Conlang Random Thread

[Xwtek]

True enough.

I guess the case system would be more to make Kala accessible, or whatever.

My favorite approach is to use direct-inverse with a person/animacy/topicality hierarchy with core arguments, so as to free up word order to mark topicality, and to use a limited set of either coverbs or adpositions with non-core arguments (with more complex meanings being achieved through the use of inalienably-possessed relational nouns, where possession is marked on the possessum)

I really wish I understood all of this.

It's much simpler that it sounds. You're probably already using passive to encode topicality. For example, if you're talking about A playing Counter-Strike, probably you will use sentence like: Then A is shot dead by B instead of B shoots A dead. Now, make the passive completely transitive, and you will get inverse system. And you really don't know anything about adpositions?

[Vijay]

*Dzanééz ashkii yiztał.
mule boy yi-kicked
'The mule kicked the boy.'

do you know, offhand, would this be understood as "The boy kicked the mule"? Just with unusual word order?

I have no idea, sorry. I was kind of wondering the same myself. EDIT2: Actually I think I remember seeing/hearing once that this word order just isn't acceptable at all, but I could just be imagining things.

It's much simpler that it sounds. You're probably already using passive to encode topicality.

I don't get that impression from his examples where he just uses word order and cases instead... EDIT: Plus it's more complicated than that. Passives don't inherently imply an animacy hierarchy and have nothing at all to do with relational nouns.

And you really don't know anything about adpositions?

I'm not surprised if that term adds to the confusion. Adposition is not nearly as well-known a term as preposition in English.

[Travis B.]

*Dzanééz ashkii yiztał.
mule boy yi-kicked
'The mule kicked the boy.'

do you know, offhand, would this be understood as "The boy kicked the mule"? Just with unusual word order?

I have no idea, sorry. I was kind of wondering the same myself. EDIT2: Actually I think I remember seeing/hearing once that this word order just isn't acceptable at all, but I could just be imagining things.

I am not familiar with Navajo syntax or morphology, but if word order marked topicalization this would mean "The mule, the boy kicked"; however Navajo may not use word order in this fashion.

It's much simpler that it sounds. You're probably already using passive to encode topicality.

I don't get that impression from his examples where he just uses word order and cases instead... EDIT: Plus it's more complicated than that. Passives don't inherently imply an animacy hierarchy and have nothing at all to do with relational nouns.

Well, direct-inverse has nothing to do with relational nouns either; I was just giving it as an example of an alternative to a large set of locational cases or adpositions that I personally like using. But yeah, direct-inverse is distinct from having a transitive passive voice because transitive passive voices are not predicated upon the relative person/animacy/topicality of the core arguments.

And you really don't know anything about adpositions?

I'm not surprised if that term adds to the confusion. Adposition is not nearly as well-known a term as preposition in English.

I'd be surprised if anyone who'd been around here for a while were not familiar with the term, as a general term covering both prepositions and postpositions.

[Xwtek]

It's much simpler that it sounds. You're probably already using passive to encode topicality.

I don't get that impression from his examples where he just uses word order and cases instead... EDIT: Plus it's more complicated than that. Passives don't inherently imply an animacy hierarchy and have nothing at all to do with relational nouns.

Not in that language, in English or your native language (unless your native language is like Hungarian or Slavic)
Well, I don't explain relational nouns. But relational noun is actually much more boring than you thought. To spice up the conlang, you would end up better using preposition like in English. Although English does have a relational noun. (On the top of ... In front of ... etc). Passives don't inherently imply an animacy hierarchy, of course, but it has an intimate relationship with discourse information (a.k.a. topicality/obviative). Basically Inverse = Symmetric Passive with obligatory passivization for 3rd person subject and 1st/2nd person object.

[akam chinjir]

Although English does have a relational noun. (On the top of ... In front of ... etc).

"sake" is another good relational noun in English ("for Pete's sake").

[Travis B.]

Relational nouns are something that should be familiar to every English-speaker, and yet are infrequently discussed in discussions about English grammar, to the point that native Engilsh-speakers likely are not familiar with the concept even though they use them every day .

[Vijay]

Although English does have a relational noun. (On the top of ... In front of ... etc).

"sake" is another good relational noun in English ("for Pete's sake").

How is that a relational noun? Isn't that just a set expression?

[Travis B.]

Although English does have a relational noun. (On the top of ... In front of ... etc).

"sake" is another good relational noun in English ("for Pete's sake").

How is that a relational noun? Isn't that just a set expression?

"Sake" as in "for X's sake" seems like a relational noun to me, but the specific phrase "for Pete's sake" is just a set expression that makes use of said relational noun.

[Vijay]

[WeepingElf]

"sake" is another good relational noun in English ("for Pete's sake").

How is that a relational noun? Isn't that just a set expression?

"Sake" as in "for X's sake" seems like a relational noun to me, but the specific phrase "for Pete's sake" is just a set expression that makes use of said relational noun.

My main conlang Old Albic has no true prepositions, so all the jobs not taken care of by the case system (which, with ten cases, is quite extensive, but not as much as those found in some Uralic, let alone Northeast Caucasian languages) are taken up by relational nouns. So for instance,"on the box" is expressed as top-LOC the-LOC box-LOC-LOC (the language has Suffixaufnahme).

[Travis B.]

Conversely, a number of my languages, including the ones I have been working on most recently, proto-Rgyam and Leha, lack both true adpositions and case, and rather rely on (for core arguments) direct-inverse and (for non-core arguments) coverbs and relational nouns (note that Leha also relies upon word order, in that core arguments other than the agent and the direct object must come after the subject and direct object). (The coverb-iness of the coverbs is clear from the fact that they can be used as main verbs as well.)

[Zaarin]

I have a script question.

So for some quick background, the classical language has a generally CV(C) syllable structure--CVCC occurs very, very rarely, chiefly in a specific verbal conjugation--with a consonant inventory on the smaller side of average and a four vowel system (/a e i u/) plus length. It has two genders (masculine, feminine), and three cases (nominative, genitive, oblique) with only a single declension--to my knowledge there are no irregular nouns. The language is written as a syllabary, and literacy is chiefly restricted to professional scribes and priests.

So question 1: Since gender and case are immutable, is it reasonable to mark them with their own specific glyphs? (I know Syriac does the former with a feminine marker.)

Now, the modern language has lost the case system, relying instead on strict word order. Writing, however, remains quite conservative, still in the hands of the scribes and priests (as the classical language also continues to be used as a liturgical language). Is it reasonable that it would continue to write the case markers?

[masako]

Is it reasonable that it would continue to write the case markers?

Of hand, I'd say yes. They might begin to look more like punctuation after a time depending on the design parameters of the syllabary.

is it reasonable to mark them with their own specific glyphs?

This depends greatly on how the syllabary evolved. If it was derived from a ideography, then it's a resounding yes. If it was extrapolated from an abugida or complex alphabet, then I would say no.

[Knit Tie]

Hey folks, stupid noob question time! Conlangs for a DnD game, I'd like to get some feedback on the phonologies and general ideas for conlangs for my D&D game.

Common, AKA Hann Basu [xanˀ pasu]

m mˀ n nˀ
p pʰ p' t tʰ t' k kʰ k' (ʔ)
f θ s x
ʋ~w j ɰ
a e i o u

Three consonant series, lenis-aspirated-ejective, allophonic voicing in the lenis series. /θ/ is retained fully only in the official/formal reɡister, in all other ones it's usually merɡed with /s/, /ɰ/ is usually elided in coda and clusters. [ʔ] is the allophone of zero onset and is inserted to break vowel hiatus, but is not really a separate phoneme. Glottalised nasals pattern with ejective plosives. (C)V(C) phonotactics, analytical with some derivational morphemes and noun compoundinɡ, especially in ɡiven names, but is otherwise isolatinɡ. Nominative - accusative, SVO, pervasive zero markinɡ and consequtively ironclad word order.


Holy speech, AKA Draconic (draɡon ɡods, nothinɡ special), AKA Hană'a Kă'aɡa [xanəʔa kəʔaɡa]

m n
p pʰ b t tʰ d k kʰ ɡ ʔ
ts~tʃ
f s ʃ x
ʋ j l
a e i o u + lenɡth and ə

The older version of the Xann Basa, artificially preserved as a liturɡical lanɡuaɡe. Morpholoɡically a bit fusional, at the very end of the transition to isolatinɡ and analytical morpholoɡy. (C)V(C) phonotactics. /ə/ elides frequently in VCəCV environment. /ʔ/ and plosive clusters are precursors to Xann Basu ejectives and /ʔ/ and nasal clusters are precursors to its ɡlottalised nasals, in all other cases /ʔ/ elided in clusters durinɡ the Hann Basu diachronic development. The affricate deaffricated, and /ʃ/ merɡed with /s/ later on, while the voiced series lenited to /w ð ɰ/. /w/ later on merɡed with /ʋ/ and /ð/ devoiced to /θ/. The vowel lenɡth was simply lost, as was the schwa.


Chiirowa sapaa - the language of a local magical warrior tribe that I made in order to have monks who aren't the usual D&D beard-stroking wise Wuxia type. The language itself predates the Hann Basu and is from a completely unrelated family, but it's been in close liguistic contact with it for many centuries and most of its users are nowadays bilingual.

m n ɲ
p pʰ t tʰ k kʰ
ts tʃ
f s ʃ h
w j l r
a e i o u + lenɡth

Phonotactics are (C)(R)V(R)(C), where R is /w j l r/. Morphologically is very very fusional, with tons of derivational morphemes, three numbers (1, 2, all~many) and pervasive agreement between everything in gender, number and tense. Six tenses, of which three are present (present continuous, present discontinuous and present sporadic, which indicate how often the action takes place, with discontinuous indicating that the action will soon end or has just began, continuous being basically the same as in English and sporadic being everything else), two are future (continuous and sporadic) and one is undefined, which is used in all verbal derivations, the subjunctive mood, the imperative mood and also to form the past. Two genders, the female one is default. SOV word order, but it's very malleable.


Iljfijskaja mowa - Slav elves, because why not and also I wanted to dick around with my native Russian.

p b t d k ɡ
m n ɲ
ts dz tʃ dʒ
f s ɕ ʂ x
l ʎ r
w j
a e i o u

(C)(C)(C)(C)V(C)(C) phonotactics, fusional everything (like Russian), but already well on its way into simplifying and becoming analytical and isolating. No genders, three cases and no tenses. Agreement mostly optional on everything except for the subject and the verb, with most adjectives already becoming limited to only two cases in singular. Distinction between perfective and imperfective aspects in verbs retained and used in lieu of proper tenses to form tenses, with the past tense additionally being formed by the addition of the particle /tada/, "back then."


Temnetskaja mawa - because Drow refugees.

p b t d k ɡ
m n ɲ
ts tɕ tʂ
f s z ɕ ʑ ʂ x
l ʎ r
w j
a e i o u

Overall, a very slightly different version of the Slav elf speech, differing mostly is minor details and being slightly more conservative. Most notable difference is retaining genders in imperfective verbs and, phonologically, deaffrication of voiced affricates and development of two postalveolar voiceless ones.

[Kuchigakatai]

Now, the modern language has lost the case system, relying instead on strict word order. Writing, however, remains quite conservative, still in the hands of the scribes and priests (as the classical language also continues to be used as a liturgical language). Is it reasonable that it would continue to write the case markers?

Yes. In fact, said scribes/priests may even pronounce the case markers at least some of the time if they aren't too phonologically awkward.

A natlang example would be standard Arabic as it's read aloud most of the time, where, in the "sound masculine plural" declension, nominative <-wn> and accusative/genitive <-yn> are distinguished as [uːn] and [iːn] (in Classical Arabic these would've been [uːnæ] and [iːnæ] a lot of the time), even though in modern colloquial Arabic there's no nominal case and only [iːn] is used. (Case distinctions in other declensions are generally not made in pronunciation if they don't have a written representation though.)

By "too phonologically awkward" I mean something like the verbal systems of Tibetan and French. Old Tibetan had verbal conjugations such as [zgrup] 'I/etc. finish', [bzgrup] 'I/etc. will finish', [bzgrups] 'I/etc. finished', [zgrups] 'finish!'. Modern Tibetan pronunciation in Lhasa has very strict phonological constraints, so while all four mentioned forms are distinguished in writing they are all merged together as [ʈʂup˨]. I suppose they could (quite artificially) distinguish them at least some of the time by saying something like [baʈʂup˨] for old [bzgrup] and [ʈʂup˨sa] for old [zgrups], but there's a strong tradition of pronouncing the old syllable groups with a single syllable in modern pronunciation, and there's no way the old clusters are going to be revived in Lhasa (*[pʈʂups˨]???).

The Old French 3PL marker in most tenses ended in -ent, originally an unstressed [ənt]. Old French admitted a variety of closed unstressed syllable types after the stressed vowel as long as they contained a schwa, such as [mət] or [bləs] (in fact, a few words even admitted two syllables due to certain constraints regarding consonant clusters: ymagenes [iˈmadʒənəs] 'images', ydeles [ˈidələs] 'idols'). Most word-final consonants were then deleted in most contexts, and the remaining final [ə] of unstressed syllables largely decayed afterwards as well, leaving the -ent 3PL marker without much phonetic correspondence other than preserving the previous consonant: aiment [ɛm] (aim-ent, but *aim would be [æ̃]), coupent [kup] (coup-ent, but coup alone would be [ku]), joignaient [ʒwaɲɛ] (joignai-ent). I suppose the French could (quite artificially) distinguish them at least some of the time saying [ɛmn^ə], [kupn^ə] and [ʒwaɲɛn], seeing that they have words like "hymne" [imn^ə], but the sound loss has been upheld, and words ending with the segments {single consonant} + [n] such as "hymne" are rare.

[Xwtek]

Is there any tips in designing preposition sets? In my language, it's decided to use preposition instead of relational noun. To make the preposition more unique. I decided to make it more verbal-like. For example:

Longi køǿn líén/líéh

Longi 0-køøn-◌́́ líén/líéh

Longi 3SG-in-3SG village.LOC/ACC.SG

Longi ɡbaaf køǿn líén/líé (Longi stays in the village)

Longi 0-gbaaf 0-køøn-◌́́ líén/líé

Longi 3SG-stay 3SG-in-3SG village.LOC/village.ACC.SG

Although, the second sentence is more naturally expressed as:
Longi ɡbaaf líén (Longi stays in the village)

Not all nouns can receive locative case, so preposition is used in those cases. This makes the alienable possession and the verb to have is identical.

Longi té xkóot (Longi has sword)
xkóot stés Longi (Longi's sword)

Also, is there any language where the distiction of here and there is encoded via verb instead of demonstrative pro-adverb.

[akam chinjir]

Is there any tips in designing preposition sets?

This is something I puzzle about, too.

One thing: I'm pretty sure adpositions never agree with a subject. (They can agree, but only ever with a complement.) So it looks like you've actually got verbs, not verblike prepositions, fwiw.

[Xwtek]

Do you have any resource on making a grammar. Preferably a dependency grammar one. Because it turns out that my language has much more complicated grammar than I thought. It involves nonconfigurationality, for example:

Longi í-géekízen k-u-dax Fet
Longi 3SG.OBV-say-3SG.OBV COMP-3SG.PROX-hunt Fet

Fet says that Longi is hunting. (Note that Longi is the subject of the í-dax) Noun-wise, the obviative is unmarked. However, it's reflected on word order.

However, you can't just scramble word order:

k-u-dax Fet en-gée-'é Longi
Is ungrammatical.

Not even after fixing obviativity
k-en-dax Fet en-gée-'é Longi

In my mind, this language has V2 word order with the first word is the proximate noun followed by relativizer. Only one noun can be extracted out from the clause. (The proximate one) The relativizer is more rigid one. Unlike most amerindian languages, this language uses gap strategy, not nonreduction. This language is mildly polysynthetic, but many things that is expressed by suffix in more polysynthetic language is either expressed by particle or periphrastically. For example the verb try in sentence "I am trying to help him" is expressed as modal verbs instead of suffix.

[Xwtek]

Is there any tips in designing preposition sets?

This is something I puzzle about, too.

One thing: I'm pretty sure adpositions never agree with a subject. (They can agree, but only ever with a complement.) So it looks like you've actually got verbs, not verblike prepositions, fwiw.

Well, when used as preposition, the subject slot is obligatory 0- (the same-subject prefix) even if the subject is actually obviate. It's not allowed to carry something like ɬ-u- (DS-3SG.PROX).

[Zaarin]

Is it reasonable that it would continue to write the case markers?

Of hand, I'd say yes. They might begin to look more like punctuation after a time depending on the design parameters of the syllabary.

is it reasonable to mark them with their own specific glyphs?

This depends greatly on how the syllabary evolved. If it was derived from a ideography, then it's a resounding yes. If it was extrapolated from an abugida or complex alphabet, then I would say no.

It was indeed derived from an ideography, so that definitely answers my question.

Now, the modern language has lost the case system, relying instead on strict word order. Writing, however, remains quite conservative, still in the hands of the scribes and priests (as the classical language also continues to be used as a liturgical language). Is it reasonable that it would continue to write the case markers?

Yes. In fact, said scribes/priests may even pronounce the case markers at least some of the time if they aren't too phonologically awkward.

Yes, I imagine they would. Thanks!