Zompist Bboard Again

I have not done much conlanging as of late, but as long as I am now (as I am kind of stuck on my latest software project), I figured I would go over Laqar some. It is a highly fusional synthetic language (you might say it is polysynthetic since it has polypersonal agreement, but I do not really like that term, so I will omit that here) with highly opaque morphology, yet it also has a good number of analytic forms, particularly with regard to aspect/mood and evidentials.

Here is the consonant inventory of Laqar:

	labial	alveolar	retroflex	palatal	velar	labiovelar	uvular	glottal
nasals	m ⟨m⟩	n ⟨n⟩
voiced plosives	b ⟨b⟩	d dz ⟨d j⟩	dʐ ⟨ǰ⟩		g ⟨g⟩
voiceless aspirated plosives	pʰ ⟨p⟩	tʰ tsʰ ⟨t c⟩	tʂʰ ⟨č⟩		kʰ ⟨k⟩		qʰ ⟨q⟩	ʔ ⟨∅⟩
ejective plosives	pʼ ⟨p'⟩	tʼ tsʼ ⟨t' c'⟩	tʂʼ ⟨č'⟩		kʼ ⟨k'⟩		qʼ ⟨q'⟩
voiced fricatives		z ⟨z⟩	ʐ ⟨ž⟩	ʑ ⟨ź⟩			ʁ ⟨ġ⟩
voiceless fricatives		s ⟨s⟩	ʂ ⟨š⟩	ɕ ⟨ś⟩			χ ⟨x⟩	h ⟨h⟩
liquids		ɾ l ⟨r l⟩
semivowels				j ⟨y⟩		w ⟨w⟩

Here is the oral vowel inventory of Laqar:

	front	central	back
close	i ⟨í/i⟩		u ⟨ú/u⟩
close-mid	e ⟨é/e⟩		o ⟨ó/o⟩
mid		ə ⟨ə⟩
open-mid	ɛ ⟨è/ê⟩		ɔ ⟨ò/ô⟩
open		a ⟨á/a⟩

Here is the nasal vowel inventory of Laqar:

	front	central	back
open-mid	ɛ̃ ⟨ẽ/ę⟩		ɔ̃ ⟨õ/ǫ⟩
open		ã ⟨ã/ą⟩

In the two tables above, the first of each orthographic pair marks the vowel when stressed, and the second of each pair marks the vowel when unstressed. /ə/ is never stressed. Also note that stress is not marked on monosyllabic words.

The basic syllable structure of Laqar is CV(C), with hiatus not being permitted and geminates being permitted between syllables.

Word order in Laqar typically marks topicality, but when it is not emphasizing a topic it tends to be SOV (the verb complex strongly tends to come at the end of the clause, while the S and O order is relatively free).

Laqar has an interesting alignment system, with three cases being used for core nominal arguments, an agentive case, a patientive case, and a null case (for lack of a better name). For transitive verbs where the agent is higher on the person/animacy/definiteness/topicality hierarchy than the patient, both are in the null case. For volitional transitive verbs where the agent is lower on said hierarchy than the patient, the patient is put in patientive case. For avolitional transitive verbs where the agent is lower on said hierarchy than the patient, the agent is put in the agentive case. For avolitional intransitive verbs, inanimate subjects are unmarked and animate subjects are put in patientive case. For volitional intransitive verbs, animate subjects are unmarked and inanimate subjects are put in agentive case.

The basic person/animacy hierarchy has the following order:

1. 1st person
2. 2nd person
3. 3rd person pronouns (i.e. demonstratives)
4. People and gods
5. Non-human, non-diety animates
6. Inanimates

This is combined with the following other hierarchies:

1. Named
2. Definite
3. Indefinite non-gerund
4. Indefinite gerund

and:

1. Singular or singulative number
2. Plural or collective number

and:

1. Topicality (word order)

Verbs have an inherent default perfective or imperfective mood, and a morphological change (in Proto-Laqar a distinct morpheme) is needed to change them from one into the other. Also, verbs have an inherent mood, with perfective verbs defaulting to past tense and needing a morphological change to become non-past tense (which for perfective verbs implies the future), and with imperfective verbs defaulting to present tense and needing a morphological change to become past tense.

I am not going to go over the details of verb morphology in this post because it is far too complex to summarize quickly.

For a more complete list of cases, see below:

null	∅
agentive	hə
patientive	no
genitive/dative	re
instrumental/comitative	lo
locative	k'ə
allative	ča
ablative	k'əl

Note that properly the cases are postpositions, as they attach to NPs and they do not influence stress placement within the nouns to which they are adjacent.

Laqar has developed an evidential particles from a past system of impersonal verbs used in an evidential fashion. These are:

direct knowledge, egophoric	go
personal observation	ze
reportative	šol
deductive	sa
dubitative	ja
assumption	lo

These go at the end of the clause, except before interrogative nə.

Laqar similarly has a number of aspect/modal particles, that go before the evidential particle. These are:

prospective aspect	nê
retrospective aspect	lê
continuative aspect	ju
inchoative aspect	si
cessative aspect	ya
they are able (to)	hi
it is possible (that)	rô
it is necessary (that)	gô
it is probable (that)	c'i
it is supposed to be (that)	lô
it tends to be (that)	źi
it is allowed (that)	tê

These particles can be combined with one another.

Looks interesting so far — I’m looking forward to seeing more of this! A couple of (very small) questions/comments:

Travis B. wrote: ↑Mon Jul 06, 2020 10:53 pm

	labial	alveolar	retroflex	palatal	velar	labiovelar	uvular	glottal
nasals	m ⟨m⟩	n ⟨n⟩
voiced plosives	b ⟨b⟩	d dz ⟨d j⟩	dʐ ⟨ǰ⟩		g ⟨g⟩
voiceless aspirated plosives	pʰ ⟨p⟩	tʰ tsʰ ⟨t c⟩	tʂʰ ⟨č⟩		kʰ ⟨k⟩		qʰ ⟨q⟩	ʔ ⟨∅⟩
ejective plosives	pʼ ⟨p'⟩	tʼ tsʼ ⟨t' c'⟩	tʂʼ ⟨č'⟩		kʼ ⟨k'⟩		qʼ ⟨q'⟩
voiced fricatives		z ⟨z⟩	ʐ ⟨ž⟩	ʑ ⟨ź⟩			ʁ ⟨ġ⟩
voiceless fricatives		s ⟨s⟩	ʂ ⟨š⟩	ɕ ⟨ś⟩			χ ⟨x⟩	h ⟨h⟩
liquids		ɾ l ⟨r l⟩
semivowels				j ⟨y⟩		w ⟨w⟩

I see that the glottal stop is left unwritten. Does that mean that its presence is predictable?

Laqar has an interesting alignment system, with three cases being used for core nominal arguments, an agentive case, a patientive case, and a null case (for lack of a better name). For transitive verbs where the agent is higher on the person/animacy/definiteness/topicality hierarchy than the patient, both are in the null case. For volitional transitive verbs where the agent is lower on said hierarchy than the patient, the patient is put in patientive case. For avolitional transitive verbs where the agent is lower on said hierarchy than the patient, the agent is put in the agentive case. For avolitional intransitive verbs, inanimate subjects are unmarked and animate subjects are put in patientive case. For volitional intransitive verbs, animate subjects are unmarked and inanimate subjects are put in agentive case.

That’s a pretty… well, interesting system there. For my own convenience, I think I’ll summarize it in a table:

	Intransitive, S animate	Intransitive, S inanimate	A>O on hierarchy	A<O on hierarchy
Volitional	No marking	S in agentive case	No marking	A unmarked, O in patientive case
Avolitional	S in patientive case	No marking	No marking	A in agentive case, O unmarked

This looks like nothing I’ve ever seen in a natlang before, yet for some reason it doesn’t strike me as being excessively unrealistic. Well done on finding a system which does that!

The basic person/animacy hierarchy has the following order:

1. 1st person
2. 2nd person
3. 3rd person pronouns (i.e. demonstratives)
4. People and gods
5. Non-human, non-diety animates
6. Inanimates

This is combined with the following other hierarchies:

1. Named
2. Definite
3. Indefinite non-gerund
4. Indefinite gerund

and:

1. Singular or singulative number
2. Plural or collective number

and:

1. Topicality (word order)

A question: how do these hierarchies interact with each other? I’d guess the procedure is something like ‘If two nouns are at the same position on the animacy hierarchy, then compare their positions on the definiteness hierarchy; if they’re at the same position on the definiteness hierarchy, then compare their positions on the number hierarchy; and if they’re at the same position on the definiteness hierarchy, then compare their positions on the topicality hierarchy’, but I’m not entirely sure about that.

Verbs have an inherent default perfective or imperfective mood, and a morphological change (in Proto-Laqar a distinct morpheme) is needed to change them from one into the other. Also, verbs have an inherent mood, with perfective verbs defaulting to past tense and needing a morphological change to become non-past tense (which for perfective verbs implies the future), and with imperfective verbs defaulting to present tense and needing a morphological change to become past tense.

Wouldn’t perfective/imperfective be an aspectual rather than modal distinction?

Whoops I miswrote aspect as mood there!

I've always loved this language, though I'm more familiar with Old Laqar, specifically its highly derived inflection paradigms. From this, all I have to say is that the animacy hierarchy is interesting.

1) I dig the lack of retroflex stops.

2) The case system is indeed an interesting case. ( ) So basically, things get marked when they go against expectation?

3) For the hierarchy, do the subsequent hierarchies only get considered if the previous doesn't resolve which constituent is considered higher? So basic list > definite list > number list > topicality?

Nice work so far!

------

As an aside, it seems like there's quite a bit of interesting & fun conlanging coming out of the ZBB lately. I think maybe COVID has something to do with that???

bradrn wrote: ↑Mon Jul 06, 2020 11:29 pm Looks interesting so far — I’m looking forward to seeing more of this! A couple of (very small) questions/comments:

Travis B. wrote: ↑Mon Jul 06, 2020 10:53 pm

	labial	alveolar	retroflex	palatal	velar	labiovelar	uvular	glottal
nasals	m ⟨m⟩	n ⟨n⟩
voiced plosives	b ⟨b⟩	d dz ⟨d j⟩	dʐ ⟨ǰ⟩		g ⟨g⟩
voiceless aspirated plosives	pʰ ⟨p⟩	tʰ tsʰ ⟨t c⟩	tʂʰ ⟨č⟩		kʰ ⟨k⟩		qʰ ⟨q⟩	ʔ ⟨∅⟩
ejective plosives	pʼ ⟨p'⟩	tʼ tsʼ ⟨t' c'⟩	tʂʼ ⟨č'⟩		kʼ ⟨k'⟩		qʼ ⟨q'⟩
voiced fricatives		z ⟨z⟩	ʐ ⟨ž⟩	ʑ ⟨ź⟩			ʁ ⟨ġ⟩
voiceless fricatives		s ⟨s⟩	ʂ ⟨š⟩	ɕ ⟨ś⟩			χ ⟨x⟩	h ⟨h⟩
liquids		ɾ l ⟨r l⟩
semivowels				j ⟨y⟩		w ⟨w⟩

I see that the glottal stop is left unwritten. Does that mean that its presence is predictable?

Its presence is predictable in that it is found wherever Laqar's phonotactics would be violated, i.e. it is only found initially or between vowels.

bradrn wrote: ↑Mon Jul 06, 2020 11:29 pm

Laqar has an interesting alignment system, with three cases being used for core nominal arguments, an agentive case, a patientive case, and a null case (for lack of a better name). For transitive verbs where the agent is higher on the person/animacy/definiteness/topicality hierarchy than the patient, both are in the null case. For volitional transitive verbs where the agent is lower on said hierarchy than the patient, the patient is put in patientive case. For avolitional transitive verbs where the agent is lower on said hierarchy than the patient, the agent is put in the agentive case. For avolitional intransitive verbs, inanimate subjects are unmarked and animate subjects are put in patientive case. For volitional intransitive verbs, animate subjects are unmarked and inanimate subjects are put in agentive case.

That’s a pretty… well, interesting system there. For my own convenience, I think I’ll summarize it in a table:

	Intransitive, S animate	Intransitive, S inanimate	A>O on hierarchy	A<O on hierarchy
Volitional	No marking	S in agentive case	No marking	A unmarked, O in patientive case
Avolitional	S in patientive case	No marking	No marking	A in agentive case, O unmarked

This looks like nothing I’ve ever seen in a natlang before, yet for some reason it doesn’t strike me as being excessively unrealistic. Well done on finding a system which does that!

Note that the verb agreement is such that the transitive verb treats its argument that is higher on the hierarchy as its "first" argument, marked the same way as the argument of an intransitive verb, and the other as its "second" argument, marked like the possessor of a noun.

bradrn wrote: ↑Mon Jul 06, 2020 11:29 pm

The basic person/animacy hierarchy has the following order:

1. 1st person
2. 2nd person
3. 3rd person pronouns (i.e. demonstratives)
4. People and gods
5. Non-human, non-diety animates
6. Inanimates

This is combined with the following other hierarchies:

1. Named
2. Definite
3. Indefinite non-gerund
4. Indefinite gerund

and:

1. Singular or singulative number
2. Plural or collective number

and:

1. Topicality (word order)

A question: how do these hierarchies interact with each other? I’d guess the procedure is something like ‘If two nouns are at the same position on the animacy hierarchy, then compare their positions on the definiteness hierarchy; if they’re at the same position on the definiteness hierarchy, then compare their positions on the number hierarchy; and if they’re at the same position on the definiteness hierarchy, then compare their positions on the topicality hierarchy’, but I’m not entirely sure about that.

The hierarchies are prioritized in the order listed, as you figured.

Pabappa wrote: ↑Tue Jul 07, 2020 7:45 am I've always loved this language, though I'm more familiar with Old Laqar, specifically its highly derived inflection paradigms. From this, all I have to say is that the animacy hierarchy is interesting.

Properly this is Early Middle Laqar, which I refer to as just Laqar as I haven't made any newer stages. And yes, it has even more opaque inflection paradigms than Old Laqar.

Vardelm wrote: ↑Tue Jul 07, 2020 7:53 am 2) The case system is indeed an interesting case. ( ) So basically, things get marked when they go against expectation?

In essence, yes - it is based on the idea of markedness, and is inspired by (but different from both of) direct-inverse and fluid-S alignments.

Vardelm wrote: ↑Tue Jul 07, 2020 7:53 am 3) For the hierarchy, do the subsequent hierarchies only get considered if the previous doesn't resolve which constituent is considered higher? So basic list > definite list > number list > topicality?

Precisely.

I was going to write something on nominal morphology in Laqar here, but after I wrote it my connection failed and I lost what I had written. Anyways, nouns in Laqar are marked for gender (masculine versus feminine), number (singular versus plural and collective versus singulative), definiteness (indefinite versus definite), and possession (1st versus 2nd m. versus 2nd f. versus 3rd m. versus 3rd f. and sg. versus pl.). The morphemes for these are suffixed in the following order: number/gender, possession, and definiteness/number/gender. The system for marking these on nouns is essentially directly inherited from Proto-Laqar, but is complicated by all the sound changes since then (even though it is still much more transparent than verbal morphology). I will go over the details of nominal morphology later, as I do not have the time to write anything detailed out again at the moment.

Travis B. wrote: ↑Tue Jul 07, 2020 5:12 pmmy connection failed and I lost what I had written.

Ugh. Always painful.

Here is the relevant Proto-Laqar nominal morphology:

Number / Gender

	sg.	pl.	col.	sgv.
m.	-∅	-itʰ	-∅	-wi
f.	-∅	-læ	-∅	-iː

Possession

	sg.	pl.
1st	-læː	-niː
2nd m.	-iː	-bu
2nd f.	-æː	-ɾæ
3rd m.	-wæ	-u
3rd f.	-ɾiː	-χɒː
refl.	-miː	-miː

Definite / Number / Gender

	sg./col./sgv.	pl.
m.	-dæ	-dæɾ
f.	-di	-dim

A possible typo: in the third person feminine plural possessive form, shouldn’t /ɒ/ be /ɔ/? (/ɒ/ isn’t listed in the phonology, but /ɔ/ is.)

EDIT And another phonology-related question: why do some of those affixes have long vowels when the phonology doesn’t list any long vowels?

Proto-Laqar adjectival morphology is rather simple:

	sg./col./sgv.	pl.
m.	-æ	-dæm
f.	-i	-im

bradrn wrote: ↑Tue Jul 07, 2020 8:12 pm A possible typo: in the third person feminine plural possessive form, shouldn’t /ɒ/ be /ɔ/? (/ɒ/ isn’t listed in the phonology, but /ɔ/ is.)

EDIT And another phonology-related question: why do some of those affixes have long vowels when the phonology doesn’t list any long vowels?

Because these are Proto-Laqar, not (Early Middle) Laqar forms; I am presenting these forms because it is hard to understand the later forms without first seeing the proto-forms.

Travis B. wrote: ↑Tue Jul 07, 2020 8:15 pm

bradrn wrote: ↑Tue Jul 07, 2020 8:12 pm A possible typo: in the third person feminine plural possessive form, shouldn’t /ɒ/ be /ɔ/? (/ɒ/ isn’t listed in the phonology, but /ɔ/ is.)

EDIT And another phonology-related question: why do some of those affixes have long vowels when the phonology doesn’t list any long vowels?

Because these are Proto-Laqar, not (Early Middle) Laqar forms; I am presenting these forms because it is hard to understand the later forms without first seeing the proto-forms.

Oh, right — I didn’t notice that, sorry!

Later stages of Laqar's relation to Proto-Laqar are what Old Irish was to Primitive Irish - hard to understand without deriving everything from the proto-forms.

Travis B. wrote: ↑Tue Jul 07, 2020 8:17 pm Later stages of Laqar's relation to Proto-Laqar are what Old Irish was to Primitive Irish - hard to understand without deriving everything from the proto-forms.

Oh, dear. Would that be this Old Irish? If so, then that doesn’t bode well for the continued sanity of Laqar speakers…

bradrn wrote: ↑Tue Jul 07, 2020 8:35 pm

Travis B. wrote: ↑Tue Jul 07, 2020 8:17 pm Later stages of Laqar's relation to Proto-Laqar are what Old Irish was to Primitive Irish - hard to understand without deriving everything from the proto-forms.

Oh, dear. Would that be this Old Irish? If so, then that doesn’t bode well for the continued sanity of Laqar speakers…

Yes, that Old Irish.

Just for an example of the insanity that lies in wait when one considers Laqar verbal morphology, consider this...

(I take back what I said about verb agreement being based on hierarchy position - I didn't remember right - it really is based on A versus P now that I remember correctly.)

Travis B. wrote: ↑Tue Jul 07, 2020 9:01 pm consider this...

OW!!!

(Consider that a conlanging compliment.)