Parts-of-Speech of Root Morphemes

A forum for all topics related to constructed languages
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Parts-of-Speech of Root Morphemes

Post by eldin raigmore »

How are your conlangs' root-morphemes or word-roots distributed among various word-classes?
According to the OED,
a bit more than half of English's almost-180,000 word-roots are noun-roots;
about a third are adjective-roots;
about a seventh are verb-roots;
and the remaining forty-second are other kinds of roots.

If your conlang is very different from that, how and why?
I can imagine several reasons:
(1) the question isn't relevant to your conlang because the distinctions between parts-of-speech don't apply at the word-root level; it applies only at the finite-word level.
(2) there's no difference between adjectives and nouns in your conlang, so that part of the question is either irrelevant or trivial.
(3) there's no difference between adjectives and verbs in your conlang, so that part of the question is either irrelevant or trivial.
(4) your conlang has very few lexical verbs (e.g. twelve or fewer?), and almost all (e.g. two-thirds or more) of its verbs are phrases consisting of a light-verb plus a content-word.
(5) it's easy for me to believe there are other obvious possibilities that just aren't all that obvious to me, at the moment.
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Parts-of-Speech of Root Morphemes

Post by eldin raigmore »

Creyeditor wrote:For Kobardon the answer is (1). Roots are category neutral and inflection/derivation makes them become verbs (transitive and intransitive), nouns, adjectives and adverbs.
Omlueuet is similar to English I guess. Most conlangs of mine do not yet have enough root to make any statistically significant statements about the categories of roots.
Thank you, Creyeditor.
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

No part of speech here, only noun roots...
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Parts-of-Speech of Root Morphemes

Post by Creyeditor »

Do they every become other oart of speech? Is there POS-changing derivational morphology? Or is it just a 'noun-only' conlang?
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
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

one can describe it as a 'noun-only' conlang...
but without opposition noun is something else...
it's a no part of speech conlang...
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Parts-of-Speech of Root Morphemes

Post by eldin raigmore »

lsd wrote:one can describe it as a 'noun-only' conlang...
but without opposition noun is something else...
it's a no part of speech conlang...
Every natural language distinguishes at least between nouns and verbs at the level of finite words. But not every natural language distinguishes them at the root level. Also, in some natural languages, the difference between a noun and a verb is nowhere near as pronounced as it is in the languages we are more familiar with. Some natural languages do not have any parts of speech other than nouns and verbs. Many do.

I think that at least some Tri-Consonantal Verb-Root conlangs, every word except a few particles maybe will be derived from a verb root.

But having a language with no parts of speech, or only one part of speech, is unnaturalistic and unrealistic. Of course, if realism and naturalism were not among the conlang's designer's goals, there's really nothing wrong with it being unrealistic or unnaturalistic.
Last edited by eldin raigmore on 20 Sep 2017 23:50, edited 1 time in total.
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

For a priori (in philosophical way) language (also called taxonomic language) parts of speech are out of the target.. It seems natural to avoid them(IMHO)...
Last edited by lsd on 23 Sep 2017 13:50, edited 1 time in total.
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

I am always amazed by the conlangs made for use (auxlangs) that have roots from different parts of speech they use differently to produce the other parts of speech, whereas regularity is their only advantage...
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Parts-of-Speech of Root Morphemes

Post by eldin raigmore »

lsd wrote:For a priori (in philosophical way) language (also called taxonomic language) parts of speech are out of the target.. Avoid them is natural (IMHO)...
I don't understand the "Avoid them is natural (IMHO)" part of your comment.

Taxonomic philosophical languages would indeed seem to concentrate on nouns.
For a constructor of such a language, it might indeed be "natural" to neglect other parts-of-speech; in particular, verbs.
That may be one reason such conlangs are seldom successful; though there are more severe problems that crop up earlier.

B.A.S.I.C. English was designed to be "without verbs"; instead it has nouns that name actions and states and events and so on.
So an auxlang can be pretty successful even if "it has no verbs" in some sense.

IMNSHO no usable language -- natural or auxiliary or philosophical or taxonomic or logical or whatever -- can fail to (in some sense) "have verbs" in some of its clauses.

Maybe it can have many, and many types of, verb-less clauses.

Maybe a word can't be classified as a particular part-of-speech except when used in a clause.
(English's "green", for instance, can be an adjective, or a noun, or a transitive verb; one must see or hear the clause it's used in to tell which PoS it is at that particular use.)

Maybe none of the roots are verb-roots, and/or none of the roots have part-of-speech--ness until inflected and/or derived into finite words that can be used in clauses.

Maybe it can have very few verbs -- only three, perhaps (e.g. I forget which, but some natlang has only "come" and "go" and "say").

But at least some of the language's clauses must require a verb as their nucleus.

lsd wrote:I am always amazed by the conlangs made for use (auxlangs) that have roots from different parts of speech they use differently to produce the other parts of speech, whereas regularity is their only advantage...
I don't understand why deriving a word that's one part-of-speech from a word that's another constitutes irregularity.
Infinitives, gerunds, and masdars, are verbal nouns.
In the case of masdars, it appears that finite verbs are derived from the masdar, usually. (Or at least sometimes, in case I'm wrong about the "usually".)

Participles are verbal adjectives.

Supines are verbal adverbs.

One can have agent-nominalization (verb--> noun for thing that does or did or will do, or might have done or might do, that verb),
patient-nominalization (verb --> noun for thing that verb was done to),
instrument-nominalization (verb --> noun for instrument to make doing that verb easier),
event-nominalization (I think I needn't explain that),
time-nominalization,
place-nominalization,
etc.

Adjectives can be "turned into" nouns;
in particular "an open" can mean "an open thing", or "a black" can mean "a black horse", etc.
Also, "open-->openness", "black-->blackness", etc.

And nouns can be turned into adjectives; child-->childish, e.g.

Adjectives can also be "turned into" verbs.
And adjectives can be turned into adverbs; e.g. angry-->angrily.

The irregularity doesn't come from deriving one part-of-speech from another.

It comes from one or more of:

1) having more than one way to derive a word that has the same meaning; "childlike" and "childish", or "manly" and "mannish", e.g.

2) assigning more than one meaning to a given word.

3) having such a wide variety of choice that one can, for instance, derive a verb from a noun, then derive a noun from that verb that means the same thing as the original noun;
especially if one could also derive a noun from a verb, then derive a verb from that noun that means the same thing as the original verb.

Situations 1) and 2) pretty much are the definition(s) of irregularity.
3) is pretty much an egregiously complexifying way of creating situation 1).

IM(NSH)O, if the same morphological operation produces the same kinds of changes in semantics every time, then everything is very regular.
In such a language, the only thing differentiating derivation from inflection, would be that derivation changes part-of-speech;
it is transparent, universally productive, and totally regular.

One might pick a particular word-class -- e.g. nouns, or verbs -- and declare that all non-particle words (all inflectible words) of other parts-of-speech must be derived from words of the picked word-class.
That might be an additional improvement in regularity.

I don't know that that would necessarily be possible, however.
Suppose your auxlang's adpositions are not particles; suppose they take inflection.
Can you always either derive all adpositions from verbs, or derive all adpositions from nouns, or derive all adpositions from modifiers (adjectives or adverbs)?
I don't know the answer.

But I'd like to know the answer, and why it is the answer.
Iyionaku
mayan
mayan
Posts: 2102
Joined: 25 May 2014 14:17

Re: Parts-of-Speech of Root Morphemes

Post by Iyionaku »

I have never counted my roots in Yélian, but I assume that I have almost as many verbal roots than noun roots. That may have been due to nonattention. Caelian roots can always be configured as nouns, verbs, adjectives or adverbs.
Ular has a closed class of verb roots; until now there are only 92 verb roots. Nouns are way more frequent.
lsd wrote:No part of speech here, only noun roots...
lsd, please take no offense of that, but is there ANY particular reason why you always finish sentences with three periods instead of one? It - drives - me - crazy.
Wipe the glass. This is the usual way to start, even in the days, day and night, only a happy one.
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

eldin raigmore wrote:ITaxonomic philosophical languages would indeed seem to concentrate on nouns. For a constructor of such a language, it might indeed be "natural" to neglect other parts-of-speech; in particular, verbs.
You undestood my broken english..
But at least some of the language's clauses must require a verb as their nucleus.
That's the result, no nucleus... no difference between words and sentences... of course it's an experimental language...
I don't understand why deriving a word that's one part-of-speech from a word that's another constitutes irregularity.
Using only one kind of root gives the big advantage in word construction to have one single construction by part of speech... no necessity of knowing the part of speech the root belong to choose the derivating way...
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

Iyionaku wrote:lsd, please take no offense of that, but is there ANY particular reason why you always finish sentences with three periods instead of one? It - drives - me - crazy.
No offense, it's probably a conlinguistic deformation... Some Sapir Whorf effect...
User avatar
qwed117
mongolian
mongolian
Posts: 4094
Joined: 20 Nov 2014 02:27

Re: Parts-of-Speech of Root Morphemes

Post by qwed117 »

lsd wrote:
Iyionaku wrote:lsd, please take no offense of that, but is there ANY particular reason why you always finish sentences with three periods instead of one? It - drives - me - crazy.
No offense, it's probably a conlinguistic deformation... Some Sapir Whorf effect...
:roll:
Spoiler:
My minicity is [http://zyphrazia.myminicity.com/xml]Zyphrazia and [http://novland.myminicity.com/xml]Novland.

Minicity has fallen :(
The SqwedgePad
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

qwed117 wrote: :roll:
...
User avatar
eldin raigmore
korean
korean
Posts: 6352
Joined: 14 Aug 2010 19:38
Location: SouthEast Michigan

Re: Parts-of-Speech of Root Morphemes

Post by eldin raigmore »

Iyionaku wrote:lsd, please take no offense of that, but is there ANY particular reason why you always finish sentences with three periods instead of one? It - drives - me - crazy.
You may have noticed that my ellipses (plural of "ellipsis" --- "elisions", maybe?) is usually four dots .... instead of three ... I started doing it that way occasionally whenever a period was the last thing before, or the first thing after, what I intended to elide. Then I found out "...." is just as proper as (though not as common as) "..." to indicate elision in general. So ....

Back on topic.
Iyionaku wrote:I have never counted my roots in Yélian, but I assume that I have almost as many verbal roots than noun roots. That may have been due to nonattention. Caelian roots can always be configured as nouns, verbs, adjectives or adverbs.
Ular has a closed class of verb roots; until now there are only 92 verb roots. Nouns are way more frequent.
Thanks for those answers, @Iyionaku!


lsd wrote:That's the result, no nucleus... no difference between words and sentences... of course it's an experimental language...
Sounds like you're trying to create what conlangers (at least here on the CBB) call an "oligosynthetic" conlang.
(Not that it's not also one or more of : "experimental", "taxonomic", "philosophical", and/or "logical"!)

IRL natlangs that have "no difference between verbs and sentences" are usually called "polysynthetic languages" (though AFAIK none really take it to the extreme that they don't have any multi-word sentences!).
These are usually analyzed as having words which are verbs, whose roots are usually verb-roots, that incorporate as inflections all the morphemes that stand for nouns and adjectives and adverbs and adpositions and pronouns and so on. Sometimes the nouns etc. (or at least some of them!) can't stand by themselves as independent words! E.g. there's some such natlang where "apple" glosses as "-k-".

Oligosynthetic conlangs differ from polysynthetic (nat- or con-)langs in that the oligosynthetic conlangs are meant to have a restricted, or small or smallish, set of fundamental morphemes. They may or may not qualify as "polysynthetic", and may or may not even try to; but it's my impression that many or even most do include a lot of "polysynthetic" features (for instance, morphology almost absorbs syntax, so that most clauses are few words and many are just one).

It's possible I have misunderstood "oligosynthetic". It's also possible that what I've said about "polysynthetic" is confusingly incomplete (or just "confusing"). Search the CBB, and maybe other places, to find out about those terms, if you want.
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

It"s a matter of point of view...

Sure, as any a priori language (in philosophical sense), we can find it oligosynthetic because of the low and finite amount of roots in it...

To call it polysynthétic, you have to imagine an huge *word and translate it in a *sentence.. But without nucleus, the *sentence is arbitrary ended... and the articulation between *sentences will be a bit artificial... After all, any word contains a story, here the story is just more explicit...

Because morphology and syntax are one, you can also see it isolating, if you want to end the *words with each root... You will find an hudge embeded speech with no center...

I prefer to see it as a picture of a moment where each root is a paintbrush touch...

Of course it is nowever still very experimental...
User avatar
Creyeditor
MVP
MVP
Posts: 5091
Joined: 14 Aug 2012 19:32

Re: Parts-of-Speech of Root Morphemes

Post by Creyeditor »

That there is no obvious, universal way to divide sentences into words is a well known position in linguistics. I like the idea of an experimental conlan based on this idea. A language without a difference between roots and sentences though, would have no double articulation.
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: Parts-of-Speech of Root Morphemes

Post by eldin raigmore »

lsd wrote: .... (your latest post) ....
I could be wrong, but it's my impression you got only some of what I was saying.
Nevermind, though, there'll be time to figure it out later.
In the meantime; Is any of your 'lang ready to show to us yet?


Creyeditor wrote:That there is no obvious, universal way to divide sentences into words is a well known position in linguistics. I like the idea of an experimental conlan based on this idea. A language without a difference between roots and sentences though, would have no double articulation.
A language where every word is a sentence and vice-versa is one thing.
(Maybe that's the same as one where there's no difference between a word and a sentence.)

But a language where there's no difference between a root and a sentence?
That would require either infinitely many roots, or a finite limit to the number of sentences.
I don't think either of those would be realistic nor naturalistic, nor even fit the definition* of "language". Especially it couldn't have the "double articulation" you talked about.

*(Whatever that is!)
User avatar
lsd
greek
greek
Posts: 740
Joined: 11 Mar 2011 21:11
Contact:

Re: Parts-of-Speech of Root Morphemes

Post by lsd »

eldin raigmore wrote:you got only some of what I was saying.
Tell me what points seem still blured...
Is any of your 'lang ready to show to us yet?
It's a kind of thought experiment, I don't save anything I produced in marginalia and others ostraca, except in mind... it's a philosophic or spiritual very personal quest...
Creyeditor wrote:That there is no obvious, universal way to divide sentences into words is a well known position in linguistics. I like the idea of an experimental conlan based on this idea. A language without a difference between roots and sentences though, would have no double articulation.
It's very common in a priori languages to loose the double articulation... Its also a common criticize...
My principal goal was on avoiding the double articulation to obtain a less arbitrary language as possible... Like Leibniz's search and as scientifics languages... The result on words limits is a collateral effect...
eldin raigmore wrote:But a language where there's no difference between a root and a sentence?
That would require either infinitely many roots, or a finite limit to the number of sentences.
IMHO we have to understand no difference between roots compounds and sentence...
Even if any language has a finite limit to the number of sentences (see the library of Babel by JLBorges...) and a more limited number of root we think (see the Wierzbicka's works...)
Iyionaku
mayan
mayan
Posts: 2102
Joined: 25 May 2014 14:17

Re: Parts-of-Speech of Root Morphemes

Post by Iyionaku »

eldin raigmore wrote: But a language where there's no difference between a root and a sentence?
That would require either infinitely many roots, or a finite limit to the number of sentences.
I don't think either of those would be realistic nor naturalistic, nor even fit the definition* of "language". Especially it couldn't have the "double articulation" you talked about.

*(Whatever that is!)
There is a theorie that the first "languages" worked that way. There was a root for "Let's go, we'll hunt moose" and an entirely different root for "Let's go, we'll hunt cheese", where both roots don't bear any phonetic resemblance whatsoever. Of course there were only few roots so speech was still very restricted. Only later the first root were cointed that denoted emotions, specific things, then adjectives, and verbs at last.

I do not know how dubious and/or accepted that theory is, though.
Wipe the glass. This is the usual way to start, even in the days, day and night, only a happy one.
Post Reply