Conlang Random Thread

[akam chinjir]

A few thoughts, not necessarily especially informed.

I think when you've got an overt complementiser, it's usually on the same side of the clause as the verb is. But there've got to be exceptions, so maybe it doesn't matter. (I know I've seen this pattern given as an explanation of why Mandarin doesn't let embedded questions---which follow the verb---use the clause-final complementiser ma 嗎, fwiw.)

Is your ergative case the same in form as, say, an instrumental? That'd be one way to make sense of your case-marking pattern with nominalisations. (From what you say it can't be the same as your genitive, I guess.)

My gut wants to say that nominalisations tend to be factive. I mean, if I say, "I'll cause your doing it," I'm presupposing that you'll do it; whereas if I say, "I'll cause you to do it," I'm implying that you'll do it, but not actually presupposing that. But maybe nominalisations don't work that way in all languages.

[*] Noonan states that ‘all languages have some sort of reduced complement type in opposition to the indicative’. But what exactly is a ‘reduced complement type’? Noonan defines it fairly straightforwardly as ‘any complement type that has fewer syntactic and inflectional possibilities than an indicative main clause’, but then never refers to this definition again and keeps on talking about ‘grammatical relations’, whatever those are.

That generalisation seems like it must have exceptions.

A reduced complement type would be a subjunctive or a nominalisation or something---anything without independent time reference would count.

Grammatical relations (as that expression is usually used) are whatever subject and object are.

[*] Should my s-type complement clause take case marking? Or is this a special case where languages usually omit case?

I think the issue for your language wouldn't be case-marking on the complement clause, it'd be ergative marking on the subject of the main clause. But to answer the question: you'd be safe treating verbs with clausal complements as intransitive, but there are certainly languages that treat them as transitive.

[*] For my nominalization construction, since the absolutive argument is marked as a possessor, I have a situation where there appears to be an ergative argument but no absolutive argument. Is this realistic?

Yes. As I implied above, I suspect it's safest to think of the ergative case-marking on the subject of a nominalisation as distinct from the ergative case-marking on the subject of a clause.

[*] Out of the two complement clause types I have described, I want one to have a restricted range of tense/aspect combinations available (which I’ve heard is pretty normal in complement clauses). For which clause type is this most plausible? What sort of restriction should I use? (I suspect the answer to this is in that Dixon article I keep on talking about — maybe I’ll check it later when I have more time.)

Your s-type complements have independent time reference, which suggests it'll be possible to mark them for tense. Your nominalisations don't have independent time reference, so they could easily do without separate tense marking. It certainly makes it seem like they could be more restricted than the s-type ones.

Also, from the fact that the object of a nominalised verb can't take regular absolutive case, this makes it seem like it's a fairly 'small' nominalisation, and that it shouldn't be possible for it to include tense. (This is a pretty theory-laden judgment, so feel free to disregard. The idea is that to include tense, a nominalisation must be 'higher' than TP, and anything that includes TP is big enough that there shouldn't be anything stopping the verb's object from getting absolutive case.)

I hope that makes sense, and is reasonable.

[Knit Tie]

So I've ended up with a conlang that, at least in the native vocabulary, has no phonemic liquids while also possessing a large consonant inventory. It does have three different flap varieties (or an approximant and two flaps, not sure yet) as intervocalic allophones of the voiced coronal plosives, but they are never found in clusters ever, which makes possible to analyse them as fundamentally plosives. Does that make sense? Is this attested, or at least plausible?

[bradrn]

So I've ended up with a conlang that, at least in the native vocabulary, has no phonemic liquids while also possessing a large consonant inventory. It does have three different flap varieties (or an approximant and two flaps, not sure yet) as intervocalic allophones of the voiced coronal plosives, but they are never found in clusters ever, which makes possible to analyse them as fundamentally plosives. Does that make sense? Is this attested, or at least plausible?

Could you write down the full inventory, plus those allophones you’re talking about? That might make it a bit easier to see how plausible those allophones are in the context of the inventory.
_____________________________________________

1. Yes, it looks plausible.
2. I would think Noonan's "reduced complement types" cover mostly the same ground as Dixon's non-clausal "complementation strategies". Noonan seems to focus on morphological differences to a finite main clause verb though; a subjunctive clause, for instance, might fit Noonan's category because it appears in a non-indicative mood, even though it is still a clause.
3. IIUC your s-type complement clauses can only appear in an absolutive role, so it seems likely they wouldn't take overt case marking. (But they probably could, if you wanted that.)
4. Yes, cf. English (otherwise not ergative, of course): Chelsea's defeat by Liverpool
5. In your language, the nominalization construction is the deranked/reduced variant because it's not a full clause, so I'd expect this construction to have fewer internal possibilities compared to the s-type construction (but obviously more external roles that it can be used for).

Thanks cedh! I’m glad my system is at least plausible, even if there are a couple of problems with it (as akam chinjir highlighted in his response).

I think when you've got an overt complementiser, it's usually on the same side of the clause as the verb is. But there've got to be exceptions, so maybe it doesn't matter. (I know I've seen this pattern given as an explanation of why Mandarin doesn't let embedded questions---which follow the verb---use the clause-final complementiser ma 嗎, fwiw.)

This is very interesting, but makes my situation a bit tricky. As I mentioned, my normal word order is SOV, but the complement clause is regularly extraposed to the end of the clause, giving SVO. So should my complementiser go after (from SOV) or before (from SVO) the complement?

Is your ergative case the same in form as, say, an instrumental? That'd be one way to make sense of your case-marking pattern with nominalisations. (From what you say it can't be the same as your genitive, I guess.)

I don’t have an instrumental. I have only ergative, accusative, genitive and dative. These are all marked differently (although I’d like to do something interesting with this system when I make a daughter language).

My gut wants to say that nominalisations tend to be factive. I mean, if I say, "I'll cause your doing it," I'm presupposing that you'll do it; whereas if I say, "I'll cause you to do it," I'm implying that you'll do it, but not actually presupposing that. But maybe nominalisations don't work that way in all languages.

From Noonan:

Another sort of two-member system is found in Squamish and other Native American languages, where the indicative complement is almost restricted to complements of utterance predicates; the nominalized complement type, which can express full tense–aspect and mood distinctions, is used elsewhere.

Which implies that Squamish usually uses nominalisations for non-factive meanings.

[*] Noonan states that ‘all languages have some sort of reduced complement type in opposition to the indicative’. But what exactly is a ‘reduced complement type’? Noonan defines it fairly straightforwardly as ‘any complement type that has fewer syntactic and inflectional possibilities than an indicative main clause’, but then never refers to this definition again and keeps on talking about ‘grammatical relations’, whatever those are.

That generalisation seems like it must have exceptions.

A reduced complement type would be a subjunctive or a nominalisation or something---anything without independent time reference would count.

Grammatical relations (as that expression is usually used) are whatever subject and object are.

That Squamish example I quoted would surely be a counterexample to your statement ‘anything without ITR would count [as a reduced complement type’. So don’t think this is a good criterion. On the other hand, now that I know what a grammatical relation is, I do think I have a working definition of a reduced complement type: A reduced complement is one where at least one grammatical relation can or must be omitted. This certainly works for nominalisations: reusing my example from earlier in the thread, The playing of the national anthem is fine as a nominalisation, although it omits the subject. To play the national anthem is fine as well.

[*] Should my s-type complement clause take case marking? Or is this a special case where languages usually omit case?

I think the issue for your language wouldn't be case-marking on the complement clause, it'd be ergative marking on the subject of the main clause. But to answer the question: you'd be safe treating verbs with clausal complements as intransitive, but there are certainly languages that treat them as transitive.

Why would ergative marking on the subject be a problem? Are you saying here that complementation is often a valency-reduction process for the main verb?

[*] Out of the two complement clause types I have described, I want one to have a restricted range of tense/aspect combinations available (which I’ve heard is pretty normal in complement clauses). For which clause type is this most plausible? What sort of restriction should I use? (I suspect the answer to this is in that Dixon article I keep on talking about — maybe I’ll check it later when I have more time.)

Your s-type complements have independent time reference, which suggests it'll be possible to mark them for tense. Your nominalisations don't have independent time reference, so they could easily do without separate tense marking. It certainly makes it seem like they could be more restricted than the s-type ones.

Also, from the fact that the object of a nominalised verb can't take regular absolutive case, this makes it seem like it's a fairly 'small' nominalisation, and that it shouldn't be possible for it to include tense. (This is a pretty theory-laden judgment, so feel free to disregard. The idea is that to include tense, a nominalisation must be 'higher' than TP, and anything that includes TP is big enough that there shouldn't be anything stopping the verb's object from getting absolutive case.)

Thanks! With regards to my other question ‘what sort of restrictions should I use’: I got around to looking at Dixon again, and he says it’s common for languages to have no tense/aspect marking at all in complement clauses. So I’ll just do that: in nominalisations, the verb cannot take tense/aspect affixes.

[akam chinjir]

This is very interesting, but makes my situation a bit tricky. As I mentioned, my normal word order is SOV, but the complement clause is regularly extraposed to the end of the clause, giving SVO. So should my complementiser go after (from SOV) or before (from SVO) the complement?

Putting it before the clause (next to the verb) is common. E.g., in Turkish, which is pretty consistently head-final in other respects, complement clauses can go before the verb or after; when after (and only then) they use the complementiser "ki," which precedes the complement clause.

My gut wants to say that nominalisations tend to be factive. I mean, if I say, "I'll cause your doing it," I'm presupposing that you'll do it; whereas if I say, "I'll cause you to do it," I'm implying that you'll do it, but not actually presupposing that. But maybe nominalisations don't work that way in all languages.

From Noonan:

Another sort of two-member system is found in Squamish and other Native American languages, where the indicative complement is almost restricted to complements of utterance predicates; the nominalized complement type, which can express full tense–aspect and mood distinctions, is used elsewhere.

Which implies that Squamish usually uses nominalisations for non-factive meanings.

Yeah, you're right about that.

A reduced complement type would be a subjunctive or a nominalisation or something---anything without independent time reference would count.

Grammatical relations (as that expression is usually used) are whatever subject and object are.

That Squamish example I quoted would surely be a counterexample to your statement ‘anything without ITR would count [as a reduced complement type’.

Does he say the nominalisations don't have independent time reference? Given that they can mark the full range of tense and mood distinctions, I wouldn't necessarily expect that.

I think the issue for your language wouldn't be case-marking on the complement clause, it'd be ergative marking on the subject of the main clause. But to answer the question: you'd be safe treating verbs with clausal complements as intransitive, but there are certainly languages that treat them as transitive.

Why would ergative marking on the subject be a problem? Are you saying here that complementation is often a valency-reduction process for the main verb?

I just meant that you don't have an accusative case, so any difference a complement clause would make to case-marking in the clause would show up on the subject. It's not a problem.

Anyway it all sounds good.

[bradrn]

This is very interesting, but makes my situation a bit tricky. As I mentioned, my normal word order is SOV, but the complement clause is regularly extraposed to the end of the clause, giving SVO. So should my complementiser go after (from SOV) or before (from SVO) the complement?

Putting it before the clause (next to the verb) is common. E.g., in Turkish, which is pretty consistently head-final in other respects, complement clauses can go before the verb or after; when after (and only then) they use the complementiser "ki," which precedes the complement clause.

I didn’t realise that even consistently head-final languages like Turkish put the complementiser before the clause! Maybe I’ll do that then. But do you have any idea how this is affected by the fact that I want to mark complement clauses with a clitic rather than a separate complementiser?

A reduced complement type would be a subjunctive or a nominalisation or something---anything without independent time reference would count.

Grammatical relations (as that expression is usually used) are whatever subject and object are.

That Squamish example I quoted would surely be a counterexample to your statement ‘anything without ITR would count [as a reduced complement type’.

Does he say the nominalisations don't have independent time reference? Given that they can mark the full range of tense and mood distinctions, I wouldn't necessarily expect that.

I believe Noonan says nothing about the relationship about nominalisations and ITR/DTR. He only states that in complement systems with two types, the s-like complement is often used for ITR and the nominalised complement is often used for DTR.

(By the way, I’m a bit confused by your second sentence. In particular, what does ‘that’ refer to in that sentence?)

I think the issue for your language wouldn't be case-marking on the complement clause, it'd be ergative marking on the subject of the main clause. But to answer the question: you'd be safe treating verbs with clausal complements as intransitive, but there are certainly languages that treat them as transitive.

Why would ergative marking on the subject be a problem? Are you saying here that complementation is often a valency-reduction process for the main verb?

I just meant that you don't have an accusative case, so any difference a complement clause would make to case-marking in the clause would show up on the subject.

I’m not sure I understand what the accusative case has to do with this. Are you saying that in a nominative-accusative language, there would be a difference in case-marking on the object? I’m sure I’m missing something important here, but I don’t know what…

[akam chinjir]

I didn’t realise that even consistently head-final languages like Turkish put the complementiser before the clause! Maybe I’ll do that then. But do you have any idea how this is affected by the fact that I want to mark complement clauses with a clitic rather than a separate complementiser?

Complement clauses often go after the verb in head-final languages (I think it's about 50/50), that's the part that's a bit at-odds with being head-final. Once you've got them after the verb, though, I think having an initial complementiser is very normal.

One issue is that if you start out with a clause-final complementiser, I guess it's likely to get reinterpreted as a suffix pretty soon. (In your case, can anything go between the verb and the complementiser? What sorts of morphophonological or prosodic differences are there between suffixation and encliticisation?)

Anyway I think the English complementiser "that" is usually a clitic. Probably this is very common for complementisers.

(By the way, I’m a bit confused by your second sentence. In particular, what does ‘that’ refer to in that sentence?)

"Given that they can mark the full range of tense and mood distinctions, I wouldn't necessarily expect [nominalisations to lack independent time reference]."

I just meant that you don't have an accusative case, so any difference a complement clause would make to case-marking in the clause would show up on the subject.

I’m not sure I understand what the accusative case has to do with this. Are you saying that in a nominative-accusative language, there would be a difference in case-marking on the object? I’m sure I’m missing something important here, but I don’t know what…

You had written "Should my s-type complement clause take case marking." I was answering that because you don't have an accusative case, you wouldn't expect case-marking on the complement clause regardless, though you might get ergative-marking on the subject of the main clause.

Edit: fixed a quotation snafu.

[Knit Tie]

So I've ended up with a conlang that, at least in the native vocabulary, has no phonemic liquids while also possessing a large consonant inventory. It does have three different flap varieties (or an approximant and two flaps, not sure yet) as intervocalic allophones of the voiced coronal plosives, but they are never found in clusters ever, which makes possible to analyse them as fundamentally plosives. Does that make sense? Is this attested, or at least plausible?

Could you write down the full inventory, plus those allophones you’re talking about? That might make it a bit easier to see how plausible those allophones are in the context of the inventory.

/p b pʰ t̪ d̪ t̪ʰ t d tʰ ʈ ɖ ʈʰ t͡ɕ d͡ʑ t͡ɕʰ k g kʰ/
/m n̪ n ɳ ɲ ŋ/
/f v s z ʂ ʐ ɣ h/
/a e i o u/

I'm currently thinking of having the voiced plosives lentie intervocalically: /b d̪ d ɖ g/ to /β ɾʲ ɾˠ ɽ ɣ/, with preexistent /v/ and /ɣ/ merging with /b/ and /g/ as a result.

[Nortaneous]

/d̪ d/ to /ɾʲ ɾˠ/

what

[bradrn]

I didn’t realise that even consistently head-final languages like Turkish put the complementiser before the clause! Maybe I’ll do that then. But do you have any idea how this is affected by the fact that I want to mark complement clauses with a clitic rather than a separate complementiser?

Complement clauses often go after the verb in head-final languages (I think it's about 50/50), that's the part that's a bit at-odds with being head-final. Once you've got them after the verb, though, I think having an initial complementiser is very normal.

Thinking about it this way does make a lot more sense to me.

One issue is that if you start out with a clause-final complementiser, I guess it's likely to get reinterpreted as a suffix pretty soon. (In your case, can anything go between the verb and the complementiser? What sorts of morphophonological or prosodic differences are there between suffixation and encliticisation?)

In the language I’m making, adverbs and prepositional phrases can go after the verb, so enclitics can attach to those as well. By contrast, a complementisation suffix would always attach to the end of a verb.

(On the other hand, in general it can be very hard to distinguish between words, clitics and affixes.)

Does he say the nominalisations don't have independent time reference? Given that they can mark the full range of tense and mood distinctions, I wouldn't necessarily expect that.

I believe Noonan says nothing about the relationship about nominalisations and ITR/DTR. He only states that in complement systems with two types, the s-like complement is often used for ITR and the nominalised complement is often used for DTR.

(By the way, I’m a bit confused by your second sentence. In particular, what does ‘that’ refer to in that sentence?)

"Given that they can mark the full range of tense and mood distinctions, I wouldn't necessarily expect [nominalisations to lack independent time reference]."

Thanks for explaining! But I thought that nominalisations can’t always mark the full range of tense and mood distinctions: in fact, you yourself said earlier that in the system I gave, nominalisations are more likely to lack the full range of distinctions. This seems a bit contradictory to me though, so I’m a bit confused about this now.

I just meant that you don't have an accusative case, so any difference a complement clause would make to case-marking in the clause would show up on the subject.

I’m not sure I understand what the accusative case has to do with this. Are you saying that in a nominative-accusative language, there would be a difference in case-marking on the object? I’m sure I’m missing something important here, but I don’t know what…

You had written "Should my s-type complement clause take case marking." I was answering that because you don't have an accusative case, you wouldn't expect case-marking on the complement clause regardless, though you might get ergative-marking on the subject of the main clause.

I can’t believe I forgot that the absolutive is unmarked. Sorry for the confusion!

[akam chinjir]

Thanks for explaining! But I thought that nominalisations can’t always mark the full range of tense and mood distinctions: in fact, you yourself said earlier that in the system I gave, nominalisations are more likely to lack the full range of distinctions. This seems a bit contradictory to me though, so I’m a bit confused about this now.

I think in general it's common for nominalisations to be restricted in that way; but apparently Noonan says that there's a Squamish nominalisation that expresses the full range of TAM distinctions; so that Squamish nominalisation seems like it's exceptional in this respect.

[bradrn]

Thanks for explaining! But I thought that nominalisations can’t always mark the full range of tense and mood distinctions: in fact, you yourself said earlier that in the system I gave, nominalisations are more likely to lack the full range of distinctions. This seems a bit contradictory to me though, so I’m a bit confused about this now.

I think in general it's common for nominalisations to be restricted in that way; but apparently Noonan says that there's a Squamish nominalisation that expresses the full range of TAM distinctions; so that Squamish nominalisation seems like it's exceptional in this respect.

Sorry! I forgot that you were talking about Squamish originally, and thought that ‘the nominalisations’ (earlier in the thread) referred to nominalisations generally rather than just in Squamish. I suppose this is the danger of long conversations like this one…


_______

Anyway, after 7 pages of questions, which I will readily admit is probably far too many, I think that’s all my issues with syntax resolved now. Thank you, everyone who helped answer my questions!


And now, on to a different topic. Over the past while I’ve been reading a fascinating book on modality, and it has occurred to me that this is actually a fairly important area of grammar which many conlangers nonetheless don’t pay attention to, in large part because it is a confusing topic with no easy reference to learn from. I’ve always loved the Polysynthesis For Novices thread on the old board (archvied copy), and I was thinking that I could write a similar thread on modality. Would anyone be interested in such a project?

(If I do end up doing it and it goes well, I might do another series on ergativity, another similarly underappreciated topic with many non-obvious subtleties. But I haven’t read anything on ergativity yet — I’m still only halfway through my book on modality.)

[Knit Tie]

/d̪ d/ to /ɾʲ ɾˠ/

what

Best I could think of, chief, sorry!

[Knit Tie]

/d̪ d/ to /ɾʲ ɾˠ/

what

Best I could think of, chief, sorry!

To be honest, this is pretty much exactly how my city's Russian works - the palatalized rhotic is a dental flap, the regular one is an alveolar flap. There's a clear acoustic and articulatory distinction between the two, so I thought it would be a good idea to make my conlang's POA distinction to be basically this.

[Ahzoh]

Relating to complement clauses. What is going on between a) "X is sad that Y Z" and b) "X want that Y Z"? Is the complement clause of a being used adverbially or is it the object of a transitivized intransitive verb?

Relatedly, the wiki article on balancing and deranking confuses me:

1. Modals and phasals (e.g. "I begin to run")
2. Purpose clauses (e.g. "I went into the phone booth in order to ring up my friend")
3. Desideratives (e.g. "I want to write a letter") and manipulatives (e.g. "I made John fight")
4. Perception (e.g. "I see the bus passing")
5. "Before", "when" and "after", plus nominative or absolutive relativisation.
6. Reason (e.g. "I cannot leave him alone, because he's gone mad") and reality condition (e.g. "If capitalism did not cause the Great Depression, government was responsible"), plus accusative or ergative relativisation.
7. Knowledge (e.g. "I know that the weather will be very hot") and propositional attitude (e.g. "I think that we should stay at home today"), plus oblique and indirect object relativisation.
8. Utterance (e.g. "He said that he was tired").

Is it saying the lower numbers (e.g. 1) are more likely to have limited forms (such as being obligatorily subjunctivized or one tense/tenseless) or the higher ones (e.g. 8)?

Also, for languages that have balancing and deranking, what is the most cross-linguistically common delineation at?

[akam chinjir]

Relating to complement clauses. What is going on between a) "X is sad that Y Z" and b) "X want that Y Z"? Is the complement clause of a being used adverbially or is it the object of a transitivized intransitive verb?

You're pretty safe just saying that it's a complement clause. I wouldn't say it's adverbial in either case. (But I know almost nothing about how languages other than English deal with your "sad" case.)

Relatedly, the wiki article on balancing and deranking confuses me:

[...]

Is it saying the lower numbers (e.g. 1) are more likely to have limited forms (such as being obligatorily subjunctivized or one tense/tenseless) or the higher ones (e.g. 8)?

It's saying you're more likely to get deranked forms with the lower numbers (or anyway it'd better be saying that).

[bradrn]

Relating to complement clauses. What is going on between a) "X is sad that Y Z" and b) "X want that Y Z"? Is the complement clause of a being used adverbially or is it the object of a transitivized intransitive verb?

Here Y Z is just a normal sentence, complementised by the complementiser that: I am sad that [you know it]. I’m not sure about want: for me, want can only take a to- complement clause, never a that- complement clause.

Relatedly, the wiki article on balancing and deranking confuses me:

1. Modals and phasals (e.g. "I begin to run")
2. Purpose clauses (e.g. "I went into the phone booth in order to ring up my friend")
3. Desideratives (e.g. "I want to write a letter") and manipulatives (e.g. "I made John fight")
4. Perception (e.g. "I see the bus passing")
5. "Before", "when" and "after", plus nominative or absolutive relativisation.
6. Reason (e.g. "I cannot leave him alone, because he's gone mad") and reality condition (e.g. "If capitalism did not cause the Great Depression, government was responsible"), plus accusative or ergative relativisation.
7. Knowledge (e.g. "I know that the weather will be very hot") and propositional attitude (e.g. "I think that we should stay at home today"), plus oblique and indirect object relativisation.
8. Utterance (e.g. "He said that he was tired").

Is it saying the lower numbers (e.g. 1) are more likely to have limited forms (such as being obligatorily subjunctivized or one tense/tenseless) or the higher ones (e.g. 8)?

Yes, I believe so. For instance, I begin to run uses the deranked clause to run (the infinitive can only occur in complement clauses, for the most part), whereas I He said that he was tired uses the balanced clause that he was tired (he was tired can occur pretty much everywhere).

Also, for languages that have balancing and deranking, what is the most cross-linguistically common delineation at?

That Noonan article akam chinjir linked talks a bit about this, although he doesn’t actually use the term ‘deranked’ in it. Noonan uses the terms independent time reference (ITR) and dependent time reference (DTR) to explain a common split:

• With ITR verbs, the verb has no impact on the time reference of the subordinate clause, e.g. know (I know you are there, I know you were there, I know you will be there). These verbs usually take a balanced complement clause.
• With DTR verbs, the verb does have an impact on the time reference of the subordinate clause, e.g. order (José is ordering João to interrogate Smith, *José is ordering João to interrogate Smith yesterday, José is ordering João to interrogate Smith tomorrow). These verbs usually take a deranked complement clause.

I believe this split corresponds to the split between items 3/4 on the deranking hierarchy.

[Knit Tie]

/p b pʰ t̪ d̪ t̪ʰ t d tʰ ʈ ɖ ʈʰ t͡ɕ d͡ʑ t͡ɕʰ k g kʰ/
/m n̪ n ɳ ɲ ŋ/
/f v s z ʂ ʐ ɣ h/
/a e i o u/

So anyway, are there any attested languages with a large inventory, but no liquids? Diachronically adding liquids to mine would be a trivial thing, but I wanted to see if I could make it special and skip them entirely.

[Ahzoh]

^i think i have seen some languages on wiki whose inventories show a lack of phonemic liquids.

Relating to complement clauses. What is going on between a) "X is sad that Y Z" and b) "X want that Y Z"? Is the complement clause of a being used adverbially or is it the object of a transitivized intransitive verb?

Here Y Z is just a normal sentence, complementised by the complementiser that: I am sad that [you know it]. I’m not sure about want: for me, want can only take a to- complement clause, never a that- complement clause.

I'm talking more theoretical syntax, or rather, syntax as it would be in my conlang where attributive adjectives like "sad" are monotransitive verb like "be.sad". Since complement clauses are usually treated like objects of verbs I can't determine if the complement is being treated as the object of be.sad (which would mean that "be.sad" would have to have an applicative voice marker attached to it to indicate it's now transitive) or if it's behaving adverbally, merely modifying the verb like an oblique argument might.

The "want" verb is just a normal transitive, so I know the complement clause is simply an extrapositioned object of want

[Nortaneous]

So anyway, are there any attested languages with a large inventory, but no liquids? Diachronically adding liquids to mine would be a trivial thing, but I wanted to see if I could make it special and skip them entirely.

Not that I know of. Kpelle and Ngishe are closest to what you have here, but they both have a rhotic. Prominent rhotic allophones of other consonants would probably be fine as an alternative. Would also expect non-syllabic allophones of /i u/.

[bradrn]

Relating to complement clauses. What is going on between a) "X is sad that Y Z" and b) "X want that Y Z"? Is the complement clause of a being used adverbially or is it the object of a transitivized intransitive verb?

Here Y Z is just a normal sentence, complementised by the complementiser that: I am sad that [you know it]. I’m not sure about want: for me, want can only take a to- complement clause, never a that- complement clause.

I'm talking more theoretical syntax, or rather, syntax as it would be in my conlang where attributive adjectives like "sad" are monotransitive verb like "be.sad". Since complement clauses are usually treated like objects of verbs I can't determine if the complement is being treated as the object of be.sad (which would mean that "be.sad" would have to have an applicative voice marker attached to it to indicate it's now transitive) or if it's behaving adverbally, merely modifying the verb like an oblique argument might.

My conlang is exactly the same with regards to adjectives. (In fact it takes the concept quite far: similarly to some Native American languages, in my conlang ‘adjectives’ can only modify nouns using relative clauses.) Now that I think about it, I think that in a construction like ‘X is sad that Y Z’, that is actually acting as a conjunction rather than a complementiser: you can replace that with a conjunction like because without changing the meaning at all. So, to answer your question, I believe that that Y Z is acting adverbally (specifically, as an adverbial clause) rather than as an object.