zompist bboard

Are there any languages in which verbs inflect differently in the third person depending on whether the subject is nominal or pronominal, or on some other factor for that matter?

Pertinent to this is the inflection -nte in Quenya, which is glossed as something like "3p when subject is not previously mentioned".

Probably not what you were looking for, but a lot of Dravidian have verbs with suffixes that agree with the subject in gender and number only in third person. This includes situations where the gender of the subject is left unspecified, possibly because using gendered pronouns or suffixes can be considered impolite. (Malayalam doesn't have this morphology, but it does have something of a taboo against using gendered pronouns).

There's also , which sound kind of like what you mention in Quenya.

I was thinking along the lines of, for example, "asks" having different morphology in the following examples:

(He/She/It) asks a stupid question
The wannabe conlanger asks a stupid question
Legalis asks a stupid question

Well, if I remember correctly, in Modern Standard Arabic, the "plural" inflections of a verb occur only when there is no explicit subject or when the subject precedes the verb (a construction that I think historically originated as a form of left-dislocation/topicalization, since Classical Arabic is considered to have VSO as the "unmarked" word order). When the subject follows the verb, the verb agrees in gender, but not in number (it takes the "singular" form).

Here are papers I found that discuss it: http://people.ds.cam.ac.uk/mtb23/NSP/Ch ... Arabic.pdf, http://webspace.buckingham.ac.uk/kbernh ... okarno.pdf

You could look at languages where the subject inflections are still clitic pronouns, at least to an extent. Then have a rule that the subject clitics can never appear together with subject nouns.

I'm out of any good natlang examples of exactly this but it felt natural enough for me to derive two 3rd person subject inflection patterns for Kišta that follow this pattern. One is the regular unmarked form -Ø that's used in most cases, notably with overt NP subjects. The other consists of the anaphoric endings -ti (SG) and -in (PL) that descend from an anaphoric extra subject clitic and are used with omitted 3rd person human subjects. Some examples to show this:

Ketti-n šami-k.
stone-FOC take-SG1
"I take a stone."

Ketti-n šammi-ti.
stone-FOC take-SG3.H
"He/she takes a stone."

Ńaarra ketti-n šammi.
boy stone-FOC take
"The boy takes a stone."

Again, it's not a natlang example, but I'm pretty sure that some amount of digging will turn up something that's reasonably close to this. And even if you don't find anything, I find that working out reasonable grammaticalisation patterns for your morphology helps a lot if you are aiming for naturalism.

Maybe I'm stating the obvious, but I mean, one thing that does happen in a lot of languages is that it's not always necessary to specify who the subject is, even without any verbal morphology to clarify that, if it's already clear from context. In fact, it may even sound awkward to use pronouns too often; this seems to be the case in Mandarin Chinese, for example (or perhaps any variety of Chinese).

You know what, you do see something like this with copulas in some languages. For example, in some English-based creoles such as Krio:

na '(he/she/it) is, (they) are'
mi na 'I am'
yu na 'you are'
wi na 'we are'
una na 'y'all are'
pikin na 'the child is'

Also, forms of the verb for 'to be' in Serbo-Croatian are typically clitics, and they always have to occupy a certain slot in a clause, e.g. (and this is just a quick-and-dirty gloss that I can conveniently make because Serbo-Croatian happens to be an Indo-European language of Europe)

On je visok.
he is tall
'He is tall.'

but

Visok je.
tall is
'He's tall.'

You can analyse Ancient Egyptian that way: sš.f "he writes" vs. sš s "the man writes". But this can also be analysed as the verb not actually inflecting at all, and the apparent personal inflection just being a nominative pronoun that happens to be suffixed to the verb.

_________________
كان يا ما كان / يا صمت العشية / قمري هاجر في الصبح بعيدا / في العيون العسلية

tà yi póbo tsùtsùr ciivà dè!

Welsh is also one of those languages where an explicit 3P form is only possible with a pronominal subject, e.g.:

Fe aeth y blant adre. "The children went home."
Fe aethon nhw adre. "They went home."

Irish is the same, except that it doesn't use a pleonastic pronoun[*]:

Chuaigh na páistí abhaile. "The children went home."
Chuadar abhaile. "They went home."

[*] Except occasionally in Munster, e.g. Táid siad sa bhaile anois. "They're home now."

_________________
كان يا ما كان / يا صمت العشية / قمري هاجر في الصبح بعيدا / في العيون العسلية

tà yi póbo tsùtsùr ciivà dè!

I think the reason that Irish is sometimes analysed as having pronominal agreement there is that throughout the Irish verbal system suffixed forms and forms with no suffix and an independent pronoun are in complementary distribution, often within the same paradigm, and cannot co-occur. So for example for the imperfect you have mhol-ainn 'I used to praise' and mhol-tá 'you used to praise' but mholadh sé 'he used to praise'.

That suffix on the pronouns is presumably cognate to our -thau/-nnau (tithau, minnau for example) which is also contrastive.

_________________
كان يا ما كان / يا صمت العشية / قمري هاجر في الصبح بعيدا / في العيون العسلية

tà yi póbo tsùtsùr ciivà dè!

_________________
Sunàqʷa the Sea Lamprey says: