(Conlangs) Q&A Thread - Quick questions go here [2010-2020]

[eldin raigmore]

I think, when the verb is transitive, the participles' principal division is one of voice, between active participles and passive participles, rather one of tense or aspect.

A case could be (and has been IIANM) made that, at least in certain languages, when the underlying verb is intransitive, the main division between participles is one of aspect, imperfective vs perfective, rather than one of tense.

But I don't see why a conlang's participles couldn't inflect for all of tense, aspect, mood, and voice, if the finite verb could.

OTOH:
There's no principled reason to make it a universal that the set of inflectional categories through which verbal adjectives and verbal nouns etc. inflect, has to be a subset or a superset of those through which the finite verb can inflect;
nor is there any reason to make it a universal, that the set of values for which a verbal noun-or-adjective-or-adverb can inflect for some category it shares with the finite verb, must be a subset or superset of those values that the finite verb can have.

For instance, maybe the finite verbs' tenses are {past anterior, past posterior, present, future anterior, future posterior}, while the partciples' tenses are {non-future, future}; or some such thing.

[Thrice Xandvii]

The Spanish forms in -ante -iente are descended from the Latin present participle in -āns /-ēns and are called present participles in traditional grammar. However, they’ve largely lost their participial functionality* in verbal constructions and are mostly used as adjectives and nouns. The gerundio in -ando -iendo has assumed the function of a present participle in forming continuous tenses, so modern grammars may prefer to refer to these forms as present participles.

*Actually, I don’t think Latin ever used the present participle to form any compounds but I’m very possibly wrong.

Regardless of their current state, I think we can agree these words are examples of what was once a participle becoming an agentive noun.

To be clear, I wasn't implying that your point was wrong... or that were wrong really at all. Just clarifying more as a sanity check that I wasn't mistaken in how some things were organized in my head. :)

[Davush]

I'm struggling to name the types of 'genitive' which Qutrussan has.

There is a 'normal' genitive express possession, e.g.

parnă mur
man-GEN house
'the man's house'

An alternative to the above is the s- prefix genitive, it is often used in 'looser' types of relationships:

gilgapatra s-Qutrussán
grammar GEN-Qutrussan
'a grammar of Qutrussan'

And the suffix -(q)qa which generally denotes 'having/possessing' (and is also often used to make nouns adjectives) is used especially when quantifying:

hússaqqa csádam
wheat-GEN field
'a field of wheat'

I also think some adjective forms could be used ambiguously for stylistic reasons:

vallarán csádam
lavender fields (where lavender could be a general description of colour/appearance, or in order to specify).

Qutrussán gilgapatra
Qutrussan grammar (i.e. the grammar of Qutrussan language, or grammar which is Qutrussan in character?)

I would call the first one the basic 'genitive', but I'm not sure about the other two? Especially the third, as it also functions as a derivational suffix. Any suggestions or comments? Does the way the adjectival form can be used seem plausible?

[Creyeditor]

Some questions that might help you:
1. Which constructions can be used predicatively (meaning 'I have a dog.') and which can be used adnominally (meaning 'my dog (is green).')?
2. Which constructions can be used for kinship relations?
3. Which constructions can be used for body parts?
4. Which constructions can be used for part-whole relations?
5. Which constructions can be used for ownership (like you bought something therefore own it)?

These are kind of the standard questions for determing the type of possession? Question one distinguishes between predicative and adnominal possession. If a construction makes you answer yes for questions 2 and 3 but not for 5 it would probably be called inalienable possession. For alienable possession it is the other way round. If the meaning of one construction is much more vague than any other, you could call it 'associative' following i.a. Bantuist tradition.

If the morphological form is different, you might call one the 'strong' genitive, if it is expressed by ablaut or something similar and another one that is expressed by an affix. If some other construction is expressed by a free morpheme, you could call it periphrastic.

---

On the other hand, I think there is no good argument against calling them 'normal genitive', 'loose genitive' and 'adjectival genitive'.

[Davush]

Thanks - those are helpful. I don't think alienable/inalienable possession is a feature of Qutrussan, and none of the above are used predicatively as there is a verb for 'to have'

'Strong' Genitive: (This is probably used mostly for inalienable possession, but isn't restricted to it, maybe it will develop that way in daughterlangs)
parnă mur - the man's house
This form is considered the most archaic.

S-Genitive: (Used for longer noun phrases or looser connections, can also be an alternative to strong genitive in colloquial language)
Phsimmu s-Qutrus 'the island of Qutrus' - 's-' is more or less like Romance 'de'

Associative/Adjectival/Quantifying Genitive: (Maybe I just won't list this as a genitive, but something else)
husaqqa(n) csádam - a field of wheat (i.e. a field having/with wheat)

[Thrice Xandvii]

Is it plausible that all nouns in a language have a non-class related suffix attached to them that is separated when the noun undergoes things like noun incorporation? I ask because I would like it to be clear why only a portion of the word (i.e. the root) is used for certain applications, and yet I don't want the language to have gender/class associated with its nouns. Would it be more plausible to say something like: the last morae/syllable is dropped?

If such a "separable suffix" thing is plausible, do you suppose it would be a smallish set of them (say, 5ish) or could I have upwards of 12?

[Creyeditor]

Is it plausible that all nouns in a language have a non-class related suffix attached to them that is separated when the noun undergoes things like noun incorporation? I ask because I would like it to be clear why only a portion of the word (i.e. the root) is used for certain applications, and yet I don't want the language to have gender/class associated with its nouns. Would it be more plausible to say something like: the last morae/syllable is dropped?

If such a "separable suffix" thing is plausible, do you suppose it would be a smallish set of them (say, 5ish) or could I have upwards of 12?

Grassfields Bantu languages have something remotely similar. The noun class prefixes are often dropped in associative constructions or if they are combined with demonstratives. I think its not much of a stretch to say that in a conlang these kind of morphemes are dropped in noun incorporations. The sets in Bantu languages are of different sizes. 5is and over 12 both make sense.

[Thrice Xandvii]

I suspected someone would bring up Bantu languages. Does it matter that these suffixes wouldn't be in any way related to a class system (maybe they are derived from an older one that is no longer productive or in use)?

[Creyeditor]

I would say that makes it even more realistic. Grassfields noun classes are not your standard Bantu noun class system, they are just ... Bantoid :D

[cedh]

Is it plausible that all nouns in a language have a non-class related suffix attached to them that is separated when the noun undergoes things like noun incorporation? I ask because I would like it to be clear why only a portion of the word (i.e. the root) is used for certain applications, and yet I don't want the language to have gender/class associated with its nouns. Would it be more plausible to say something like: the last morae/syllable is dropped?

If such a "separable suffix" thing is plausible, do you suppose it would be a smallish set of them (say, 5ish) or could I have upwards of 12?

In my conlang Omari, many nouns have a fossilized animacy prefix that must be present when the noun is used on its own (o-čet 'bird'; h-ami 'axe'; i-nųǫgu 'news, knowledge, facts') but which is removed when the noun is incorporated into a verb (tičetšehe 'I went bird-hunting'; šunamekam o-psut 'I felled the tree with an axe'; unųǫgoššok 'we learned [something]'). Granted, these prefixes are also associated with noun class in some way, but this not anywhere near as elaborate and systematic as in Bantu languages. The origin of these prefixes is in 3rd person markers similar to the so-called 'absolutive' affixes in Nahuatl, and the incorporation system (including the feature that incorporated nouns are often phonologically a bit reduced compared to their free counterparts) is modelled on a variety of North American natlangs, although I don't remember which ones specifically. Anyway, I think what you have in mid is quite plausible.

[Ehesh]

On Noun Class systems there tends to be a "catch all" class where stuff that doesn't fit on other languages tend to end up on. Is this a process of languages where Noun Classes tend to become obsolete? I would believe so but not sure. Languages tend to move out from orderly things into messy ones

[Inkcube-Revolver]

Adding on to Ehesh's question, would it be possible for noun class systems getting messy with "catch-all" or "generic" classes like that to result in languages with grammatical gender, such as those with masculine-feminine or animate-inanimate contrasts?

An example to illustrate the thing I'm asking:

agglutinative lang with dozens of noun classes > infected lang with three genders

[Dormouse559]

Adding on to Ehesh's question, would it be possible for noun class systems getting messy with "catch-all" or "generic" classes like that to result in languages with grammatical gender, such as those with masculine-feminine or animate-inanimate contrasts?

An example to illustrate the thing I'm asking:

agglutinative lang with dozens of noun classes > infected lang with three genders

"Class" and "gender" can be thought of as two names for the same phenomenon. They both mean agreement with a noun based on its membership in a category. Your scenario is plausible; there are plenty of examples of languages reducing the number of genders they have (Latin masculine-feminine-neuter to Romance masculine-feminine being one of the classics).

I'd dispute the idea that languages have a tendency to become more "messy". Certainly, sound changes and other alterations can cause older systems to break down, but new systems develop all the time because of processes like analogy.

[Thrice Xandvii]

What are some suggestions to represent creaky voice phonation in a Romanization? (Note: an underdot is already in use, and will need to be able to combine with whatever is suggested.)

[Vlürch]

What are some suggestions to represent creaky voice phonation in a Romanization? (Note: an underdot is already in use, and will need to be able to combine with whatever is suggested.)

You could use digraphs with <ġ>, <ḡ>, <ɋ> or basically any other letter. Personally, I'd go with whatever is the least likely to be confused with another letter, eg. if you use <ġ> for /ɣ/ or whatever, then using that one as part of digraphs like <aġ> might not be a good idea.

[cedh]

What are some suggestions to represent creaky voice phonation in a Romanization? (Note: an underdot is already in use, and will need to be able to combine with whatever is suggested.)

What about using vowel letters with grave (à è ì ò ù) or circumflex (â ê î ô û)?

[Frislander]

I'd dispute the idea that languages have a tendency to become more "messy". Certainly, sound changes and other alterations can cause older systems to break down, but new systems develop all the time because of processes like analogy.

Actually according to Trudgill's Sociolinguistic Typology, there are actually both complexifying and simplifying tendencies, and which ones are active depends on both the size and complexity of the speakerbase as well as the amount of contact a language is under. So high-contact societies with large speakerbases (e.g. English) tend towards simplification, high-contact societies with small speakersbases tend towards complexification by the addition of categories (e.g. Tariana gaining evidentials from Tucanoan contact) while low-contact languages tend towards more complex and irregular exponence of existing categories. The book goes into more detail as to why, and also assesses the repercussions of these ideas. It's a book I'd really recommend taking a look at, it's very readable and isn't soaked through with heavy abstract theory.

[Davush]

Oh thanks for the recommendation Frislander, that sounds like a good read! I always want to know more about the ‘larger’ processes at work but don’t know where to start.

[Dormouse559]

Actually according to Trudgill's Sociolinguistic Typology, there are actually both complexifying and simplifying tendencies, and which ones are active depends on both the size and complexity of the speakerbase as well as the amount of contact a language is under. So high-contact societies with large speakerbases (e.g. English) tend towards simplification, high-contact societies with small speakersbases tend towards complexification by the addition of categories (e.g. Tariana gaining evidentials from Tucanoan contact) while low-contact languages tend towards more complex and irregular exponence of existing categories. The book goes into more detail as to why, and also assesses the repercussions of these ideas. It's a book I'd really recommend taking a look at, it's very readable and isn't soaked through with heavy abstract theory.

I didn't know that. Thanks for the recommendation. Though, I don't know if it makes sense to equate "simple" and "complex" with "messy" and "orderly" (not necessarily respectively).

[Frislander]

Though, I don't know if it makes sense to equate "simple" and "complex" with "messy" and "orderly" (not necessarily respectively).

Well on of the good thing about the book is that is discusses how there are actually different kinds of complexity, only one of which (the increase in irregularity) can actually really be described as being "messy", so yeah I did slightly mishandle my words there, which is all the more reason to read the book because then you're not getting the ideas through the distorting lens of a third party reporter who makes mistakes such as myself.