Another Almost-Cartesian Consonant Inventory

A forum for all topics related to constructed languages
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Another Almost-Cartesian Consonant Inventory

Post by Creyeditor »

Oh, I see. I understood the parts of Seins-art (no relation to "sort", IINM), since German is my native language. I just never had heard the term applied to language. I had mostly heard it in a sense close to "mode of living" especially when applied to national stereotypes of locals in touristic places. Duden Online uses it as a classification of "Realität" and "Idealität". One could translate the Duden usage to linguistics by talking about realis and irrealis nouns. Biak is a natlang that marks this distinction on its determiners.
BUT, I can totally see why one could use it in parallel to Aktionsart for gender, mass/count, animacy and similar stuff. I would have probably still called it either gender or noun class, since I did not know about the term Seinsart in linguistics.
Intuitively, I can see why you want the first to consonants to express Art (the head noun of both Seinsart and Aktionsart) on the first two letters. On the other hand, does that make words more easily confusable. If, let's say both "rice" and "wheat" (inanimate mass nouns) start with the similar letters and have similar meanings, does that make them more easily confusable. IIRC, someone once criticized something similar about Wilkins' Real Character.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

There’s no way it can be perfect. There are only eight PoAs and I think there are eight +- a couple different sub variables of seinsart. And some of the inflections of a nominal root will promote the second PoA into the first position.

Mutation of a word’s initial consonant encodes syntactic information about the noun into the MoA of its first consonant.
I was thinking maybe the initial PoA could encode non-grammatical, semantic information, about the noun’s referent; information that would be lexically inherent in the root, and not change with inflection. It should, however, be relevant to how likely the noun is to occupy certain morphosyntactically-assigned argument positions. That’s why animacy and concreteness and ability to sense the environment and to speak would be parts of seinsart. I’ve changed my mind about mass-vs-count and common-vs-proper. Or, rather, I’m considering changing my mind.

In the classical languages having a final consonant of a verb be a liquid MoA told certain syntactic and/or morphological information about those so-called liquid verbs.
I thought the final consonant’s PoA could help hint at lexically-inherent semantic information about the verb, that would make the semantic interpretation of certain morphological markings more clear. The one that jumped to my mind was aktionsart.

Sorry about the sart vs art error. German isn’t my L1. Its not even my L2, assuming I have one.

Thanks for reading this and thinking about it!

.....

Of course at such an early date I don’t have any roots I could assign meaning to, nor any inflections I could assign meaning to. I think I may need fewer than five vowels; perhaps just three.

I’ll return to answering your other questions that are pending, when I can find the time and brainpower to do so.
Last edited by eldin raigmore on 11 Aug 2021 02:21, edited 1 time in total.
User avatar
WeepingElf
greek
greek
Posts: 531
Joined: 23 Feb 2016 18:42
Location: Braunschweig, Germany
Contact:

Re: Another Almost-Cartesian Consonant Inventory

Post by WeepingElf »

eldin raigmore wrote: 11 Jun 2020 12:22 “Seinsart” is German for “mode of being”.
“Sart ” is cognate with “sort” and means something like “type”.
Correction (from a native German speaker): Sart is not a German word. The second member of Seinsart is Art 'type, species'. But your translation of Seinsart is correct, even if it is a very uncommon word.
... brought to you by the Weeping Elf
My conlang pages
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

I don’t think seinsart and gender are the same thing.

I think they may overlap, but neither contains the other.

Gender is concordial noun-class. It may or may not be marked on the noun, but some other word has to be marked to agree with it. Otherwise, while it could be noun-class, it couldn’t be gender. Gender is usually considered lexically-inherent, and for some languages some genders are largely semantic, but as a general rule some nouns are assigned to certain genders for non-semantic reasons.

Seinsart otoh is some lexically-inherent semantic feature of the noun’s referent, that influences what roles it is likely to play in clauses or verb-phrases (or possibly other phrases? maybe not). Perhaps it is to case what aktionsart is to aspect. (Maybe I could call it “lexically-inherent casual class”. Or not.)

Adjectives etc. rarely have to agree with most of the features that make up seinsart.
Seinsart more closely resembles case, in the way we might say that certain adpositions govern certain cases, rather than agree with them; or that certain verbs demand subjects or objects in certain cases (the verbs don’t have to “agree” with the quirkily-cased subjects nor the differentially-marked objects). I think.
But since it is not always case, adjectives don’t always have to agree with it.

If a language has animacy-based gender, for that language animacy would be both part of seinsart and part of gender.
Animate referents are likelier to be agents.

If a language has sentience-based gender, for that language sentience would be part of both seinsart and of gender.
Sentient referents are likelier to be the experiencers in sense-verb clauses.

Likewise, the ability to form mental judgements, or experience emotions, influences a noun’s probability of being the evaluator or emoter in clauses of judgement or emotion.

Solid, tangible, visible, concrete objects, that take up a definite amount of space for a non-trivial time, are likelier to be the subjects or objects of locational predications. And so on.

Referents who can speak and understand language are likelier to be tellers and showers and hearers and addressees and viewers in clauses of telling or showing or speaking or listening etc.

I think these are likelier to be parts of seinsart in most languages. But most wont be part of gender even in most languages that have gender.

For one reason, it’s rare that any other word would have to agree with such a nominal feature.
Though I suppose for most of those seinsart features, there is some natlang that has a corresponding gender!

That’s what I think.
That’s how I want to use the term seinsart in linguistics, as opposed to in philosophy.
I think between the 1970s and the 1990s , the linguistic use of the term “seinsart “ was trending to be used that way; lexically-inherent semantic features of a noun’s referent that influenced the likelihood that it would occupy certain core argument roles in clauses; for instance, agent, patient, recipient; maybe also instigator, causee, beneficiary.

__________

(However I would be only mildly surprised to discover I’ve been wrong!)
Last edited by eldin raigmore on 11 Aug 2021 02:23, edited 1 time in total.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

Creyeditor wrote: 10 Jun 2020 15:30 .... Actually, in order to be easier to understand, could you give a concrete example of two possible roots with their meaning and two possible inflections? No need to commit yourself, just a concrete example. ....
Except for assigning meaning, I might be able to do something like that.
Consider the four MoAs {voiceless fricative, voiced fricative, nasal, voiceless stop}
and the four PoAs {alveolar, retroflex, palatal, velar}
Consider a 4-consonant root
whose first two consonants are a voiceless fricative and a voiced fricative,
and an alveolar and a retroflex;
and whose last two consonants are a nasal and a voiceless plosive,
and a palatal and a velar.

Its first two consonants might be
/s-ʐ-/ or /ʐ-s-/ or /z-ʂ-/ or /ʂ-z-/.
Its last two consonants might be
/-c-ŋ/ or /-ŋ-c/ or /-k-ɲ/ or /-ɲ-k/.

These sixteen different sequences of four consonants, would be different inflected or derived forms of the same root lexeme.
User avatar
WeepingElf
greek
greek
Posts: 531
Joined: 23 Feb 2016 18:42
Location: Braunschweig, Germany
Contact:

Re: Another Almost-Cartesian Consonant Inventory

Post by WeepingElf »

eldin raigmore wrote: 13 Jun 2020 18:04 I don’t think seinsart and gender are the same thing.
They aren't. In fact, Seinsart is not a commonly used German word, rather something to find in the writings of German-language philosophers. The German word for 'gender' is Geschlecht.
... brought to you by the Weeping Elf
My conlang pages
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Another Almost-Cartesian Consonant Inventory

Post by Creyeditor »

eldin raigmore wrote: 13 Jun 2020 18:04 I don’t think seinsart and gender are the same thing.

I think they may overlap, but neither contains the other.

Gender is concordial noun-class.

[...]
Seinsart otoh is some lexically-inherent semantic feature of the noun’s referent, that influences what roles it is likely to play in clauses or verb-phrases (or possibly other phrases? maybe not).

[...]

I think between the 1970s and the 1990s , the linguistic use of the term “seinsart “ was trending to be used that way; lexically-inherent semantic features of a noun’s referent that influenced the likelihood that it would occupy certain core argument roles in clauses; for instance, agent, patient, recipient; maybe also instigator, causee, beneficiary.

__________

(However I would be only mildly surprised to discover I’ve been wrong!)
Ah, I see that now. So they are potentially the same features semantically, but they influence the morphosyntactic grammar in different ways. Whereas gender is relevant for agreement and concord, seinsart is relevant for argument roles.

So, if we have animacy-based differential object marking what would that be? Suppose there is no agreement/concord for animacy and animate direct objects are marked with an overt accusative case suffix, but inanimate direct objects are unmarked for case. Using your terminology, would that be gender, seinsart or something else?


eldin raigmore wrote: 13 Jun 2020 18:47
Creyeditor wrote: 10 Jun 2020 15:30 .... Actually, in order to be easier to understand, could you give a concrete example of two possible roots with their meaning and two possible inflections? No need to commit yourself, just a concrete example. ....
Except for assigning meaning, I might be able to do something like that.
Consider the four MoAs {voiceless fricative, voiced fricative, nasal, voiceless stop}
and the four PoAs {alveolar, retroflex, palatal, velar}
Consider a 4-consonant root
whose first two consonants are a voiceless fricative and a voiced fricative,
and an alveolar and a retroflex;
and whose last two consonants are a nasal and a voiceless plosive,
and a palatal and a velar.

Its first two consonants might be
/s-ʐ-/ or /ʐ-s-/ or /z-ʂ-/ or /ʂ-z-/.
Its last two consonants might be
/-c-ŋ/ or /-ŋ-c/ or /-k-ɲ/ or /-ɲ-k/.

These sixteen different sequences of four consonants, would be different inflected or derived forms of the same root lexeme.

So, to be even more conrete: /saʐacaŋ/ and /ʂazaɲak/ can be two different inflected forms of the same word. I think this is already an astonishing result. (I was just guessing vowels.)

I noticed that this is similar to an idea that I came up with earlier, where the idea of floating tones in autosegmental phonology is expanded to (almost) all phonological features and to be the only way to express grammatical features. And a related engelang idea, where lexical information is expressed by prosodic phonological features and grammatical information by segmental features. For both of of them, I have to admit, the hardest part is coming up with concrete examples. So, I really respect the fact, that you are able to do that at this point.




WeepingElf wrote: 13 Jun 2020 19:20
eldin raigmore wrote: 13 Jun 2020 18:04 I don’t think seinsart and gender are the same thing.
They aren't. In fact, Seinsart is not a commonly used German word, rather something to find in the writings of German-language philosophers. The German word for 'gender' is Geschlecht.
I think eldin is not talking about the German (philosophical) term "Seinsart", but about an English term "seinsart" used in some linguistic research traditions.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

So, if we have animacy-based differential object marking what would that be? Suppose there is no agreement/concord for animacy and animate direct objects are marked with an overt accusative case suffix, but inanimate direct objects are unmarked for case. Using your terminology, would that be gender, seinsart or something else?
The animacy would be part of the seinsart if I have my way.
The differential object marking would be an effect seinsart has in this particular language.
Does that make sense?

Edit: It’s a bit like the difference between diathesis and voice. Diathesis is semantic; voice is the morphological marking of diathesis. At least in some linguisticians’ terminology!

—————
I noticed that this is similar to an idea that I came up with earlier, where the idea of floating tones in autosegmental phonology is expanded to (almost) all phonological features and to be the only way to express grammatical features. And a related engelang idea, where lexical information is expressed by prosodic phonological features and grammatical information by segmental features.
That sounds fascinating! Care to say more?

Edit: BTW as for the rest of your post; yes, you understood me!
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

@WeepingElf:
Your comments are also appreciated!
I haven’t replied to them: but that reflects the fragmentation of my attention; not any idea that they weren’t helpful!
Thanks!
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Another Almost-Cartesian Consonant Inventory

Post by Creyeditor »

eldin raigmore wrote: 14 Jun 2020 13:43
So, if we have animacy-based differential object marking what would that be? Suppose there is no agreement/concord for animacy and animate direct objects are marked with an overt accusative case suffix, but inanimate direct objects are unmarked for case. Using your terminology, would that be gender, seinsart or something else?
The animacy would be part of the seinsart if I have my way.
The differential object marking would be an effect seinsart has in this particular language.
Does that make sense?
It does and it really helped me understand what people use the term seinsart for.

eldin raigmore wrote: 14 Jun 2020 13:43
I noticed that this is similar to an idea that I came up with earlier, where the idea of floating tones in autosegmental phonology is expanded to (almost) all phonological features and to be the only way to express grammatical features. And a related engelang idea, where lexical information is expressed by prosodic phonological features and grammatical information by segmental features.
That sounds fascinating! Care to say more?
I started a new thread, so I don't derail the disussion here. I had prepared this post for a few months anyway, so not much work involved. I would love to hear your comments or questions.
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

It does and it really helped me understand what people use the term seinsart for.
That is, what I think linguisticians writing in English use it for, and/or what I want us to use it for.
Like I said, I’m not sure I’m right; but I could be and I hope I am!

Glad it helped!
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

Creyeditor wrote: 14 Jun 2020 14:48
I’m going to continue to use or abuse* the term “seinsart”, at least in this thread, because I haven’t thought of or run across a better (in my opinion) term.
I thought of “lexically-inherent casual class”; but even if that’s a well-formed term, I think it’s kinda cumbersome.
*(I may be the first one to use it this way! I don’t think so, but —— maybe!) ..... ..... ..... ..... .....
What I’ve been thinking is the following.
One-and-the-same semantic fact about a noun’s referent, can be part both of gender and of seinsart, in the same language.
(E.g. Creyeditor showed that animacy could be.)
.... .... .... ....
Clearly, if it’s part of gender, that’s language-specific.
But I think, if it’s part of seinsart, that’s probably cross-linguistic.
Last edited by eldin raigmore on 19 Feb 2021 14:30, edited 1 time in total.
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Another Almost-Cartesian Consonant Inventory

Post by Creyeditor »

I just remembered that I once heard "sortal properties" for the semantic basis of grammatical "gender", but I doubt that this is common use either. Googling gives a lot of alternative usages for "sortal properties of nouns".
Creyeditor
"Thoughts are free."
Produce, Analyze, Manipulate
1 :deu: 2 :eng: 3 :idn: 4 :fra: 4 :esp:
:con: Ook & Omlűt & Nautli languages & Sperenjas
[<3] Papuan languages, Morphophonology, Lexical Semantics [<3]
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

Creyeditor wrote: 02 Sep 2020 21:42
Pretty hard to understand in one glance.
Thanks!
Wonder if what I was talking about are “essential” properties or not;
also whether they are “sortal” or not.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Another Almost-Cartesian Consonant Inventory

Post by eldin raigmore »

In this 38-consonant inventory:
7 nasals: /m ɱ n ɳ ɲ ŋ ɴ/
7 voiced fricatives: /β v z ʐ ʝ ɣ ʁ/
7 voiceless fricatives: /ɸ f s ʂ ç x x̩/
6 voiced stops/plosives: /b d ɖ ɟ g ɢ/
6 voiceless stops/plosives: /p t ʈ c k q/
5 approximants: /ʋ ɹ ɻ j ɰ /

:::::

6 alveolars: /t d n s z ɹ /
6 retroflexes: /ʈ ɖ ɳ ʂ ʐ ɻ/
6 palatals: /c ɟ ɲ ç ʝ j/
6 velars: /k g ŋ x ɣ ɰ /
5 uvulars: /q ɢ ɴ x̩ ʁ/
5 bilabials: /p b m ɸ β/
4 labiodentals: /ɱ f v ʋ/

If there are 7*6/2 = 21 unordered pairs of places-of-articulation and 6*5/2 = 15 unordered pairs of manners-of-articulation (counting voicing as part of manner), there are 21*15 = 315 pairs of a pair of PoAs and a pair of MoAs.
If we were lucky enough (but we aren’t!) that each pair of PoAs could go with each pair of MoAs, and any pair for the first two consonants of a 4-consonant root could go with any pair for the last two consonants of a 4-consonant root, we could get 315*315 = 441*225 = 99,225 pairs-of-pairs-of-pairs. That’s less than the OED’s 177,000 but more than Kankonian’s current record.

….. ….. ….. ….. …..

If the PoAs are drawn from the four that intersect all 6 of the MoAs we could get (4*3)/2 = 6 pairs of PoAs; giving 6*15=90 pairs of a pair of PoAs and a pair of MoAs that are guaranteed to have all four intersections filled in.
If the MoAs are drawn from the three that intersect all 7 of the PoAs we could get 21*3=63 pairs of a pair of PoAs and a pair of MoAs that are guaranteed to have all four intersections filled in.
18 of those were counted twice. So that’s a minimum of 90+63-18 = 135 pairs of pairs that are guaranteed “locally-Cartesian”.
If I am not mistaken any locally-Cartesian pair for the first two consonants is compatible with any locally-Cartesian pair for the last two consonants.
So there could be at least 135*135 = 18,225 roots. That’s a lot but I’d like more.

…. …. …. ….

If I were willing to include consonants that aren’t on the IPA chart, I might use 8 PoAs and 6 MoAs, or 7 PoAs and 7 MoAs, and get either 28*15 = 420 or 21*21 = 441 pairs-of-pairs for the first two or last two consonants, and have them be locally Cartesian and compatible with each other; resulting in 420*420 = 176,400 or 441*441 = 194,481 roots; rivaling the OED.

………………..

If four different PoAs are used, or four different MoAs are used, then the pair of pairs for the first two consonants is compatible with the pair of pairs for the last two consonants.
Post Reply