My Avrelang

[vec]

I've been updating the avrelang Afula lately and I'm wondering what to do with regards to derivation. What I've been able to get from WALS is the following:

• when it comes to inflection, an avrelang is mostly suffixing, and there's no reason why this shouldn't apply to derivation as well
• ordinal numerals are formed with a suffix, which gives us further indication that suffixation is preferred
• adjectives are nominalized, verbalized and adverbialized through conversion, not through affixes
• there are no action nominal constructions, so we shall take this as there being no action nominals altogether, they are kept verbal
• reduplication exists as a mechanism

I'm leaning towards a "Sumerian model", mostly because I've been reading about Sumerian a lot and because it seems to fit in with what we know so far. This would mean adjectives and verbs are roughly a closed class, though verbs can be expanded through the use of object nouns with postpositional arguments so examine would be "see close" and so forth. In the Sumerian model, nouns are productive through compounding but there are few derivational affixes. Verbs could be nominalized through the incorporation of a noun object so say "kitchen" would be food-cook and "pick" (as in the tool) would be stone-break (given Afula's object-verb order).

Does this sound feasible? How common is compounding cross-linguistically and how common are closed-class adjectives and verbs? Does anyone have any data that I could use to fill in WALS's blanks?

[Melteor]

I didn't respond to this initially because its tangential, but I noticed that another Vietnamese based conlang, Frater, does something similar instead of affixes.

[hwhatting]

[*]there are no action nominal constructions, so we shall take this as there being no action nominals altogether, they are kept verbal

What do you mean by that (or what does WALS mean by that)?

[vec]

Action nominals are nouns derived from verbs such as destruction or cooking. Action nominal constructions are constructions that allow the noun to retain its arguments. English uses possessives for both subject and object so in English you can say "his destruction of the city" as an action nominal construction derived from "he destroyed the city".

[Melteor]

What would "him destroying the city" be then?

[Ser]

[*]there are no action nominal constructions, so we shall take this as there being no action nominals altogether, they are kept verbal

What do you mean by that (or what does WALS mean by that)?

I suppose it might mean that there's probably nouns that refer to actions (buying, a purchase, an exchange, a compromise...), but they can't be modified to specify the agent, patient or theme of such actions. Neither with possession constructions, nor relative clauses, nor with the usual markers of finite clauses, nor with particular oblique cases or prepositions, nor MAYBE with noun incorporation.

(I'm not sure if their inclusion of "clausal nominalizations in [...] many North American languages" among the things that don't count for an action nominal construction is supposed to refer to noun incorporation of both subject and object.)

What would "him destroying the city" be then?

One of the various minor patterns, I suppose, with accusative for both agent and patient (us destroying him).

[vec]

Sorry, Serafín gave a better answer than me. I answered a question that wasn't really asked


But yeah, instead of "his destruction of the city" would be "that he destroyed the city" or something like that instead.

[hwhatting]

[*]there are no action nominal constructions, so we shall take this as there being no action nominals altogether, they are kept verbal

What do you mean by that (or what does WALS mean by that)?

I suppose it might mean that there's probably nouns that refer to actions (buying, a purchase, an exchange, a compromise...), but they can't be modified to specify the agent, patient or theme of such actions. Neither with possession constructions, nor relative clauses, nor with the usual markers of finite clauses, nor with particular oblique cases or prepositions, nor MAYBE with noun incorporation.

(I'm not sure if their inclusion of "clausal nominalizations in [...] many North American languages" among the things that don't count for an action nominal construction is supposed to refer to noun incorporation of both subject and object.)

What would "him destroying the city" be then?

One of the various minor patterns, I suppose, with accusative for both agent and patient (us destroying him).

Thanks, that's what I was interested in. So there may be action nounst, they're just not used as replacement for clauses.

[Rainchild]

Great Jumping Jeehosephat! I have independently started a conlang called "Gorokon" that also uses features identified as common on the WALS, though perhaps not as carefully as Averlang does. For example, I use partial reduplication to mark the plural, whereas the plural suffix is actually more common, and so shows up in Averlang.

I suppose I should feel that I've been "scooped," but finding people who are working on a project similar to my own actually warms my heart, and that's no joke.

I'll be following this thread.

[vec]

Thanks for checking in! Reminds me I need to post my updated Avrelang grammar.