1 
This would be highly bizarre in Swedish, but could be cool in a conlang.