Some Nahuatl Links

A friend of mine has a blog on nahuatl stuff, which might be inspiring for a conlanger to read about.

And this reminded me of an article about a no-root verb in some Nahuatl dialect - a verb consisting of two affixes, and nothing else. It's quite an interesting essay.

The Former Noun Morphology of Tatediem

Back when I first came up with the name and the concept of Tatediem, I wanted to have cases in it. In fact, my first post about it on this blog assumes it has the old case system that I originally envisaged for it; however, I never spelled it out in detail, and I figure doing so might be a good idea, because it's not entirely half bad.

	abs	erg	dat	instr	part
singular
def	-te	-k:ı	-k:ı	-istik:	-istik:
indef	-∅	-∅		-isti	-isti
dual
def	-ǎt	-dəki	-dnǎ	-dak	-dnǎ
plural
def	-aftin	-ambun	-ambun	-ǎmbust	-ǎmbust
indef	-ǎ	-abui		-ǎbuist	-ǎ
mass
indef	-o/-e	-o/-e	-osti/-esti	-ǎbuist	-o/-e

There are some things I'd do differently, now ten years later. I might rather have conflated ergative and instrumental in part of the system, than ergative and dative. However, I might also have omitted mass as a 'number', and rather described mass nouns as having their own declension. The omission of indefinite dative makes some kind of sense - indefinite beneficients would call for a slight periphrasis. I think it's fairly likely that the indefinite plural would conflate absolutive and ergative if the indefinite singular does so - in fact, I find it more likely that the plural would do so than that the singular would.

Still, I like this system a lot even though it dropped out of Tatediem as it grew more noun-class-centered and less reliant on case. Might revive it sometime.

In many ways, playing around with a less case-centered language is pretty interesting.

Tatediem: Numbers

The numbers in Tatediem are somewhat inconsistent in their morphological behavior. Generally speaking, one to four take singular marking; usually, two is only used in counting and the morphological dual is used instead. With masculine, feminine and neuter I, it is not unusual for the dual to be used with three and four, but not mandatory. It is uncommon with neuters III and IV, due to the length of the gender-number prefixes (setem-, geme-) . The noun prefixes are used.

one	-t	-xuns
two	-kint	-xera
three	-párt
four	-sèlx

Thus, counting masculine things would go net, kint, wanpárt or nepárt, wansèlx or nesèlx, whereas counting feminine things would go sart, kint, xanpárt or sarpárt, xansèlx or sarsèlx. The right column gives the forms used when compounding for larger numbers.

A pair of things can be formed by compounding the dual number prefix to 'one', or, remarkably enough, by prefixing the singular prefix to kint.

Five, six and seven take the plural marking instead.

5	-barà
6	-kenà
7	-pelì

Eight, nine, ten and eleven do not take any morphological marking whatsoever, but force plural marking on the noun.

8 - buns (with the form -tikàr when compounding with su-, rupu-, sader-, etc, which will be introduced a bit down)
9 - xerans
10 - mbártans
11 - sanáns

from eleven upwards, however, all nouns take plural markers, and the least significant digit takes the gender congruence, except if that number is buns:

12 wan|sélx suxuns
13 wan|bara suxuns
14 wan|kena suxuns
15 wan|peli suxuns
16 buns suxuns
17 wan|khun suxera
18 wan|xera suxera
19 wan|párt suxera
20 wan|sèlx suxera
21 wan|barà suxera
22wan|kenà suxera
23 wan|pelì suxera
24 buns suxera

su- is a prefix that basically forms 'bunches of eight'; rupu- forms 'bunches of forty'. sader- forms 'bunches of 120'. kurber- forms 'bunches of 480'. When speaking of numbers as numbers, the grammatical class is used, i.e.

kurbértíkàr ya-dekàr l-mónta re-l-kàlù
ku:b:ɛ́:tíkà: yadekà: lónta relkàlù
8*480 neut2-number/amount  neut4-big gram-neut4-reach
3840 is a big number / reaches a big amount

Ordinals are formed by using the adjectival prefixes instead of the nominal ones, with one exception: the neuter II. The neuter II uses the same prefix as the neuter I. Here's 'the first', which also uses xuns, exceptionally, maybe because et, rat, ... would be too short:

masc	exuns
fem	raxuns
neut I	kexuns
neut II	kexuns
neut III	yexuns
neut IV	yexuns
grammar	urxuns

Et and rat do occur in some contexts - mainly in numbering the volumes of scholarly tomes, the points in the agenda of a meeting. Counterintuitively it also appears at times in faux-foreign accents, where often the -t/-xuns pair are intentionally mistaken.

Notice that the neut III and IV ordinals are identical to neut II cardinals.

The other number with exceptional ordinals is buns, eight. Its ordinal is formed using tíkàr.

The 'grammatical gender' version 'urxuns' is basically an analogous formation to English 'firstly', although its meaning is somewhat distinct - it is not so much used to state importance as it is to state temporal priority.
'urkint' essentially means 'next, thereafter, then, subsequently'. 'urxera' means 'secondly, in order of importance' - sometimes, it is compounded with gender prefixes forming eyurxera, rayurxera, kexurxera, etc, which signify 'second in order of importance'. Urpárt and ursèlx sometimes are used to weaken the causal connection or the time-span between the events in a narrative or discussion.

Some Ideas Regarding Ordinal Numbers

One morphological way of forming ordinal numbers that somehow makes sense would be the verbal perfect aspect marker or a perfect, active participle. Consider 'the fourth item' is the item that 'brings' the number to four, thus the item that 'reached' four, the item that caused the items to number as much as four.

However, there are other things to do with ordinals than just their formation. We can consider what other things than numbers can take ordinal marking. In Finnish, the word 'monta' (many) permits forming an ordinal - mones (manyeth), which mainly is used in combination with the interrogative particle -ko. Monesko? 'How manyeth?' Of course, we could imagine a language that has forms like:

I don't recall which, but in someth part of that series they really jump the shark.

He's not the first to do this, but at least fewth.

One could maybe extend it to meanings like 'every so many|often|':

the oftenth day, we throw a party.

no matter the manieth the ephemeral obsessions of the youth, there will always be a next.

Of course, next is sort of ordinal-like as well, and one could also imagine offset-ordinals: the n:th from this one, as well as periodically offset-ordinals: every n:th from this one, (versus every n:th).

But, to go further and create some semantic fun, consider a language where ordinals can go on both sides of their noun.

the n:th thing := the thing that is n:th in order
the thing n:th := a thing that completes a culturally sort of significant set of that particular type of thing; i.e. 'the seventh out of a set of seven unfortunate events', the third out of three deaths in a village in a short span of time, the twelfth month out of a year, the sixth beer in a six-pack - so month twelfth, death third, disaster seventh would all be somewhat grammaticalized phrases.

But we could go on even further. Any out of these nouns, if in the plural with the ordinal after it, would signify the whole set of months or disasters; with the ordinal in the plural, it would be such sets of months or disasters or six-packs.

 

 

 

 

 

 

Music Theory for Conworlders

I recently decided to write a series of posts about another favourite topic of mine - music! Many conworlders do mention the music of their concultures. Some even go so far as to compose some works set in those conworlds. However, most such works remain within the trappings of Euro-American styles.

Obviously, no co world can be an entire world, and no conworlder is omniscient. However, music is a field where innovation  is not all that unattainable.

At the tuning tarpit I present a primer to some type of generalised music theory, with audio examples to illustrate the concepts presented.

Bryatesle: Case Usage III: the Dative

This post is part of a series on the case system of Bryatesle

I. The nominative and some subcases

II. Gaps

III. The Dative

IV. The Ablative

V. The Ergative

Besides the nominative and accusative (which later will be described), Bryatelse uses dative and ablative quite prominently. For some combined forms, the dative and ablative are conflated into a single 'oblique'.

The dative is comparable in some sense to a classical dative - it marks the recipient of an action. Examples:

emi mersi nalei
she him.dat sing-3sg - she sings for him

nek es urek siber
me.dat it strange appear.3sgN - it appears strange to me

xnërsi xus lrama
you.pl.dat xus fall-3sgN - something bad happened to you

However, beyond this, it also serves as quirky case subject with a handful of verbs:

tënk bulyr nïsr kevyk du inrahat
who.dat believe.3sg I.^2ndsubj that like behave.1sg? - Who(DAT!) believes that I behave like that?

 The dative further can serve as a genitive:

Erkube tebuxu vybara
Erkub.dat cake.neut.nom suffice.telic3sgN (the verb is vyberet in the atelic form); Erkub's cake sufficed. 

The dative is the case used when the possessor is not marked as subject, secondary subject or object of the clause. Topical secondary subjects can own any participant of the clause, subjects can own any participants, objects can own oblique arguments, and oblique arguments can own other oblique arguments. A dative or ablative that is directly in front of a possessed noun, however, can own that noun. Datives are preferred with human possessors, ablatives with inanimates and non-human animates.

Directions with nouns are not given by the dative, most of the time - generally, those are formed by postpositions with the accusative.

In combination with the secondary cases, the dative further has some interesting deals - with the partitive it and the ablative merge and form a single oblique case. With the definite the implied telicity, pastness, 'realisness', absolutely-likely-to-happen futureness, etc, of the statement is slightly strengthened.

With the partitive, the opposite implication is generally acquired, and there's a very weakened definiteness. (However, as noted, it then merges with the ablative.) The possessive should be fairly well understood by now. The reciprocal object will have its own post sooner or later. The secondary subject appears on the dative for two quite opposite reasons:

Promoted Dative Secondary Subjects

Subordinate Dative Secondary Subjects 

The previous kind basically are causers, parties involved indirectly but who cause things to come about. The subordinate dative secondary subjects are embedded in VPs that are objects of the main verb; they either are quirky case subjects or they are datives that serve a topic-like function in the embedded subclause.

Erkub taidënisr rulmunti ake sigi
Erkub wife-dat-2ndsubj tobacco.obliq-part not drink.atelic-3sg - Erkub doesn't smoke tobacco (because his wife doesn't like it)

Garkeb kendynnyx bumal sega
Garkeb desire.dat-pl.neut.2ndsubj water drink.atelic-3sg.neut
Garkeb felt like drinking water, so he did

The subordinate kind would be exemplified by

Firgyk naskannyx barlei ydrer
Firgyk field.fem-dat-plur.2ndsubj-plur dry_up.atelic-3sg fear-3sg.neut
Firgyk fears the field will dry up

The embedded verb usually comes before the main verb, but if the semantic difference of the kinds of verbs involved is big enough, the order can be reversed, and the change can signal differences along these lines:

Firgyk naskannyx ydrei barlei: Firgyk, on the account of the field is worried - it might dry up.

Firgyk ydrei naskannyx barlei: Firgyk worries, the field may dry up

naskannyx barlei Firgyk ydrei: the field drying up has Firgyk worried.

The remaining forms - both main objects and secondary objects - will be covered in the next post.

Dairwueh: Phrasal Irrealis Verbs

Actions such as look for, hunt, propose (marriage), ask for permission, and even plan have no roots directly corresponding to them. These, all of which have a clear corresponding 'perfective' verb root in Dairwueh - bargət find, sgaran kill game, riksəl marry, srusnam be given permission, sŋalək carry out - are generally expressed as an irrealis with one of the adverbs 'kudku' (present), 'kudku hve' (past), 'kudku həy' (future), replacing kudku with 'kudkunš' negates the adverbial complex, but also calls for a negative main verb. The tense marking moves to the main verb in the negative, however:

vnu kudku hve riksiŋ - he proposed to her. He would marry her (if the proposal goes through).

vnu riksiŋ - he married her 

te kudkunš bargeyš - I did not look for it/I was not going to find it

te bargeyš - I did not find it

Even when there is a clear atelic lexeme, such as 'ratmək' ('build'), it is not uncommon to use this construction with a more telic verb instead, and sometimes even a somewhat distant one as far as meaning goes - for instance 'hargiyn' ('settle, move in')

bare ratmas
house:acc build-Isg - I build a house

bare kudku hargiyas
house:acc ADVB settle-1sg - I would/will settle a house (the one that incidentally is being built)

tsernu vidko
food.acc cook.3sgFem - she cooks food

tsernu kudku sperit kagino
food.acc ADVB front-FEM.PL.DAT cook.3sgFem - she would/will place forth (~serve) food

Thus, kudku reduces the irrealisness significantly, and rather creates an aspectual/temporal distinction, where the fulfillment of the verb is in the future of the time referred to.

Bryatesle: Case Usage II: Gaps

This post is part of a series on the case system of Bryatesle

I. The nominative and some subcases

II. Gaps

III. The Dative

IV. The Ablative

V. The Ergative

Lots of Bryatesle nouns are defective in some sense or other.

A small number of nouns lack a nominative altogether, but otherwise have a more full set of forms. The term 'naked' marks the use of the primary cases without secondary case marking - some nouns lack a primary case except when it's marked for secondary cases as well. Here, I give the root and a vowel that gives some indication as to the forms used by other nouns:

*seig(e) (masc) extended family, a set of relatives, clan; lacks nominatives. igan which has a similar meaning can be any case. However, there are differences in meaning - igan is more matter-of-factly, whereas seige includes a variety of obligations and expectations.

*ritë(n) (masc) temple

*kurb(i) (fem) a common cereal

*pryb(e) (masc) an excuse

Those above never appear in nominative, neither with or without secondary cases. Some nouns, however, only lack the 'naked' nominative, and usually all other 'naked cases' as well. Some secondary cases may also be missing:

pargu- (masc, pl) - lacks singular as well. Money. Only appears in partitive, possessive, negative and definite in combination with the other cases.

bras- (neut) - power, strength. (Almost never appears with the negative, and seldom with the partitive.)

xryg- (neut) -  fear. (Primarily partitive or definite).

kirga- (fem) - zeal

kust- (fem) - sufficient space (lacks partitive)

sydne- (masc) - stench (lacks definite, but does appear in the 'naked' accusative, dative and ablative)

kulyt- (neut) -  test, attempt (has a naked accusative, but no naked dative or ablative)

A bunch of mass nouns exclusively have forms with the partitive case:

jagur, jagur, jagyr  (neut: nom, acc, obl) dirt (as in 'a spot of uncleanness')

kragu, kraguze, kragux (fem: nom, acc, obl) ash. This has a related noun which exclusively has possessive forms - kragela, kragei, kragir, kraging - the ashes of a dead person.

stertu, stertëze, stertër (masc: nom, acc, obl): earth, dirt

sardur, sardur, sardyr (neut): a certain poisonous fruit (that is used for colouring fabrics)

Some nouns only have possessive forms:

ragd|unë, -an, -ar, -ënt; -uvu, -ëkux, -emxi, -emxi, debt (masc, given as sg nom acc dat abl, pl nom acc dat abl) (a synonym that has a full paradigm is kugyda, (fem))

 kynt|ela, -ei, -ir, -ing, -ivi, -eki, -ersi, -ersi: inheritance (fem). A fuller paradigm can be obtained by kynurdi (fem)

drig|ela, -ei, -ir, -ing, ... the wife of the king. Unlike the other 'possessum tantum' nouns here, drigela can be used without a syntactic possessor present.

Some only have definite forms:

driginë (fem) a ruling queen (as opposed to a queen by marriage, who exclusively appears in the possessed form)

 ktirtes (neut) - any castle

sandes - the direction 'east' (the other directions are not restricted to definite)

Two nouns only have reciprocal object forms:

sakivn- the two participants in a duel, whenever their actions can be expressed in reciprocal ways. For obvious reasons also always plural.

lidan- a plural term denoting the seasons in their cycle of one following the other. Of course, only appears when grammatically possible. Lidanisr sagita, lidanisr sagitër - seasons have followed one on the other, seasons will follow one on the other.

A few nouns only have  secondary subject forms. All of these have synonyms that have full paradigms:

harkansus, -aksus, -esus, -itysus, (no excl or voc), -ersus, -uksus, -umxus, -ursus; a teacher can be referred to by this term if he  does something that grammatically calls for it - which he often will, since he often is a causer. The more common term is girme.

kurmasus, -aksus, ...  a military officer. The more common term is sudur.

ginsus, -iksus, ... a king. The more common term is parge.

 Further, there are nouns that exclusively have a nominative:

seban (masc) - one of the designations for God.

lavyk (masc) - 'something characterized by being sufficient for the situation'. (Only combines with reciprocal object and secondary subject markings).

Further, some nouns lack accusatives:

pyrida - a thought

stirmu (neut) - a lifetime

kidsan (fem) - a regular prayer

More regularly, most nouns designating stretches of time lack accusatives.

Lack of ablative and dative forms is unusual; lack of exclamative and vocative is, however, rather usual.

A large number of adverbs are basically nouns that only exist in some particular combination of case and secondary case:

tabiring! - a profanity, 'the demon of!' (the possessor being implied by syntactical cues, but not being a proper possessor.) The meaning is comparable to 'the gall to do such a thing'. Tabir has historically been used more widely but now only survives in this particular exclamation.

kybavux - luckily

susragemxi -  unfortunately

This only gives a sample of the gaps in Bryatesle morphology. Examples of the main types of gaps have been given, and a fair share of vocabulary to boot.

Grammatical Gender

A question that regularly pops up in conlang fora and groups is that of whether grammatical gender is good for anything.  Obviously, grammatical gender is not good enough to be universal. A few examples of languages that lack grammatical gender are

• English (with marginal gender - i.e. humans are still gendered, and you might find dialects where things that aren't neuter exist), Armenian, northern Ostrobothnian Swedish (at least Karleby - the situation is comparable to English), some Danish dialects
• Finnish, Estonian, Hungarian and related languages
• Turkish, Azeri and related languages
• Basque
• Georgian, Svan, Mingrelian

These are just the gender-less languages in or close to Europe. We find a whole slew of them elsewhere as well. Clearly a language can exist without gender, and as in the cases of Armenian and English, languages can lose gender as well. (Other examples of languages that have lost grammatical gender are Persian and some dialects of Swedish that have done so under influence from Finnish. The Danish dialects' loss of gender are more difficult to account for: the nearby region for those is quite well stacked with languages that have grammatical gender.) 

However, gender has been a remarkably stable feature of the Indo-European languages, the Semitic languages, and the Niger-Congo languages for millennia. There must be some benefits to having it. I do not posit that these benefits only can be gained from having gender, other strategies may also give the same benefits, but these benefits come bundled when you have gender - and that might be one reason why there's few signs of it disappearing from IE, Semitic and the Bantu branch of Niger-Congo, Ket, and a variety of other families.

I posit there are at least two major important advantages: added redundancy and distinguishing referents of anaphora. In addition, some minor advantages may also appear.

When we listen to, or even read a statement, most humans don't hear or register every phoneme or letter. In fact, we really just sample the utterance, and our brain fills out the remainder by some very good guesswork. However, we sometimes guess wrong, and if we are in a noisy environment, our sample may be too distorted by the background noise for us to reliably reconstruct what was uttered. (In texts, consider reading through dirty glasses, or a text that's been very worn).

In many languages with gender, gender congruence adds redundancy somewhat systematically, which helps the brain reconstruct the utterance more reliably. Errors still may happen, but the increase in redundancy reduces the risk for error. Other ways exist to reduce the risk ­- longer words, repetition of words, more 'context-sensitive' lexemes, etc, but gender is fairly efficient at it. Obviously, if no adjective or article is around to carry congruence, and there's no verb congruence, this added redundancy won't help - but chances are it'll appear often enough to make a difference often enough to be advantageous.

Another thing gender enables is a simple way of getting some of the benefits of proximate-obviate distinctions, i.e. distinguishing two (or more) third person referents. Of course gender is not the only way of resolving this: proximate-obviate systems may resolve which referent is proximate and which is obviate more reliably (i.e. there's always a way of figuring out which is which, unlike when two participants of the same gender co-occur), but this greater reliability generally requires a more complicated marking strategy, and thus the speaker must have planned-ahead far enough (unless obviative marking is mandatory on all non-proximate nouns). Obviously, gender will not enable such distinctions for every possible sentence, but there's a significant probability that two nouns you need to refer to will have different genders, and [you may sufficiently often derive enough benefit from it when they happen to be of different genders] that having grammatical genders becomes a good deal.

Conversely, gender may help distinguish homophonous nouns by means of the distinct congruence they trigger (thanks to Salmoneus for reminding me of that!). Congruence of course also is a powerful thing, see e.g. how the gender system in the Niger-Congo languages enables free word order yet no case marking is needed - the verb agrees with subject and with object, and that is generally sufficient. Some ways of achieving free word order without case and gender congruence exist (i.e. animacy hierarchies), but gender seems a rather convenient method.

These are the purely 'utilitarian' advantages of gender. Other ways of achieving the same advantages exist, but gender is a neat way of achieving several such benefits at once.

Detail #145: Splitting Nouns into Two Word Classes

I have previously mused a bit about splitting adjectives into two separate, new word classes. In part, this is me being contrarian - conlangers seem to like merging word classes, so why not go the other way.

So, let's get on it! In some way, it seems nouns, verbs and adjectives all semantically are somewhat distinct groups, so it would make sense to try and find a semantic distinction within nouns that we use for leverage in splitting the nouns in half. One very obvious distinction we could use is humans vs. non-humans or animates vs. inanimates. We'll see later that the term 'human' (or even 'animate') is misleading.

Humans (or animates) lack definiteness marking, but have a case system. Humans can be arguments of verbs.

Nonhumans (or inanimates) lack case marking, but do have definiteness marking. When definite, there's also a number distinction. They are not "arguments" of verbs. Nonhumans have a negative form.

In some sense, the nonhuman noun is somewhere inbetween adverb and noun; its relation to the verb is less clear-cut. Nonhumans often, however, combine with various adverbials to express existential things or statements about the thing

	perper	jin
(there is)	rain	around

it is raining, there is rain around

kuso	ha
water(-level)	low

the water is low (in the lake or well or whathaveyou)

tama	-de	bisi
house	DEF.PLUR	cold(ly)

the house is cold
kendi-ly

	menta	-k
rabbit	plenti

rabbits plentily - there are lots of rabbits kendinol - no rabbits!

Humans, when topics or subjects always trigger some congruence somewhere:

ergar ki jin 'soldier is around' (a/the soldier is around here somewhere)

fargan kei migi 'women are weird'

ergan kei menta 'the soldiers are many'

ba ergan kei - there are no soldiers

-p marks reduced transitivity:

'ergar tar-ki-p' - a|the soldier fights 

ergan tar-kei-p - 'soldiers fight' 

mobran tar-kei-p - 'a barbarian fights' 

ergan tar-kei-da mobrat - 'soldiers fight barbarians'

ergar tar-kei-di mobrav - 'soldiers fight a barbarian' 

ergar tar-kei-p vidgale - a soldier fights for the king

mobran suar-kei-di ba vidgat - barbarians have no kings 

In a sentence where both kinds appear, a transitive verb is usually marked for reduced transitivity:

ergar tar-ki-p kendi - soldier fights rabbit

ergar tar-di-p kendi - a rabbit fights a soldier

Verbs that we would perceive as transitive, but that normally take a nounoid object are intransitive:

lengu sea-ki kendi nagat - lengu eats rabbit (for) lunch

sea-p kendiol - there is no rabbit to eat

Anipi sea-ki kendiol - Anipi doesn't eat rabbit 

A cannibal, who eats a human, would get a marker of increased transitivity:

rauduxun sea-kei-le Dehrav - cannibals ate Dehra

Since non-humans take no case marking nor adpositions, their function often has to be figured out from the context:

ergar tar-ki-di vidgav tasdak (the soldier fights the hostile king)

ergar tar-ki-di vidgav tasdak gurba (the soldier fights the hostile king sword(ily?))

Syntactic location often gives parsing hints: non-humans often are close to human constituents that control them. There is one particle that is used with non-humans, that only serves to break such 'control'. This particle can go on either side of the word, and obviously breaks control in that direction. The case of the human that controls the non-human may actually tell us more about the role of the non-human than about the role of the human:

bliga sim-ki-di jungav piri - thief stole wizard.acc wand

junga ŋumi ra-diska-ki -di bligav - the wizard punished the thief with his (great) magic, the wizard punished the thief greatly with his magic (ra is a prefix that marks 'greatness', either of the verb or of a prominent nounoid)

The culture is very focused on humans (or at least animate things) and their affairs. Non-human (or inanimate) participants are not considered central to the events in the same way that humans are, and therefore are relegated to a separate function entirely grammatically speaking. A lot of the inanimate nounoids are derived from verbs, and the language has a rich verbal vocabulary for doing things by various means or for events that can happen. Body parts, the moon, the sun and the stars are also 'nouns' proper, whereas almost all natural phenomena and formations are relegated to being nounoids.

Ways of turning these nounoids into real nouns do exist (which makes the human<>nonhuman terminology somewhat misleading), but such constructions sound stilted to most speakers. Certain verbal nouns are in common use, but most verbs rather turn to verbal nounoids.

Nounoids generally don't take adjectives (although they can be described by adverbs); their qualities are rather implied by the verb used - the language has a number of verbal affixes that give implications as to various attributes of the most prominent nounoid. (Generally one that is either controlled by the subject or the object or one that is just left of the verb.)

This idea isn't perfectly worked out, but it's a start. I throw the idea out there mainly so that someone else would get some more impressive ideas as to how to split some common word class into two, rather than just to merge them. Maybe redraw the line in some other way, so half of the adjectives and half of the nouns form a new third class? Or maybe even half of the nouns and half of the verbs?

Bryatesle: Case Usage: The Nominative and Some Subcases

This post is part of a series on the case system of Bryatesle

I. The nominative and some subcases

II. Gaps

III. The Dative

IV. The Ablative

V. The Ergative

The case system of Bryatesle has never been quite exhaustively described - the very obsolete grammar pdf that still can be found online basically amounts to me jotting down what I had planned to do with them. So, I figure it is time to go and present it in some detail. This is the first out of several posts.

	DEF	POSS	SEC.SUBJ.	RECPRC.OBJ.	NEG.AGR.	PART	SUGGESTION
NOM	xx	xx	xx	x	-	x	x
(ERG)	x	x	x	x	x	-	x
ACC	xx	xx	x	x	x	x	x
DAT	x	x	xx	x	x	x	x
ABL	x		x	x	x	x	x
VOC	x						-
EXCL	x						-

Color coding: black x show that a form is by and large unique (with at most mergers for some gender-number combinations), any other color is shared by at least two cells, and indicates that the forms are identical, or widely identical for those cells. If a cell has both a black x and another x, this usually means that the neuter conflates these forms, but other genders do not.

Since the nominative is a somewhat exceptional case, the subcases may behave somewhat distinctly with it compared to how they behave with the other three cases. Among those three, the accusative is sometimes also likely to be exceptional. For this reason, some of the case combinations are dealt with in their own subheadings.

The Nominative

The nominative marks most subjects, neuter objects, objects of some comparisons, subject complements and sometimes things or persons that accompany the subject (a kind of morphologically unmarked comitative or instrumental) of an intransitive verb or a verb for which the noun semantically doesn't fit as subject. Such nominatives often go on the right side of the verb.

Eŗym biŗai : Erym(.NOM) wait.3sg-TELIC ≃ Erym waited, Erym will wait, Erym waits (faithfully)

Vŗih tir guşpu : rain fall.3sg.neutATELIC thunder.neut-nom ≃ rain is falling/was falling and there is/was thunder

 Eŗym zeļei tebyhaņ: Erym dance.3sg-ATELIC sandal.plur-masc-nom ≃ Erym dances (wearing) sandals

In noun phrases, the an attribute of a noun is sometimes in the nominative, regardless of the case of the head of the NP. This is mostly the situation with names before titles or other designations.

Kaŗum maxbuşŗ : Karum the baron

Gyţŗi ņedvuļy : Gytri the priest 

Lepxi gudze : Lepxi the hunter 

The titles in these  may be in any case, depending on what role the full NP has in the sentence.

The Definite

The definite corresponds fairly closely to English 'the' but is more restricted in where it can go in a sentence. The same usages as can be found with the 'naked' nominative is permissible with the definite nominative as well. However, it often implies stronger telicity, stronger likelihood of implicit past tense as well as stronger likelihood of perfective aspect. On nominal attributes, it can only be used for possessive attributes, never on other attributes.

maxbuşŗyņe : the baron, maxbusurven: the barons

guşpeş : the thunder, the thunderstrike, guspuvu: the thunderstrikes 

vŗiheş : the rain

tebyhveņ : the sandals, tebyhune, tebuhune : the sandal

With cases other than the nominative, it only has an effect on the implicit telicity or tense if the noun is a direct object or a quirky case subject or object.

The Partitive

The partitive nominative implies atelicity, less volitionality, a lower likelihood for perfectiveness, slightly reduced likelihood for implicit past tense. It may convey that the subject is not particularly committed to the action, or in the plural that only some of the group participate.

maxbuşŗu: some baron, a baron, (the baron), maxbuşŗute: some barons, barons, (the barons)

guşpur: some thunder, a thunderstorm, 'thunder (sometimes)', guşpuxa: some thunderstorms, some thunderstrikes, thunderstorms or thunderstrikes, (the) ...

vŗihuŗ: some rain, rain, a rainstorm, vŗihuxa: some rainstorms, a huge lot of rain

If there is a complement, the partitive on both the subject and complement can mark that the subject consists of the noun given as complement:

tebuxur deļyņu gahniŗ - cake.neut_part dough.fem_part bake.3sg_neut

cake is baked from dough

With copulas, subjects in the partitive might also imply that the subject only 'shares in' the qualities of the noun or that the subject only partially is such a thing. If the subject is not in the partitive but the nominal complement is, the implication rather is that the subject is increasingly such a thing.

The partitive has a somewhat similar function with regards to the direct object. With the oblique case, it mainly agrees in partitive marking with the subject or object, unless the oblique noun is marked for some other secondary case or it is individuated and known to the speaker or a topic. Such partitive agreement to subjects or objects only extends to arguments of the verb and not to attributes of nouns.

The Possessed Noun

Possession is marked on the possessee in Bryatesle. This marking is fusional with all the four main cases. The possessor may appear in the (sometimes definite) nominative, dative or ablative depending on various factors.

maxbuşŗ|maxbusze|maxbuszity ţeneļa : the wife of the baron

(ţaiņi|ţaiņe)|ţaineţa şaŗḑuvu : the nun's vows

ŗaga kinivi : the people's rights
people rights-POSS

Bryatesle avoids chains of possession - if one has to express that something is the X of the Y of Z, this would normally be done by apposition:

ŗagaveņ kinivi, (ţivi) Pargi du 

people-DEF rights-POSS, (they) Pargi associated_with

the rights of the people of Pargi

In this construction, the definite article is almost always used on the possessor. As for combinations of the possessee-marking and the other three cases, there is not particularly much to say.

Any non-subject may be marked as possessums without an explicit possessor. This implies that the subject (which may even be pro-dropped) or a topic or something marked by the secondary-subject case is its possessor.

 The Secondary Subject / Nominative

A secondary subject is a noun-phrase that carries out some action, but is not the primary agent OR whose agency is limited in some other fashion OR sometimes whose agency is just somehow not the default agency it is expected to have (may be increased!). Primacy as for agents is determined by several different factors:

• the causer in a causative construction is mostly primary
• the possessor in a construction where the possessee is unmarked for possession is secondary (e.g. kaļbunişŗ duşŗu faxḑa - fisherman-NOM-2ndsubj boat-NOM sink-TELIC), this especially if the possessor is greatly affected but unlikely to have any control over the event. If the possessor instigates his possessums to do things, the possessum is more likely to be marked for this case, and the possessor is then subject. In both situations, the possessor is likely to precede the possessum - word order helps distinguish their relation, and case marking distinguishes the agency and effect.
• subjects of verbs that are more adverb-like than predicate-like in the information structure of the sentence are always secondary (tegunişŗ bizei raga baḑies - musician-SECSUBJ play-3sg people dance-3PL - while the musician plays, the people dance; the musician playing, the people dance - this would be a secondary subject if the point of mentioning the musician and his playing simply was to provide contextual information, and not to say something about the musician himself.)
• marking the agent if an instrument has been promoted to subject (tegunişŗ beļţa haute bizeŗ: musician's lute well play ~ the lute plays well (~in the hands of the musician).
• the subject of an embedded verb (e.g. run in 'I saw her run' - (nem) vemişŗ ļelei biţaţ  ((I) her-2ndsubj run-3sg_atelic see-1sg_telic)
• subjects demoted due to voice operations
• non-subject possessors of the object. (These non-subject possessors are generally left-dislocated)

With the other cases, the secondary subject behaves sufficiently differently that it will be dealt with in separate entries.

Ćwarmin: Some Marginal Cases

I previously mentioned that Ćwarmin has a number of marginal cases. Some case forms fill gaps in the regular case system for a few nouns, others are limited to just a few nouns altogether. Here are three such marginal cases:

1. Postpositional Case

This case exists for:

• Many nouns that refer to locations that designate parts of a town or village or buildings in general (tretke - town, city, badku - village, wundarś - marketplace, open square, winćə - city gates, rumb - temple, wicxə - house, wruŋna - castle, sirni - temporary building
• agricultural locations (acres - karaž, granaries - perəc, barns - birsi, fallow land - saŋmoru, pasture land - grundu)
• waterways (iməl, river; udug lake; žewə a sound or a narrow maritime passage in general, rəige bay, cove)
• the noun 'ćan', table.

Its suffixes are: singular: -(i)ši|-(u)šu, plural: -(i)žwi|-(u)žwu, paucal: -(i)žmi|-(u)žmu

Morphologically, there's a tendence for the forms to be slightly irregular:

ćan - canšu - ćažwu - ćažmu

winćə - winši - wižwi - wižmi 

wicxe - wiči - witwi -  wimi

karaž - karažžu - karažwu - karažmu

iməl - imiši - imižwi - imižmi

grundu - grundu - gružwu - gružmu

birsi - birsi - biržwi - biržmi

tretke - treči - trežwi - trežmi 

udug - urušu - uružwu - uružmu

It is used with most postpositions.

2. By Order|Permission Of
Titles of authority, family members. There's also the word 'har' ("mr", "mrs", "ms") which serves to combine with person names and can carry this morpheme.

singular indefinite: -sada|-sədə plural indefinite: -samu|-səmi

singular definite: -sandan|-sənden plural definite: -samman|-səmmin

Thus, we get forms like 

tergisədə = by order|authority of a clergyman

jehisəmmin = by (long-standing) obligation to or grant by the royal crown

jehisənden = by order or obligation from the ruling king

Even though notionally definite, har takes the indefinite forms when combined with a name

Geres hasada - by order of (mister) Geres

tretkitite hasamman = by permission|order from the town's men

3.  'Along'
The postposition 'along' - šimšim - has merged with a few nouns.

iməl, river: singular: imitčimš plural: imiwšimš

runa,  road: singular runšumš , plural: runawšumš

wundarś, marketplace: wundruśšumš (more generally 'in the market'), no plural

grundu, pasture - grunšumš:  following a herd on a foraging trip

A Closer Look at Barxáw Spatial Prepositions

I have already provided some Barxáw prepositions, but I have not described in any closer detail the use of these. In addition, Barxáw has a bunch of pronouns / anaphoric adverbs that basically fuse 'there, that' or 'here, this' with the meaning of the prepositions.

sún ~ on

sit ~ in 

However, things are sún one's hand, not sit one's hand. Similar things go with regards  to cups, ships, etc. However, a glove is not on a hand, it is tùð that hand.

This is due to the thought model that sún, sit and tùð are selected through.

The hand has a primary surface (the palm), likewise a boat has a primary surface (the deck or the inside of the hull), a house has a primary surface (the roof), a town has a primary surface (its land area), a country likewise, a mountain likewise. Being enclosed by, covered by, or on top of the primary surface calls for sún. Thus, the drink is sún kahið, on the glass.

The primary surfaces can be somewhat odd - for a wall, the primary surface is really anything cut into the wall or on top of the wall - thus anything that is not a vertical surface. The vertical surface is an un-primary surface with regards to walls. Thus, a door is sún gahké, 'on the wall', as is a window. A painting is sit gahké, 'in the wall'.

As for the glove, its primary surface is analogous to that of the hand - viz the palm on the outside. As the primary surface is not the inside, the glove is seen as surrounding the hand and therefore tùð is used. The hand is sit the glove.

For most large things, the primary surface is basically the topmost outside surface (which can be rather large). This is less often true of smaller things, but often enough to be a good inference. Culturally significant things, however, often have lexically determined primary surfaces, and whenever one is in contact with that surface, sún is used, and sit is used for most other types of relation to the surfaces of a thing. Finally, tùð is used for certain situations where only part of a thing is covered, but no primary surface is involved with regards to the object of the preposition. Since the hand only is part of the human, the hand in the glove is a hand tùð the glove. An item hidden in the glove would be sit the glove.

Edit: inspired by Salmoneus' Rawang Ata - Prepositions

Detail #144: A Thing with Case Morphology and Adjectives

In a case system along the lines of your typical traditional Indo-European fare, how about:

use the case suffixes normally used with nouns when the adjective is a complement of a verb, and the adjectival case suffixes go on nouns when they are attributes.

Bringing Children up in your Conlang: Answers to Installment I

I have seen two comments on this post elsewhere, (viz. badconlangideas), and I figured I should probably address the comments.

I invite people to comment here if they have objections, questions and likewise.

Mathionalist writes

Interesting! I imagine the same analysis would apply to teaching your children a natural language that you are not yourself fluent in. This is why I used to worry about bringing my children up speaking Italian, since it was important to me, but I don’t have the same level of fluency as a native speaker.

I intend to deal with that later on, in a post whose topic also is applicable to exactly that situation. We have some fairly hard data on the results of doing that, and it seems it's not recommendable. Most speakers are less spontaneous in their language use in languages they have acquired after their childhood, and this reduced spontaneity affects the interaction with children in ways that might stunt their emotional development.

Lingumaniac writes

But that is only a problem when you are unwilling to alter the language in reaction to what the childs does to it. When a child acquires a language it does not simply copy it but it adjusts it. By speaking with your child in your conlang it would automatically adjust it to its brain structure, overhaul the mistakes and make it a speakable language. This is how creole languages evolved. When two parents are speaking two different languages in a special environment and are mixing them up the child gets a plethora of inconsistencies and errors. But still, what the child will start to speak is a complete, elaborated and speakable language. Teaching a child a possibly incomplete language wouldn’t lead your child to have a deficiant way of thinking. The child would make your language complete automatically, as language doesn’t shape speakability but our language facility shapes language and makes it the way it needs it. Making a conlang an active language would require the step being acquired by a child as a mother tongue. That’s how Esperanto or creoles got shaped to be speakable today.

I have italicized two points that I think fail to appreciate parts of the problem.

Yes, but languages have evolved over _large sample sizes_, whereas you'll be using a sample of one or at most a dozen or so if you're a prolific breeder. However, I already pointed out that you'll need to be somewhat consistent in how you adjust your language - children will do things 'wrong' anyway ("cutted"), so just because a kid does not get your construction right away doesn't mean it's impossible to get - you can't adjust them too early (because then your kids first attempt consistently ends up deciding how your language should be), and on the other hand you have unlearnable shit going on.

Children's language acquisition is somewhat slow - even when learning natural languages, they make errors systematically at first. They fail to place the negation particle in the right place, they fail to use morphology systematically, they overgeneralize morphology ... and only after a while do they end up constructing sentences right, and the bits fall in place slowly.

So, your child not learning the right structures might not be a sign of it being unlearnable - it might be a sign of normal linguistic development. Thus, adjusting your language too quickly is a problem, as it also reduces the consistency in the data that the child learns from. But then again, you don't know whether it's learnable until you might already have stunted the kid's development. Trying to find a balance there is not a thing you should even be doing, because there's no reasonable methodology by which it could be done without risking the child's linguistic development.

As for teaching a child an incomplete language not affecting its way of thinking, we have some pretty solid evidence that this is not so. Children that are taught a language by not-very-competent speakers are more likely to be severely stunted when it comes to emotional development, as well as the ability to discuss more complicated issues. The next post, in fact, will deal with this (and will come with sources).

Detail #143: More Subtle Inverse Number

Inverse number as described in detail #141 could be done in some more subtle way:

Have morphological number on nouns follow a regular system, but have object number congruence follow an inverse system:

pərəŋ-ga lixr-wi d-gu-miť-ig : the men have picked the fruit

pərn = man, (class I, 'primarily singular'), -ga = plural definite nominative class I

lixr = fruit (class II, 'primarily plural'), -wi = accusative class II

miť = pick, gather, harvest

 d- = object is its typical number, -gu- = plural class I subject, -ig = past tense perfect

 lixr-pe pərəŋ-žu d-pu-xxan-ig : a fruit poisoned the man

-xxan- = to poison, cause sickness, d- = object has typical number, -pu = singular class II subject

-pe = singular indefinite nominative class II, -žu = class I accusative

In this case, I've gone so far as to omit number altogether in the accusative case marking (whereas other cases will retain plural marking), but having the verb mark for number in a way that differs from the singular marking is not entirely impossible.

In fact, Georgian distinguishes 'semantic number' from 'syntactic number', and marks for both on occasion. Nouns with a numeral are always in singular form, and therefore have singular verb congruence - but some verbs have suppletive roots for  any semantically plural nouns. Thus, it has verbs that simultaneously agree with the subject in two opposite ways  - singular and plural. The system given above would seem to be, in fact, less out there than the Georgian system.

Detail #142: Phonologically Absent Inanimate Pronoun

In a language with a gender system consisting of, say, masculine, feminine, inanimate or just animate, inanimate, it is possible to pull off a system whereby the inanimates simply lack a pronoun.

If the language marks verbs for transitivity (possibly the marker is a remainder of a pronoun that historically was present), the absence of a pronoun implies inanimate object. For subjects, absence of a subject has the same effect. Prepositions/postpositions are assumed to have an object, so absence again amounts to third person inanimate.

Of course, an inanimate pronoun is likely soon to return into such a language by repurposing demonstratives or even some particle of some kind.

Detail #141: Combining Inverse Number Systems with Another System

The Inverse Number System is a typological oddity, apparently restricted to one family of languages. For slightly more information on it, read wikipedia's entry on grammatical number.

Could this be combined with some other number system? Certainly. However, there's a slight caveat - we can't really decide which system to use on the basis of which number is being marked (since that gets a bit circular and also undoes the effect entirely), deciding it based on gender also seems to undo the effect (or at least, one gender then has to have some kind of subclass system).

However, we could have some case marking follow an inverse number system. One case that might work very well for that could be the instrumental, for instance, and another imaginably good case for it could maybe be the absolutive.

So:

		singular	dual	plural
abs	I	-	-s	-s
	II	-s	-	-s
	III	-s	-s	-
	mass	-s	N/A	N/A
erg	I	-in	-il	-il
	II	-in	-ir	-il
	III	-in	-ir	-il
	mass	-i(n)	N/A	N/A
dat	I	-ek	-wel	-wel
	II	-ek	-wer	-wel
	III	-ek	-wer	-wel
	mass	-ek	N/A	N/A
loc	I	-ku	-kul	-kul
	II	-ku	-kur	-kul
	III	-ku	-kur	-kul
	mass	-kul

Another possibility would be to have the inverse system apply to definite forms, or to have definite articles' congruence go by an inverse system (but not the nouns themselves).

A Small Idea: Cardinal Directions

Four cardinal directions is the obvious idea we get from the Germanic languages. However, Finnish has eight cardinal directions (or at least, eight named directions whose names are not derived from the other directions).

All languages I know seem to form intermediate directions by compounding them - even Finnish, although it starts compounding them one step later than most (and thus no silly things like west-by-northwest or västnordväst, you just get pohjois-luode). Is this the only manner by which we could form intermediates? Some other approaches could reasonably well work - morphemes meaning 'half', 'up/over', 'down/under', 'behind', 'in front of', 'little', 'big', 'off', etc could work. Of course, in these cases arbitrary meanings have to be assigned, but that's the point, really. There isn't even necessarily any need for a consistent usage.

Further, in some places, there may even be less of a need for precise directions in part of the compass, and thus compound-terms are coined as needed for that direction, whereas greater granularity is used for another part of the compass.