Random ideas: Morphosyntax

[Czwartek]

While playing around with Merthic, I came up with an idea for negation in Merthic. Basically, there is none.

I like this idea because it makes the speaker be a little more expressive and less ambiguous. 'I can't see you', then, could be translated:

'You're not here'
'I'm looking for you'
'You're invisible'
'I'm blind'
'I'm closing my eyes'
'Something between you and me is blocking my field of vision'
'I'm at the wrong airport!'

Recently I've been toying with the idea of having several 'versions' of the essence of a subject of a sentence, according to what type of involvement they have in the verb. One indicates the subject's choice, free will and personal involvement in the verb, one indicates his/her mental, emotional or spiritual involvement, one indicates his/her authoritative involvement, one indicates that they did a automatic, involuntary action, and one indicates that only their presence is involved. Examples:

Direct involvement: The cop stopped the gang in their tracks (he arrested/dispersed them), My dad build this house (he was a builder), I've been looking for you (with my own eyes), Everyone watched in horror.

Mental involvement: My dad believes in ghosts, The man who knew too much, I wish you'd shut up, You promised me the world (until its fulfilled, a promise is just an idea)

Authoritative involvement: My dad built this house (he was in charge of its construction), I've been looking for you (I sent out a search party), I hereby sentence you to death, The king commanded that all prisoners be released.

Involuntary involvement: The baby cried all night, I nearly tripped over, People keep bumping into each other, I love you.

Presence only: The cop stopped the gang in their tracks (the gang stopped when they saw the cop), The boss kept everyone in the office respectful (everyone in the office was respectful on account of the boss being there), The new baby took up all their time. It could also be used for inanimate objects which, by their very presence have an effect on a situation; The wind blew the house down, The scorching heat drew everyone to the beach, The music helps me think. It could also be used for statements such as There's a cat on the table, There's a hole in my pocket, etc.

Any thoughts?

[MrKrov]

Verbal negation via nominals.

[negative noun] ['s] [nominalized verb] = [negated verb]

I suppose a copula would be necessary.

[cybrxkhan]

Alright, so I just remembered I had this idea, and I dunno if I can take it anywhere or whether it's just stupid.

Anyhow, I'm sure most of you have come across "sentences" like these before, spoken or written:

"I'm so angry I want to [insert expletive or very violent action]."
"I think I'm going to [insert weird noises]."
"Oh, I am so clever." "[Insert witty remark]."

Basically, I was wondering if there would be a possible way to grammaticalize the "insert X", so it might be translated as something like "I want to do [unspecified things of category X]. X would usually be some vague category of thing, but I don't know whether this could possibly be done through what kind of grammatical feature, like maybe some weird case or something. Any ideas? I wish I had a better conception of the concept. :-|

[Micamo]

You mean as a replacement for when people actually say "[insert expletive]" or as a generic replacement for expletives?

Either way, you could probably pull this off with a categorizer derivation: An affix expletive-atolo which resolves roughly to "a single expletive, but undefined as to which one." The same function however can be pulled off by articles, so to justify the specific feature's presence I'd leave out articles altogether in a language like this.

[eldin raigmore]

"I'm so angry I want to [insert expletive or very violent action]."

You are aware, aren't you, that that's probably a mis-use of the word "expletive"? Unless you mean the post-Watergate non-linguistic meaning of "bad word".

"The linguistic meanings of "expletive" come from the Latin verb explere, meaning "to fill", via expletivus, "filling out". It was introduced into English in the seventeenth century to refer to various kinds of padding—the padding out of a book with peripheral material, the addition of syllables to a line of poetry for metrical purposes, and so forth. .... expletive is a term in linguistics for a meaningless word filling a syntactic vacancy (syntactic expletives)."

"Syntactic expletives are words that perform a syntactic role but contribute nothing to meaning. Expletive subjects are part of the grammar of many non-pro-drop languages such as English, whose clauses normally require overt provision of subject even when the subject can be pragmatically inferred (for an alternative theory considering expletives like there as a dummy predicate rather than a dummy subject based on the analysis of the copula see Moro 1997 in the list of references cited here)."

In sentences such as
"I forgot to pay the phone bill twice running, so the bloody line was cut off."
bloody contributes nothing to the meaning. Rather, it suggests the strength of feeling (usually anger or irritation, but often admiration, etc.) of the speaker. In having no meaning, it resembles the syntactic expletives discussed above; in these uses, bloody is an expletive. An expletive attributive is a grammatical intensifier.

Other words that are never thought of as offensive can be used in similar ways. For example:
"I forgot to pay the phone bill twice running, so the wretched line was cut off."
The phone line discussed may (before it was cut off) have been just as good as any other, and therefore would not have been wretched in the dictionary senses of "extremely shoddy", "devoid of hope" or similar. Rather, wretched serves here as a politer equivalent of expletive bloody and the like.

[manu]

Idea for a TAM system I'm playing around with for Áressa: Aspect is marked on predicates through inflectional/derivational morphology. There are several aspectual classes of predicates, which determine the types of grammatical aspectual categories the predicates can be marked for. Tense and mode are marked outside of predicates through particles and enclitics. The only obligatory tense distinction is non-future / future. Past tense(s?) are marked through enclitics, and are optional. Mode will be imperative (only marked on verbs), interrogative, and conditional or something like that. This area is probably going to be very loosely connected to the rest of the system.

I don't know, we'll see how this develops.

Interesting, i was actually having a similar idea. If all the TAM markers are placed "around" the verb (as clitics, suffixes, particles), i think the language would still be natural for human beings, although i don't know any language which have the system you described.

I was thinking something more complex and "unnatural": why not to marked TAM on other phrases of the sentence ?? Maybe the object, or (even) the subject. Or at least other elements can agree with the tam system:

example:
John(present) talk(no suffix) with Mary(progressive) = John is talking with Mary...

[roninbodhisattva]

So for a language idea I'm playing around with to test out a classifier system (which I've never done before), I'm coming up with this.

Nouns may not be directly modified by a numeral or demonstrative. Instead, the Num of Dem must be followed by a classifier. Classifiers may also be used with a noun when there is a no Dem or Num, in which case they indicate definiteness (of some sort). Many classifiers in this use will take a special definite form. So they kind of act like articles as well. The definite form will be required if a noun is modified by a relative clause. Some classifiers may also contract with a preceding numeral or demonstrative.

I haven't quite figured out the kind of semantic ground I want the 'definite' form of the classifiers to be. I definitely don't want them to act simply as definite articles. I may end up having them be some form of 'weak' demonstrative- but I haven't figured out what that means exactly. Perhaps they'll just end up signifying specificity.

[Nortaneous]

Read through this entire thread out of boredom.

On the topic of trivocalic roots: Arve has something vaguely similar in its auxiliary verb conjugation, where the verb root conjugates for person with mostly consonantal circumfixes:

1S = r̥-s
2S = ∅-n
3S = ç-ɐ
1P = vʏr̥-s
2P = xʷ-s
3P = ʂ-s

So, to take an arbitrary example, zigg, unconjugated form /t͡sɛj/, conjugates like this: /r̥ɛç ˈt͡sɛjɪn ˈçɛjɐ vʏˈr̥ɛç xʷɛç ʂɛç/. (there's also some fun shit with epenthetic vowels, but that's not important)

With the proper root forms to start, and maybe some infixes, I could see something like a trivocalic (more likely, bivocalic) root system developing.

[Ear of the Sphinx]

In sentences such as
"I forgot to pay the phone bill twice running, so the bloody line was cut off."
bloody contributes nothing to the meaning. Rather, it suggests the strength of feeling (usually anger or irritation, but often admiration, etc.) of the speaker.

BTW, on topic: a conlang declining nouns by "feelings".

[Ceresz]

Oh, I want to try that sometime.

[Micamo]

I once tried something like this with Midhera, but it didn't really work out. It definitely has to be semantic.

...Though this DOES give me the idea of doing this with verb agreement, rather than with noun affixes as I had done before. Hmm...

[roninbodhisattva]

The discussion on clitics makes me want to make a language whose only morphological device would be cliticization.

[Micamo]

You mean English? ;p

[roninbodhisattva]

You mean English? ;p

Touche.

[Micamo]

Here's my idea for the Midhera voice system I came up with today:

Verbs are divided into 7 conjugation classes, depending on what arguments they require.

Impersonal verbs, which take no arguments.
Agentitive intransitives (A)
Patientitive intransitives (P)
Transitives (A + P)
A + Benefactive
P + B
A + P + B

"Voice" markers in Midhera can be interpreted as verbal reclassifiers: They turn a verb of a more-defined class to a less-defined class. This leads to 2 underlying voices: The passive, which removes or merges the A argument, and the antipassive, which removes or merges the P argument. However the voicing is actually done *backwards*, with only one marker. The new intransitive verb is inflected according to its new conjugation, with an extra voicing marker indicating the verb is, by default, transitive. You can tell whether the passive or antipassive voice is being used by the rest of the verb's conjugation.

Notice the benefactive argument, where required, cannot be removed: A semantically identical verb without a required benefactive must be done through lexical choice rather than a voice marker.

[roninbodhisattva]

So verbs can only move down, and not up the transitivity levels? How would causation be marked? Periphrastically?

[Micamo]

I was planning periphrastic causatives, but I still want to do non-periphrastic causatives if I can figure out a way that doesn't suck. (My previous attempts at making non-periphrastic causatives sucked.)

[roninbodhisattva]

Well I mean really all you have to do is have a suffix that add's an A...

[cybrxkhan]

A little idea I thought up a while ago, but I think it's probably exactly like or close to something in a natlang... Anyhow, the original idea was that noun classes / genders would be based on politeness, so instead of say, having a masculine/feminine/neuter distinction, you'd have a noble/commoner/peasant or superior/equal/inferior distinction, with certain objects being classified in a certain group because they're associated with that group, so maybe a throne would be of the superior noun class, while a pile of shit would be of the inferior noun class. I suppose this could be extended further to make the gender classification dynamic and dependent on the situation, so for instance, a king might refer to a noble with the equal noun class, while a peasant might refer to that same noble with the superior noun class. It's sort of like Japan's honorific verb forms, except for nouns... Although I get the feeling it's more or less the same, somehow...

[roninbodhisattva]

. I suppose this could be extended further to make the gender classification dynamic and dependent on the situation, so for instance, a king might refer to a noble with the equal noun class, while a peasant might refer to that same noble with the superior noun class.

I think this would be the most interesting part of a system like this.

It's not exactly the same, but something like the multiple vocabulary levels in languages like Javanese could be a way you could implement this in a naturalistic way. You could say, have different lexical items in certain registers for nouns, and agreement would agree with the register. Perhaps there would be a separate adjective but then on verbs you might get an agreement morpheme based on the register being used in a sentence or something.

Could be cool.