Zompist Bboard Again

Can you even be precise about how many registers a language has? I would have thought whether a language has, say, three or four registers is a matter of interpretation.

Raphael wrote: ↑Sun May 19, 2024 12:10 pm Can you even be precise about how many registers a language has? I would have thought whether a language has, say, three or four registers is a matter of interpretation.

This is especially true of a language like English which does not really have grammaticalized registers.

Well, of course it's a matter of interpretation. Joos interpreted the English data as falling into five categories based on a set of criteria. Different criteria would probably produce two registers, or a hundred. It's like asking how many colors there are. Just because it's a spectrum doesn't mean we can't come up with an answer that makes sense within a carefully described context.

For Joos, he based his categories on things like public information (which distinguishes intimate from casual), level of spontaneity (which distinguishes frozen from formal), and many others. But his work is from sixty years ago. I just mentioned him because I stumbled across his book recently. No doubt more recent linguists have come up with more sophisticated criteria.

Travis B. wrote: ↑Sun May 19, 2024 12:46 pm

Raphael wrote: ↑Sun May 19, 2024 12:10 pm Can you even be precise about how many registers a language has? I would have thought whether a language has, say, three or four registers is a matter of interpretation.

This is especially true of a language like English which does not really have grammaticalized registers.

'snot true. Elipsis of function words is very carefully regulated by rules that vary by register.

Glass Half Baked wrote: ↑Sun May 19, 2024 8:22 pm

Travis B. wrote: ↑Sun May 19, 2024 12:46 pm

Raphael wrote: ↑Sun May 19, 2024 12:10 pm Can you even be precise about how many registers a language has? I would have thought whether a language has, say, three or four registers is a matter of interpretation.

This is especially true of a language like English which does not really have grammaticalized registers.

'snot true. Elipsis of function words is very carefully regulated by rules that vary by register.

I meant in contrast to, say, Japanese, which has highly grammaticalized registers. However, it should also be noted that it is hard to interpret English registers as being discrete, countable registers, as opposed to a continuum of forms from the most basilectal to the most acrolectal with no sharp delineations in between.

What do you think are some crosslinguistically common or even "essential" conjunctions.

I want to come up with a small but versatile set of conjunctive proclitics to employ at the clause-level

So far, I can only think of: "and", "but/yet", "or", "so", "thus/therefore", "if", "then", "because", and maybe "nor". But beyond that I'm not sure what else.

Ahzoh wrote: ↑Sat May 25, 2024 6:10 pm What do you think are some crosslinguistically common or even "essential" conjunctions.

None: I don’t see why a language should necessarily need any conjunctions at all. I haven’t really investigated the subject at all, but I would be unsurprised to discover a language which has no dedicated conjunctions.

All languages have some way of expressing basical logical operations like "and" and "or". But I wouldn't be surprised if there were some languages without a separate word class of conjunctions. I remember hearing about some (Papuan?) language where conjunctions were just a subclass of verbs. Or they could be entirely morphologically expressed which would probably fit in a polylang.

In a somewhat similar—though not identical—way to how I unified CT and O into a single language family (Beheic), I’ve gotten¹ an idea of how to merge two of my other languages (to wit, Kgáweq’ and Täptäg, both now part of the Macro-Kgáweq’ family). I’m quite pleased.

1. The recent discussion in the Linguistic Miscellany Thread is very much brought to mind.

Darren wrote: ↑Sat May 25, 2024 7:14 pm All languages have some way of expressing basical logical operations like "and" and "or".

In desperation, perhaps. But the concepts are not usually expressed in Middle Egyptian. And in some languages a preposition meaning 'with' can take the place of a conjunction meaning 'and'.

Richard W wrote: ↑Sat May 25, 2024 8:25 pm

Darren wrote: ↑Sat May 25, 2024 7:14 pm All languages have some way of expressing basical logical operations like "and" and "or".

But the concepts are not usually expressed in Middle Egyptian.

Could this not be an accident of attestation? I find it rather difficult to believe that people wouldn't have needed to talk about multiple things going on at once.

Are there any natlangs that, like my new conlang fad'ami, have a complete lack of adpositions, expressing equivalent ideas with applicatives, relational nouns, genitive case, and relative clauses? Common motifs in fad'ami include using an applicative on a verb with a relational noun belonging to a noun in genitive case, and in many cases this verb is an existential copula in a relative clause. How common are these kinds of strategies?

Travis B. wrote: ↑Sat May 25, 2024 9:11 pm Are there any natlangs that, like my new conlang fad'ami, have a complete lack of adpositions, expressing equivalent ideas with applicatives, relational nouns, genitive case, and relative clauses? Common motifs in fad'ami include using an applicative on a verb with a relational noun belonging to a noun in genitive case, and in many cases this verb is an existential copula in a relative clause. How common are these kinds of strategies?

As far as I can tell, Dyirbal has neither conjunctions nor adpositions. I can't find any mention of them in Dixon's grammar.

bradrn wrote: ↑Sat May 25, 2024 6:34 pm

Ahzoh wrote: ↑Sat May 25, 2024 6:10 pm What do you think are some crosslinguistically common or even "essential" conjunctions.

None: I don’t see why a language should necessarily need any conjunctions at all. I haven’t really investigated the subject at all, but I would be unsurprised to discover a language which has no dedicated conjunctions.

Sure, but this doesn't help with my question. That's why I first said "cross-linguistically common". And there ain't no way that "complete lack of conjunctions/words with meanings often covered by conjunctions" is crosslinguistically common either.

The goal is to avoid recreating ALL the English conjunction words and their specific distinctions (e.g. but vs yet)

They don't have to be a formal word class, just express the concepts that conjunctions typically do. Cuz I'm quite aware some use adpostions or special verb forms to serve conjunctive functions . Like "with" also being used to mean "and" or the Ethiopian Semitic "gerundives" which also serve to link clauses in a manner akin to "and"

Ahzoh wrote: ↑Sat May 25, 2024 10:06 pmThe goal is to avoid recreating ALL the English conjunction words and their specific distinctions (e.g. but vs yet)

Maybe emulate Yiddish, with only one conjunction word, which I think was ne....

I am smart ne Einstein was too.

I am smart ne you are smarter.

I am smart ne rocks are heavy.

etc.

Rounin Ryuuji wrote: ↑Sat May 25, 2024 9:10 pm

Richard W wrote: ↑Sat May 25, 2024 8:25 pm

Darren wrote: ↑Sat May 25, 2024 7:14 pm All languages have some way of expressing basical logical operations like "and" and "or".

But the concepts are not usually expressed in Middle Egyptian.

Could this not be an accident of attestation? I find it rather difficult to believe that people wouldn't have needed to talk about multiple things going on at once.

Sure… but you don’t need a dedicated conjunctional word to do that! Like Richard W says, it’s quite common to use ‘with’ in the sense of ‘and’. And plain juxtaposition is always a popular strategy for all kinds of things, even if English doesn’t use it.

(And of course there are more complicated strategies, like clause-chaining systems, which are quite elaborate while not needing conjunctions.)

Travis B. wrote: ↑Sat May 25, 2024 9:11 pm Are there any natlangs that, like my new conlang fad'ami, have a complete lack of adpositions

Yes, quite a lot of them. Many serialising languages tend to do this: serial verb constructions replace adpositions perfectly well. (I seem to recall this being the case for some Chinese languages with their ‘coverbs’, which are more or less prepositional.) Yimas strongly prefers an applicative strategy, like your conlang — I don’t recall if it might have one or two adpositions, but I think it may have none. In fact, there’s even a subsection in Language Typology and Syntactic Description entitled ‘Languages without adpositions’ (although it doesn’t say much).

Ahzoh wrote: ↑Sat May 25, 2024 10:06 pm They don't have to be a formal word class, just express the concepts that conjunctions typically do. Cuz I'm quite aware some use adpostions or special verb forms to serve conjunctive functions . Like "with" also being used to mean "and" or the Ethiopian Semitic "gerundives" which also serve to link clauses in a manner akin to "and"

OK, thanks for clarifying. Then I’d say that at least ‘and’ is probably universal. I wouldn’t be so bold to claim that any others are. (I seem to recall reading somewhere that many language get by just fine without ‘or’, for example.)

keenir wrote: ↑Sun May 26, 2024 2:12 am Maybe emulate Yiddish, with only one conjunction word, which I think was ne....

This is just rubbish. Wiktionary, for one thing, lists no less than 39 Yiddish conjunctions. Yiddish is not that grammatically distant from any of the various other German languages!

bradrn wrote: ↑Sun May 26, 2024 4:30 am

keenir wrote: ↑Sun May 26, 2024 2:12 am Maybe emulate Yiddish, with only one conjunction word, which I think was ne....

This is just rubbish. Wiktionary, for one thing, lists no less than 39 Yiddish conjunctions. Yiddish is not that grammatically distant from any of the various other German languages!

Well, the principle of one conjunction serving for multiple functions, could still help Ahzoh. I remembered the multiple uses of (ne) from one of the books I used to own about Yiddish, and I didn't delve into seeing what other conjunctions it has...tho I likely would have encountered them had I tried to learn Yiddish.

keenir wrote: ↑Sun May 26, 2024 5:02 am

bradrn wrote: ↑Sun May 26, 2024 4:30 am

keenir wrote: ↑Sun May 26, 2024 2:12 am Maybe emulate Yiddish, with only one conjunction word, which I think was ne....

This is just rubbish. Wiktionary, for one thing, lists no less than 39 Yiddish conjunctions. Yiddish is not that grammatically distant from any of the various other German languages!

Well, the principle of one conjunction serving for multiple functions, could still help Ahzoh. I remembered the multiple uses of (ne) from one of the books I used to own about Yiddish, and I didn't delve into seeing what other conjunctions it has...tho I likely would have encountered them had I tried to learn Yiddish.

Was hoping to have more than just one, which is also kind of boring

I'm more thinking in terms of functions, such as illatives like "presents rationale or purpose" (for, in order to) or "presents cause" (because) or "presents consequence" (so, thus, therefore), or adversatives like "presents contrast" (but, yet).

Like it make sense to me to not make a distinction between "but" and "yet" which differ in emphasis/degree of contrastiveness, and it also make sense that "so" and "thus" need not be distinct either.
But I know of one language where the word for "and" also means "but", which is strange and I don't know how it resolves the ambiguity, since it is essentially autoantonymous.

bradrn wrote: ↑Sun May 26, 2024 4:25 am Sure… but you don’t need a dedicated conjunctional word to do that! Like Richard W says, it’s quite common to use ‘with’ in the sense of ‘and’. And plain juxtaposition is always a popular strategy for all kinds of things, even if English doesn’t use it.

And Middle Egyptian usually uses plain juxtaposition where English uses 'and' or 'or'.


(And of course there are more complicated strategies, like clause-chaining systems, which are quite elaborate while not needing conjunctions.)

Travis B. wrote: ↑Sat May 25, 2024 9:11 pm Are there any natlangs that, like my new conlang fad'ami, have a complete lack of adpositions

bradrn wrote: ↑Sun May 26, 2024 4:25 am Yes, quite a lot of them. Many serialising languages tend to do this: serial verb constructions replace adpositions perfectly well.

On the other hand, Thai does seem to have a few prepositions, though most words translating English prepositions are verbs or nouns.