Monday, December 31, 2012

Detail #20: Complements and implicit case

A hierarchy-like thing where nouns, when operating as complements assume implicit case, so e.g.

I am town = I am in town
He becomes town = He goes to town 
This would probably interact with nouns and verbs in various complicated ways, and could possibly fit well with an isolating language, so I may include it in the isolating language I am planning. This differs from topic-comment in some ways, which I should probably explain in greater detail but this'll be enough for tonight.

Notice that this should only go for complements, and not for adverbs, in the language in question.

Thursday, December 20, 2012

A longer-term project: Bringing some ideas together

Few conlangers (except engelangers and auxlangers) ever make isolating languages. I surmise the reason for this is that isolating languages are felt to have less of a potential for things to develop - the lack of morphology feels stifling, and all isolating languages are felt to be grammatically the same as Mandarin.

I have long intended to make an isolating language that differs from Mandarin and such without being too much like English either. So let me start out.

These features I have already mentioned will be included:

Seems like a set of features that could combine nicely and provide for a rich and detailed grammar.

Wednesday, December 19, 2012

Detail #19: Instrumentals and Possessives

Many languages express X has Y somewhat along the lines of With X is Y. What case or adposition appears with X is language-specific, and can be locative or comitative or anything. Both Finnish and Russian use (mainly) locative expressions.

Some languages mark the object as some kind of oblique, though. English has a phrase that is somewhat similar to it:
She is with child.
Now, that is a rather special case, but illustrates the kind of marking I am discussing here. wals.info informs that this type of construction actually appears. Personally, I find some kind of instrumental marking to be very suitable for this (except maybe in certain specific types of phrases or under some circumstances - restricting this to non-human objects seems rather reasonable, for one, or some similar cultural restriction).
is.1sg  cup.instr - "I have a/the cup"
is.3pl dogs.instr = "they have dogs"
Furthermore, I find it reasonable that the instrumental case could be the case marking for any possessed noun -  the instrumental case doubling as head marking in possessive constructions. Why instrumental? Well, there seems to be a vague correlation between using a thing as a tool and owning it.

The idea by which it becomes evident that I am a one-trick pony is this: reflexive possession of objects marked by the noun object being marked by the instrumental case.
He drive.past.3sg car.instr = he drove his own car
He drive.past.3sg car.acc = he drove a/the car 
Alternatively, the distinction could work like this:
He drive.past.3sg his car.instr = he drove his own car

EDIT: reading through the earlier posts, seems this is fairly close to another post in content. Meh. It is an ok rephrasing that may be easier to understand for people not that familiar with ergativity and such. Well, I will leave it behind and try to avoid similar embarrassment in the future.
He drive.past.3sg his car.acc = he drove his (someone else's) car

Tuesday, December 18, 2012

Detail #18: Weird things to do with adverbs

In some languages where adjectives are verb-like, let adjective-verbs form auxiliaries that mark all kinds of manner. Even permit numbers to form auxiliaries that mark "to do n times", or "to do the n:th time" or "to do to N objects" (in effect extracting the number from the object NP to the VP).

When many adverbs are called for, some may remain chained with the main verb, and if several are highly salient they can be coordinated with a shared complement VP along these lines:
John quicks and carelesses stack[infinitive] the books
John carelesses stack the books and quick
John carelesses stack the books quick
John stacks the books, carelesses and quicks
If the adverbial has little to do with the subject, they may take a dummy subject:
John's house burned down, [dummy] unfortunates. 
Maybe a causative would be nice there?
[dummy] unfortunate.causative.3sg John's house burned down.
Would maybe work best in a language where infinitives can mark subject agreement? Alternatively, a language where there is some but rather sparse verb morphology, e.g. something along the system in standard Swedish with an explicitly marked present tense distinct from the infinitive.

As for numbers, some pronouns could also work:
I many.1sg have ideas - I have many ideas, "I profusely have ideas"
You howmany.2sg run laps? - You ran how many laps?
They how.3pl.past? - How did they do it?

 On the other hand, if done badly - without making a really thought-through morphosyntax here, it would easily end up a silly way of marking things in the wrong place.

Detail #17: Marking for Atypical Arguments

Consider a verb such as to eat, and consider some possible arguments: to eat food, to eat cheese, to eat dust, to eat a lot.

In Hungarian, verbs mark whether the object is definite or indefinite (although there are further complications: 1st and 2nd person objects trigger indefinite verbs). The indefinite conjugation is also present on intransitive verbs.

I was thinking what else one could do with verbs, and having typical objects vs. atypical objects could maybe be a reasonable thing to mark. Which one is more likely to take a marking? Somehow, atypical objects feel like the more likely to trigger a marking here, as they're less likely to occur.

An atypical object would be anything outside the usual scope of objects for the verb, which clearly makes this a rather culturally influenced thing. In the examples above, eat dust would probably have the unusual object marking if meant literally. Of course, that particular idiom would probably not exist in the language, but similar considerations apply.

However, what if the same verbal marking applies to any salient argument - not necessarily only objects?

Saturday, December 15, 2012

Detail #16: Wackernagel Adpositions

This must exist in some language, for sure. The Wackernagel position is the second spot in a phrase (and according to Newmeyer's Possible and Probable Languages a position beyond which human grammar cannot count - you can have things go in the first position, in the second position, but never a rule that says that something has to go in the third position). If we changed the rules of English so that adpositions went in the Wackernagel position it would make this paragraph come out like this:
This must exist some in language, sure for. The Wackernagel position is the second spot an in phrase (and according Newmeyer's to Possible and Probably Languages a position which beyond human grammar cannot count - you can have things go the in first position, the in second position, but never a rule that says that something has to go the in third position. If we changed the rules English of so that adpositions went the in Wackernagel position would make this paragraph come out this like.

Such a system could easily be a special case in some language, i.e. only under some circumstances is this enforced. Maybe only a subset of all adpositions go in the Wackernagel position, or only when the adpositional phrases are arguments of the verb rather than adverbials, or maybe the adposition is made a preposition (or postposition) if it's especially salient, e.g. correcting which adposition was said or somesuch.

Friday, December 14, 2012

Detail #15: Non-trivial Verbal Discongruence

Discongruence is a thing that really interests me. Congruence, of course, is a rather obvious way of increasing redundancy in a language. Marking gender on verbs or adjectives, case on adjectives as well as nouns, gender on both adjectives and nouns [...] are all ways of having some information appear twice or having some extra information appear that may help in case noise appears along the line (which it will).

Some languages have weird discongruences though, such as the Semitic inverse gender marking with some numbers. Some languages have a bunch of set phrases (or half-set, as the construction type may be somewhat productive): Finnish has a bunch of adjective + noun phrases where the adjective is in nearly any of the spatial + role-related cases and the noun is in either the partitive or the instructive (= instrumental) case. Pitkäksi aikaa = long.translative time.part = for a long time, tällä tavoin = this.allative manner.instrument = by this manner, in this manner, like this (adverbial).

Further, some languages permit some discongruence in the verb phrase, like having the verb sometimes be third-person singular rather than the proper, expected person marking (even when not counting things like quirky-case subjects not triggering congruence for their person, or whatever along those lines). I find it likely most languages that do stuff like that tend to permit third person verb marking for non-third person subjects, but how about a language where this is not the case - where any person can get any person's verb marking under some circumstances, like, some lexically determined verbs that just *permit that kind of thing* or with certain phrases or whatnot? (Note: English dialects where the congruence system doesn't line up with standard English verb morphology are not an example.) Preferrably, the exceptional markings should occur seldom enough that they don't affect the default interpretation of the suffixes, but often enough that they are learned and could be part of the language for a wide speech community.


Wednesday, December 5, 2012

Detail #14: Suppletive auxiliary-based agreement

This idea is kind of a snapshot-in-language-history kind of idea, as I find it very unlikely it would remain for long.

A constructed isolating language where certain auxiliaries suppletively are changed to other ones depending on the class of the noun. Noun class is otherwise only marked by pronouns. Other than that, the noun class system has been losing ground in the language. Certain semantic differences between the suppletive auxiliaries may be evolving, depending on the lines along which the noun classes are distinguished.

Some of the auxiliaries may have incomplete 'paradigms' of suppletive forms, and for the noun classes that lack some specific auxiliary, some manner of periphrasis is needed, or whatever the auxiliary marks is simply avoided.

Tuesday, December 4, 2012

Detail #13: Indefinite pronouns

Ever since reading Ye Merrye Conlangre's short description of the typology of indefinite pronouns, I have returned to it on occasion. Now it - alas - is not available as such anywhere else than through archive.org and maybe google cache. The machines hosting the copies available at archive.org seem a bit unstable, but at the writing moment at least this link worked. If it does not, search for "http://math.berkeley.edu/~apollo/my-conlangs/mcindpro.html" in archive.org.

The sample sentences I will use in this post are partially based on that short essay.

However, the categories given for different correlatives are as follows:
  • known to speaker, not to listener
  • not known to either speaker or listener
  • irrealis, non-specific
  • polar question
  • conditional protasis
  • direct negation
  • indirect negation
  • comparison (e.g. bigger than any other)
  • free choice
The article provided lists some universals about which of these tend to be conflated - and a list of sample systems present in actual real-world languages.

I am thinking rather of doing something else with them - we could, of course, get rid of the pronouns and make them affixes or even give them their own verbal congruence markers or whatever, but that seems a bit boring as well.

I will end up suggesting an example system though, just so I can do something with it:

  • I: direct and indirect negation are conflated, as well as polar questions
  • II: free choice, standard of comparison, conditional protasis and irrealis, non specific likewise form one category
  • III: specific, known to speaker forms one group
  • IV: specific, unknown to speaker, conditional protasis and irrealis, non specific form one group (that partially overlaps with group #2)
Group I have their own congruence morpheme, third person negative object/subject. Group II simply lack congruence for the relevant role (note: the language has verb-like adpositions with some congruence, although the congruence is rather impoverished compared to the language in general, only marking one participant). Some verbs do introduce a linking vowel for morphophonotactical reasons. 

So, "bigger than my house" : big exceed.[3sg obj, appropriate gender etc] house.mine 
"bigger than any house": big exceed.[no marker] house
"bigger than the house": big exceed[3sg obj, appropriate gender etc] the house

Group III is marked by third person pronouns without person congruence. (Note that this limits specifik, known-to-speaker nouns to subjects and objects.)

He called[no person congruence], guess who! 
≃ someone called, guess who!

Group IV would be a separate pronoun, with optional third person marking on the verb, here the pronoun is ey :

if ey calls[3sg?] while I am out, tell[2sg subj, 3pl obj] to call[3pl subj, 1sg obj] later 
please go[2nd p. sg. subj] arrive[3sg object?] eywhere
ey stole[3sg?, 3sg obj] my car! 
Both could potentially omit the congruence element on the verb. Some irrealis modal marking on the verb could also contribute in example sentences one and two - possibly even omitting the pronoun?, whereas the third example is clearly indicative.

How to mark possession could be a bit more complex - but would probably have to interact with some strategies suggested elsewhere in this blog, maybe even such that the different pronouns use different approaches - possessive suffixes, possessive pronouns, voice rearrangements where subjects are owners of objects, ... possibly even omission of mandatory suffixes (analogous to the omission of verb congruence) ...

Other ways of arranging this could also be found - bigger than any house would seem to be rather a natural place to place an irrealis modal marker on the verb of comparison.

This would fit in some fairly synthetic language with loads of verbal congruence, and possibly possessive suffixes (and even more) on nouns. 

Friday, November 30, 2012

Detail #12: Entirely unmarked comitatives

In a language that marks the person and number of the subject on the verb, as well as having an accusative case, it would be cool to mark comitatives this way: nominative nouns that never trigger subject agreement (note: the nominative in such a language would basically just be used for subjects, complements and this kind of unmarked comitative). The comitative nominative is likely to be dislocated from the subject:
I.nom go.1sg fishing brother.nom
I go fishing with my brother
However, this is not used for comitatives semantically associated with the object; if the language also had object agreement on its verb, a similar object that does not trigger object agreement could be a neat 'parallelism'?

Detail #11: Lexical Hierarchies, again

Some isolating language where a word can be used both verbally and nominally probably already does something at least sufficiently like this to accuse me of plagiarism of real-world languages, but here we go anyway. Assume free word order, and rank words as to likelihood to have different roles in a way where verb, object, and subject are the three primary poles of likelihood, and maybe every word also has an orientation (e.g. verb -> object, verb -> subject, object -> subject, object -> verb, etc) and a distance from the centre. For simplicity, the distance from the centre in the examples will be given as two different: O and o, S and s, V and v, capital standing for greater objecthood/subjecthood/verbhood likelihood, and o,s and v being weaker. The orientation will not distinguish distance. Any word's quality will be given as Position→Orientation. In a situation where several nouns form a sentence, say:

bread? oven? coal?
bread might be O → s→ v, oven O → v → S, and coal s → v → o. We notice coal is the least likely to be an object of these three, and thus we  reassign it to subject:

coalo→s breadO→s ovenO→v U+21E8.gif coalS [breadO→s ovenO→v]
the two remaining alternatives are both equally likely to be objects, apparently, but they both prefer different secondary roles. bread as a subject is already covered, so it cannot be that, and is most likely to be the object, in which case oven would be the verb.
coalS breadO ovenV
This would parse as something like "coal bakes/heats (the) bread". It is likely certain tuplets of roots would acquire idiomatic uses that may violate the normal hierarchy-interpretation, and it is also likely certain tuplets of words would develop greater granularity than the average system, due to how often collocations occur in actual speech.

Exceptional "poles" can be considered to exist, but most instrumentals etc are marked by adpositions. A surplus of very similar arguments - such as several humans - can be considered to share the same role, although explicit such marking can be achieved using a conjunction.

I find it likely this violates some universals on account of lack of learnability? Could be an interesting thing to ponder.

EDIT: Rereading this post about two and a half years later, I find that this post never got around to explaining the "shape" of the, well, the hierarchy involved. It's essentially a circle! On this circle reside three points, that are equidistant along the circumference of it. In it, ellipses with a starting point and directions exist:
as you can tell, I am no graphical artist.

Now, in this graph the ellipses (with a direction) depict the probabilities for three words:
you have black corresponding to VoS, i.e. strong likelihood of being a verb relative to other words, followed by low likelihood of being an object relative to other nouns, followed by strong likelihood of being a subject. the red circle starts at S, goes to O and then passes v in a distance. The blueish circle starts at O, and passes s and v at a distance. 

Wednesday, November 28, 2012

Detail #10: Impermissible possession

Some languages have nouns that mandatorily have to be possessed. These often include body parts, family relations, and so on. How about nouns that cannot be marked as possessed using any nominal marking, but possessing them is possible by some uses of verbal voice, such as having the possessor be the causative subject, or the dative of a transitive verb, or even have the possessed subject demoted to object of an otherwise intransitive verb?

The language could even permit several of these periphrastic constructions and assign them to different registers or encode information about social status and deference in them.

Marking a permissibly possessed noun by any of these would be highly marked, and could indicate both disowning, distance or respect, depending on which construction is used.

I envision something along this table:

possessor <-> possessum
causative subject <-> the actor instigated to act on something (with transitive verbs that have been made causative)
subject of intr. verb <-> object of intr. verb (basically; there may be transitivizing morphology on the verb; still, the object is not properly or even syntactically an object in the language, as it should fail some syntactical as well as possibly other objecthood tests)
indirect object <-> subject (with both intransitive and transitive verbs)
oblique comitative* <-> causative subject
oblique comitative* <-> any non-subject


* in case the language has several different comitatives, the "most oblique" one, for some arbitrary measurement of obliqueness.

Detail #9: Unusual ordinals (again)

Getting back to the ordinal-vs-cardinal bit, how about a language which forms its cardinals a bit like Finnish:

1-7 [unanalyzable things]
8-9 [seem related to 2 and 1, with a suffix not found anywhere else]
10

I am not sure this is how eight and nine have been formed in Finnish, but a simple hypothesis would be that their meanings originally have been something along the lines of two away from (ten) and one away from (ten). In a language where the situation is more transparent, say something like base-12, sub-base 6:

1 em
2 da
3 tchi
4 fij
5 sa dun
6 dun
7 kha
8 ssem
9 dutchi
10 dufij
11 sa lop
12 lop
Notice how 9 and 10 are formed using the sub-base dun, whereas 7 and 8 are unanalyzable; 5 and 11 are formed from the base and sub-base by a morpheme that marks incompleteness (and also has verbal and adjectival uses). Now, ordinals are formed by simple applying the adjectival suffix -((a)r)a

ema, dara, tchira, fira, ..., duna(ra), khara, ssema, duratchira, durafira, ... lopara.
sa dun and sa lop instead are formed by a different formulation, applying a different prefix and different adjectivization:
ereduni, erelopi

Of course, first and second may also differ, along the lines of first, second, so something like
toma, seva or whatever. Once lop is passed, larger numbers are formed by juxtaposition: lop em, lop da, ... lopara-toma, lopara-seva, lopara-tchira, lopara-fira, ...

24 is formed as dalop, thirty as tchilop, etc.

This seems a more likely and realistic thing than having the entire ordinal system offset by one, even though I have a soft spot for that weird construction.

Tuesday, November 27, 2012

Further possession-related things

A thing monoglot conlangers tend not to realize is that possession often is used somewhat differently in different languages.

Consider sentences like
I talked to my mom earlier today. 
I need to wash my hands.
My legs hurt.I have it in my pocket. 
Put it in your wallet. 
My brother is an engineer. 
His brother is an engineer.
Oftentimes, which wallet, whose head, whose pockets, whose mother or brother etc is obvious from the context, and in say Swedish, you'd usually say these along the lines of
I talked to (the) mom earlier today (mamma, morsan; the latter being a diminutive form of moder) 
I need to wash the hands 
The legs hurt 
I have it in the pocket 
Put it in the wallet 
My brother is an engineer (or even the brother, but in that case generally in the diminutive form brorsan
The brother is an engineer (generally not using the diminutive, but rather along the lines of brodern; hans bror is also an option, and can be used as well.)
In Finnish, it's not even unusual that definiteness would be ignored in contexts like these, given that Finnish does not usually mark definiteness at all. It does, of course, have possessive suffixes and can thereby mark things like "my brother" as veljeni, but in contexts where whose brother is being spoken of is relatively obvious, it is not unusual to omit this.

There is a lot of possibilities here:

  • lexical distinctions along the lines of the use of diminutives or non-diminutives to distinguish owner
  • derivative suffixes developing towards being possessive suffixes? (Is that possible?)
  • reflexively possessed object marking developing through some odd grammaticalization of the situation described in Swedish?
  • extending how often things will be marked for first or second (or even third person) possession - you do not usually say "it's in my street" when something's in the street by which you live, "this time of my day, I usually take my nap", "I am heading out to get my bottle of milk"... this could be very common in another language.


Monday, November 26, 2012

Detail #8: A gender-like thing but with verbs instead

A language where there are three or so classes of verbs. These classes differ in

  • what affixes they take to mark different things
  • what they mark (e.g. class I marks tense-aspect but no mood, class II marks tense, aspect and mood, but has a less fine-grained tense-aspect system, and class III marks tense, and every verb in it has inherent aspect that cannot be changed morphologically)
  • details in the congruence system (class I has gender, person, number agreement for the subject; class II has no agreement, class III has person, number and gender agreement for subject, as well as a really minimal object-related agreement detail, as well as volition marking by means of reduced agreement)
They do not differ by 
  • meaning - there's no set of meanings that unite all verbs in one of the class or exclude verbs from some class; auxiliaries can be found in all three. With these exceptions: all verbs of movement are class I verbs, all verbs of possession, exchange of possession and agreements and contracts are in class III.
  • intensity, whether they are kinetic or not, transitivity, number of arguments in general, etc.
However, the classes also differ in some other ways:
  • subject and object case; class II is somewhat ergative in behavior. Coordinations using nominative and/or accusative is permissible, though, for all classes.
  • certain derivative affixes assign a derived verb to a specific class
  • infinitives are marked for case according to what the auxiliary's class calls for
Several verbs can be finite in a clause, and which subjects and objects belong to which verb can be distinguished by case (as well as proximity to verbs, if ambiguity has to be counteracted by phrasing). The case distinctions are fewer in the plural than the singular, so this works best with singular subjects.

Sunday, November 25, 2012

Short sketch #5: Possession and ergative syntax (again)

I have recently described something along the lines of a split-dechticaetiative system, and split-ergative should be familiar to pretty much every conlanger by now, I guess. How about split-possessive systems?

Some languages, such as English, mark possession on the owner, others mark it on the possessum, some mark it on both, some don't mark either; some have other systems than those even. How about a system that utilizes different approaches in different circumstances.

Let's say the language normally marks the possessum, but in transitive subjects marks the possessor.
John house-[poss] is big
We burned down John house-[poss] 
vs.
John-[gen] dog attacked us!

Let us imagine some further complications:

  1.  Transitive subjects possessing the objects: John built house-[poss] by himself
  2. Possessors of transitive subjects possessing the object as well: John-[gen] dog attacked cattle-[poss]
Alternatively we could turn it all around:
John house-[poss] is big ← with free word order this'd be cool
We burned down John-[gen] house
John dog-[poss] attacked us!
and turn the complications around as well, although getting a slightly weird thing:

  1. John-[gen] built house
  2. John dog-[poss] attacked cattle-[poss]? Joh-[gen] dog-[poss] attacked cattle?
The last option seems unnatural somehow, in a way the others don't quite.

Further, let's have -[poss] be identical to the instrumental marking in the language.

Saturday, November 24, 2012

Detail #7: splitting the adjectives in two classes: "adjectives vs. adjectoids"

Many conlangers like reducing the number of word classes in their languages. One popular and relatively easy target of this reduction are the adjectives.

Let us do something else, let's increase the number of word classes, and let us do that by splitting the adjectives in two!

How would one go about doing something like that? First, let us make them behave differently morphosyntactically; have one of the classes mark congruence with the noun as far as case goes. Let us have the other category redundantly mark some kind of noun class system along the lines of the system of Chinese.

Let's call the two adjective kinds adjectives and adjectoids. The difference given above seems a bit small, let us come up with some other differences, both notional and syntactical.

A simple difference that could be relevant to the speakers of the language would be whether a word is understood as denoting that the noun has a quality, or denoting that the noun is a member of a class with a distinguishing trait.


Adjectives

  • mark case and number
  • productively can form verbs both or 'turning increasingly x' or 'make something x'
  • use location-like metaphors (go (to) x, g from x, stand at x)
  • have case-congruence even as verbal complements
  • come closer to the head noun

Adjectoids
  • mark noun class
  • tend towards more periphrastic expressions
  • use membership-like metaphors (join x,  be of x, be with x, ) 
  • have no congruence as verbal complements
  • go further from the head noun
Adjectives have a closer syntactical affinity to verbs, adjectoids to nouns; still, both have (partly separate) morphology for forming nouns and verbs; verbs more easily form adjectoids, nouns more easily adjectives.
Both classes have some exceptional derived forms among nouns and verbs. Some adjectives have synonymous adjectoids, but far from all. Prepositional phrases easily form adjectives as well, whereas pure nouns easily form adjectoids. 

What word order this language otherwise would prefer depends a bit on what other syntactical features it has - I think something along the lines of det NOUN adjt. adjd.  would be neat.

Thursday, November 22, 2012

Onwards with Detail #6: More on split dechticaetiatives

(If I pull this post off without a single typo, I am happy. My spellchecker doesn't know dechticaetiativity.)

Dechticaetiative alignment, for those who do not know the term, is a bit like ergative and absolutive for indirect and direct objects. This means the direct object is demoted to some other case whenever an indirect object - which then has the same marking as the direct object of a transitive verb.

Rephrasing what one source has to say on terminology:
The recipient of a ditransitive verb and the direct object of a transitive verb are known as the primary object, whereas the direct object of a ditransitive verb is known as the secondary object.[1]
So basically
A.nom sees B.acc
A.nom gives B.acc C.dec => B=recipient, C=direct object

In the previous post on this detail, I described a split-dechticaetiative system. Is such a system sufficient to be able to distinguish grammatical information? For a system where some lexical items trigger one system to be entered into, there are two basic options - either the lexical items trigger dative-like behavior, or dechticaetiative-like behavior.

Dative as the triggered option

Let us assume it triggers dative-like behavior. In that language, the normal construction would be
A.nom Verb B.acc C.dec (essentially conveying what in dative languages would be A verbs B.dat C.acc)

For no reason whatsoever except that I need to be able to refer to something that triggers the behavior, I will go for reflexive pronouns being the trigger.
In that case we get either
A.nom verb rflx.dat B.acc
or
A.nom Verb rflx.acc B.dat

The presence of .dat itself (or whatever other marking, be it on the verb or some particle or whatever) is sufficient to show that something odd is going down. We can come up with a further restriction: what if dative behavior only is triggered by the reflexive pronoun acting as direct object?

  • A.nom verb rflx.dat B.acc
  • A.nom verb rflx.dec B.acc

This seems interesting. The .acc-marking on B remains the same no matter whether the reflexive pronoun is dative or dechticaetiative. A neat benefit of this is that only the pronouns that trigger the dative-like behavior require any dative morpheme.

What if the same pronoun is used for other third person as well as reflexive third person? Is there sufficient information available to tell whether it's reflexive or not based on which alignment is triggered?

In merely transitive sentences, no - and there could very well be some extra marker appearing in those. In ditransitive sentences it would be sufficient, though, but not in combination with the previous rule (which restricted dative behavior to the pronoun acting as direct object).

  • A.nom verb pron[A].dat B.acc
  • A.nom verb pron[not-A].acc B.dec

vs.

  • A.nom verb B.dat pron[A].acc
  • A.nom verb B.acc pron[not-A].dec

Dechticaetiativity as the triggered option


Let us investigate the other option, dechticaetiativity triggered by certain pronouns:

  • A.nom verb B.dat C.acc
  • A.nom verb rflx.acc C.dec
  • (A.nom verb C.acc rflx.dec)
I would guess the third sense there is unusual - for some reason, "I give you myself" and such seems a rather unusual thing to say, unless there are very specific cultural reasons for it - which there well may be, and maybe I just happen to live in a culture where such utterances seem highly marked, who knows.

What we notice here is that it seems nearly impossible to get rid of having to mark every noun for every case - there must be dec, acc and dat available for each.

Of course, I'd deal with that either by having these also mark some other case, such as
dat = gen
dec = loc

or something along those lines. Another option, of course, is to use adpositions or have a fully productive case morphology, or marking on the verbs instead. What also could be fun would be to have either dechticaetiative or dative be identical to accusative. 

Lexical Hierarchies?

How about having a hierarchy of which noun is likely to be direct object and which is likely to be indirect object, such that they both are syntactically treated as direct objects. When an unusual order is desired, some synonym that is higher up in the hierarchy is used. This is of course a direct ripoff of animacy-hierarchy object marking systems. However, I would almost wager that this is unlikely to develop in this part of the language, and more likely to appear with regard to the direct object vs. subject distinction.


Split dechticaetiativity would be interesting to combine with some voice system and have the ways things shift around get somewhat unpredictable. Dechticaetiativity, of course, is the new ergative, so I guess I am late on that bandwagon. Another thing, of course, would be to chain together ergative and dechticaetiative systems, but I know I am not the first to propose that. I suspect strongly that no such language exists "in nature", so to speak, since dechticaetiative languages are fairly uncommon outside of Africa, and ergative languages are fairly uncommon inside of Africa. If God exists, he's holding out on us.

Further complications?
Of course, there is nothing that says the presence itself of the pronoun has to trigger it - maybe only the presence of the trigger lexemes in the right syntactical role triggers it? Or it is verb-specific, or relates to how the speaker feels about the event referred to.

[1] Wikipedia

Wednesday, November 21, 2012

Detail #6: Split dechticaetiativity

In this constructed language sketch, certain pronouns and nouns follow a dechticaetiative pattern while some follow a normal accusative-dative pattern.

Pronouns that would follow the dechticaetiative pattern could include reflexive pronouns and 1st and 2nd person pronouns in general. This way, reflexive pronouns would only be distinct from normal third person pronouns by the presence of a different case and slightly different morphosyntax in certain situations.

Since it's late in the evening, and this idea occurred to me just now,  I figured I might just as well jot this down and elaborate on what this means and how it could work in later posts. For the readers that do not know what dechicaetiative is, I guess I will explain that in the next post.


Monday, November 19, 2012

Detail #5: Intrinsic, implicit referents

Does any natural language use obviative pronouns also to refer to nouns that have not been mentioned as such, but which are kind of intrinsically related to prominent nouns in a sentence?

Something like, say,

children look like them.obv = children look like their parents
house had no him.obv = the house had no master
sheep.plur know.3pl him.obv = the sheep know their shepherd

Not all nouns, necessarily, have their own associated nouns such as these; some have many, differentiated by gender and numbers.

When this is supposedly "normally" obviative or an indirect reference like this may be unclear from context.

Also, challenge #1 still stands. I have personally made no progress on it.

Possibilities #1: Comparatives on non-adjectives

Fun things with comparatives and superlatives


Superlatives and comparatives are not universal in human languages, c.f. Biblical Hebrew and such. Certainly there's ways of expressing whether something has more or less of some quality than something else, but there is not even any guarantee that this would be a set periphrastic construction - there may well be one, but it could also be the case that one can come up with any variety of ways of expressing it. 

How about using the same morphemes with verbs as well? This could lead to a variety of shifts in meaning:
  • doing something more intensely, or with greater [whatever] than what usually characterizes the verb, e.g. run faster, shout louder, play more melodiously, play louder, preach convincingly, preach annoyingly, preach with even greater hypocrisy than usual, sleep tight, sleep such that it is difficult to wake the sleeper up, run fast, run clumsily/trample,  ...
  • doing something to a greater number of objects (or indirect objects)
  • doing something more often, more frequently, more deliberately, more intentionally, more ...
  • ...

Adpositions obviously could be compared:
close to - closer to - closest to
above - higher above - on top of
It would be of some interest to maybe have fewer grammatical distinctions in the comparative or somewhat; as if positive and superlative adpositions form triplets distinguishing from - at - towards, whereas the comparatives are defective in lacking the distinction between at and towards?

Finally, for nouns, one could easily obtain meanings like
  • greater importance
  • greater stature
  • greater size in general
  • greater in some essential quality (harder stone, etc)
The important thing I would go for with a system like this is not any kind of unified meaning - not just picking one and using that for the entire noun system or even locking every specific noun to its preferred alternative. Rather, some nouns or verbs may favor just one of them for the majority of use - so like, 95% of the use of house.comparative would relate to size, whereas a similar percentage of house.superlative might refer to some kind of religious building, functionally important building or otherwise remarkable building.

That is, it should be very contextual rather than fixed.

[EDIT:

Finally, in Finnish, comparatives sometimes go on nouns with locative endings, indicating 'closer to'.)

Sunday, November 18, 2012

Cultural detail: A singing style

In some tonal languages, tone carries more information than vowel quality does. To understand what this means, say we have five vowels (/a e i o u/) and three tones (let us mark them a, á, â, for convenience). If we were to take a sentence such as
ka lé pê mindâ lâktú patínmerông
and we conflated all tones and obtained either of
ka le pe minda laktu patinmerong
ká lé pé ...
kâ lê pê ...
or alternatively conflated the vowel qualities but kept the tone:
ka lá pâ mandâ lâktá patánmarâng
ki lí pî ...
...
Of these two alternatives, in those languages the latter would be easier to understand for speakers of the language. This is part of how drum language and whistled language works - although some extra redundancy often is added to ensure that there is sufficiently little ambiguity.

In many cultures with tonal language, musical tone is permitted to supplant linguistic tone in musical settings, sometimes making lyrics difficult to understand. I would assume a good composer-poet in those cultures would manage to make the musical and  linguistic tone line up throughout - or often enough that misunderstanding is unlikely or potentially even a productive thing (puns and such).

However, what if a style of music or musical poetry went the other way around, and instead of having melodicy tone trump linguistic tone, it had timbral melodies trumping linguistic timbre (= vowel quality), while the linguistic tone remains untouched? Essentially, a timbral melody, as far as song goes, would be something like
a e a i o
a e a i o
o a o e
a e a i o

so instead of, say
ka le pe minda~
ka le pe laktu~
patinmerong~
ka le pe minda~
 Where each syllable's tone is determined by the melody, you'd have

ka lé pâ mindô
ka lé pâ lîktó
potánmorêng
ka lé pâ mindô
where each syllable's vowel is determined by the vowel-sequence.

Timbre would be an important thing in the music of such a culture so instruments with adjustable resonance boxes or other mechanisms of adjusting the timbre would maybe be common. The most obvious adjustable resonance box is the mouth of course, so various things like this, this as well as jew's harps and such. If the vowel system is sufficiently small - five or seven vowels, maybe a simple musical scale of five tones and seven timbres would be doable as a xylophone or metallophone with rows corresponding to timbre and columns corresponding to tones. The full cartesian product of tones x vowels might not be necessary.

Friday, November 16, 2012

Detail #4: Verb-congruence failure

A language where verb congruence fails in these cases:

In the following circumstances, the person congruence is conflated with a third person verb form:
  • negative reflexive verbs
  • any existential verbs (including verbs of motion or location, when used in an existential manner, such as "there are so-and-so running around on our back yard")
  • presence of non-subject topics

This specific third person form is also used with third person subjects - both singular and plural - for expressing:
  • with verbs of physical actions it marks plural subjects not acting in a concerted manner
  • most often with copulas for any subject person
  • lack of volition for third person
  • nonspecificity of singular subject
Verb congruence also fails when a noun is used to denote a full class of things, so that the plural verb is used even with singular subjects, e.g.
  • caviar is fish eggs
  • cars have engines
  • the antelope is an African mammal
all would have plural congruence on the verb. When the upper set of rules and this set of rules conflict, which one is chosen is in free variation.

Challenge #1

Imagine a system that forms ordinal numbers as follows:
first := suppletive
second := after one (could be suppletive as well)
third := after two
fourth := after three
...
The origin for this system is along the line of specifying the position in some line-up: the one after five others is obviously the sixth one. The numerals may have congruence or some such to mark gender or class agreement with the noun - class agreement along the lines of Chinese would probably work out really well with this.

This may later grammaticalize, and the morpheme marking "behind" (as well as the class marker) may merge into the numerals. In this case, we would obtain a system where each ordinal is offset by -1 compared to the corresponding cardinal.

The challenge itself:
Come up with a reasonable and likely development of a system where the offset is +1 instead. Extra challenging would be reasonable systems with an even greater offset in either direction.

An obvious attempt would be to use "before three" to mark the second, but this doesn't make logical sense: there is no guarantee the n:th element will be part of a group of n+1 things.

Contribute through the comment-field if you have an idea.

Detail #3: some analytical tense constructions

In some language where the normal indicative present (progressive) is the simplest tense, morphologically, the active participle is used with copula and whatever verb construction is used in place of to have, to obtain  various TAMs by various constructions. To be is used with intransitive verbs, to have with transitive ones.

subj: nom, (obj: acc),  participle.acc = inchoative
subj: comitative, (obj: acc),  participle.nom = habitual [exceptional in using to be even when an object is present]
subj: dat, obj: instr, participle.nom = indicates that the subject would like to do something. Intransitives require a dummy object, which normally is the numeral "one" or an indefinite pronoun.
subj. dat, (obj: nom or acc), is + participle(comitative) = obligation

The piece of paper on which I wrote this idea down years ago was tiny, and therefore there's a lot of really short ad hoc abbreviations in it. Hence, exactly what I meant is a bit unclear to me.

Still, I think a language with something like this would be best with, say, 4-5 different constructions along these lines, maybe one or two of which only appear with objects present. Finally, they shouldn't completely cover a certain type of thing - like each construction encoding a particular tense or a particular aspect or a particular mood; they should each express some combination of tenses, aspects and moods, and not necessarily even just one of them; and preferrably, they would be somewhat orthogonal to each other in what they express, so that they not form a neat paradigm.

Not every combination of subj.[case1], obj[case2], participle[case3] should be used, and some constructions where two of them have the same case could very well be permitted.


Wednesday, November 14, 2012

Short Sketch #4: Definite and indefinite verbs

Verbs come in two main forms, sometimes differentiated by suppletion, mostly by some simple but not entirely predictable morphology. (Probably such that dialects differ greatly in which affix which verb takes).

Indefinite verbs are more coarse-grained as well dividing semantic space generally into fewer slices, and may be less permissive in whether they accept direct objects at all.

Definite verbs, however, do not permit topicalization of objects, except through passivization. Passivization of a definite verb does not demote the subject to oblique, but both the subject and object of the finite active verb behave as independent subjects of the finite passive verb. Passives do not have congruence.


Other unrelated idea:
Have a few verbs require the object to be the topic.

Tuesday, November 13, 2012

Detail #2: Multiple distinctions encoded in conjunctions

An isolating language with a multitude of 'and'-like particles with mainly TAM-like differences (some applicable to both noun and verb phrases, some only applicable to verb phrases or object NPs), such as

  • denoting intertwined, simultaneous, forming unified whole (can also denote comitative-like meanings).
  • habitually intertwined, alternating in time.
  • completing the first action, after which on to the next.
  • irresultative, simultaneous or negative.
  • one action without completion, simultaneously with another action without completion.
  • ...

Monday, November 12, 2012

Detail #1: a possessive and a causative structure

In some language,
"nom1 has an adjective nom2" turns out
[nom1] [verbalized adjective] [nom2, oblique object]

whereas "nom1 makes nom2 adjective" turns out as
[nom1] [verbalized adjective] [nom2, direct object]
The marking as direct object or oblique object may be obtained by some kind of adposition if the language is primarily isolating (in which case I'd suggest the oblique object be the only one marked at all), and in a more inflecting or agglutinating language, an accusative or somesuch would probably fit the bill fairly well.

Short sketch #3: Dummy objects

Imagine a language where verbs are made more 'intensive' by adding an object of some kind.

Giving glosses, where "dummy" stands for "some dummy of some kind"
hear.2pl   = we (do) hear 
hear.2pl   it.acc = we listen 
hear.2pl   dummy.acc = (thisnoise is annoying us

However, when introducing a real semantic object, this language does not permit the dummy to be the same case as the real object, and demotes it to some other case:
*hear.2pl throbbing gristle.acc dummy*.acc   ----> hear.2pl throbbing gristle.acc dummy.dat 
The same process repeats when an indirect object is added, etc. Different dummy cases have different hierarchies and will be demoted along different paths.

Sometimes, a dummy is repeated in different spots to increase the intensity:

rain.3.sg.  =  it rains
rain.3.sg it.acc = lit. "it rains it", ~it rains intensely
rain.3.sg it.acc it.instr = lit "it rains it with it", ~it's pouring down
rain.3.sg it.acc it.instr it.comitative = it rains it with it and it too, ~this rivals the deluge.
it.nom rain.3.sg. it.acc = the subject dummy probably works to make the rain somewhat more specific, such as "right now, right here, it rains intensely". In combination with a previously established location or timeframe or both being somewhat topical, it would be more likely to pertain to that location or time or both.

As for describing persons, adding such objects could be used both for positive or negative effects:
play (pertaining to instrument).3sg dummy.acc = he/she plays (his/her instrument) intensely well.
 give.3.sg dummy1.dat = put out
give.3sg dummy2.dat = he/she is slutty; the verb is only interpreted as referring to sex when with a dummy dative.

These dummies originate as nouns, often specifically differently strong profanities. This idea occured to me after I heard my brother use "satan" - a common Swedish curse word - as an intensifying particle after 'basa' (to emit bass tones), as in "an basar satan" to describe the volume of the bass sound emitting from an amplifier and its speakers.

Short sketch #2: Participle craziness

A language with more than three dozen different kinds of participles-gerunds, marking things like future religious obligation, current religious obligation, obligation to close kin, active carrying out of an action in contravention of agreement, being the object of an action carried out in contravention of an agreement, having been the witness to a contravention that has been violated, intense desire for performing an act,  etc. No verb can be inflected for all of these participles though, the average verb having less than ten of them, the occasional verb having upwards of twenty. Tense, aspect and even mood feature in various not very regular ('cartesian') combinations, and cultural things factor into the huge set of available participles.

The most basic ones would include:

  • actively, currently carrying out an action ((with much|with some|without] physical exertion)
  • recently having finished carrying out an action (with|without physical exertion)
  • intending to carry out an action (with|without physical exertion)
  • being obliged to carry out an action
The participles can be used as adjectives, as nouns describing the state (with some syntactic restrictions on how they can be used), and as complements of some auxiliaries. Adverbial use is also possible.

Secondary idea that goes well: some auxiliary-like verbs lacking finite forms altogether, and being used for somewhat anaphora-like uses and in some idioms. 

I envision this as being in a language with a relatively small nominal morphology, with a simple TAM system. 

Short sketch #1: Case system with ergative detail

A language with a somewhat latinate case system, say nominative, genitive, accusative, dative, ablative.

This language descends from a clearly ergative stage, where the ergative case, as in many ergative languages, was identical to the genitive. The indefinite and the intransitive subject are both in the nominative. However, definite third person transitive subjects still are marked by the genitive, except for the pronouns.

The genitive/nominative distinction has carried over to the third person pronouns, but in the pronominal system, a genitive subject is obviative and a nominative one proximative. Genitive 1st and 2nd person subjects tend to mark seriousness.

The accusative is used to mark the object of verbs deemed not beneficial to the first or the second person, sometimes even a third person of personal relevance to either; the pragmatics and sociolinguistics of when the first or second person is relevant are slightly complicated. Beneficial or neutral objects tend to be marked with the dative.

The verb has the following set of forms: [singular, plural] x [1st, 2nd, 3rd] + [3rd II], where II just serves to disambiguate it from the explicitly 3rd person singular one; 3rd person II is used both with singular and plural indefinite subjects, some definite subjects under some circumstances, as well as with subjectless verbs. It has a smaller set of TAM-distinctions than the other persons, and cannot combine with auxiliaries to the same extent either.

Miniature conlangs

The construction of languages for hobbyist purposes tends to be a very private hobby, in that a lot of thinking and tinkering goes into every tiny bit of produced quality conlanging. Many clever or interesting ideas may never reach any kind of audience whatsoever, as the conlanger strives for a full reference-grammar style work to see the light of day. Many ideas never are worked into that reference-grammar.

The intention here is to post short descriptions of grammatical features I come up with that just might get included in some actual more coherent, full conlang grammar. Some of these ideas will not be destined to merge with anything such. As ideas add up, some of them might fit together. I might mention if some ideas are gravitating together into a single language.

If someone spots an idea that they know to be attested, point it out! I'd be glad to hear of such coincidences - although possibly, it can be the case I am accidentally plagiarizing some reference grammar or such.

Ideas may pertain to any field of language - most will probably be somewhat syntax and morphosyntax-related, possibly some lexical ideas, occasionally morphology, even less often morphophonology and phonology. I do hope to have some imaginative ideas for fictional sociolinguistics and pragmatics as well.

Some posts may qualify as "microblogging"-sized nuggets. Some may be a bit more researched with even books on typology being quoted.