zompist bboard

I've heard somewhere that there are, but I can't remember where. Nor can I think of any other things that can be catagorized under voice.

Antipassive for Ergative langs

And Medium/ Mediopassive, for example in Greek.

My Kazakh grammar (by Bekturov & Bekturova), in addition to active and passive, lists the causative ("impelling"), reflexive, and co-reciprocal as voices (the latter can refer either to two actors peforming an action to each other, or helping each other do something), but I don't think the latter three are described as "voices" in most languages. I suspect that Bekturov and Bekturova call them voices because they, and the passive voice, are all formed by verb affixes in a similar fashion.

p@,
Glenn

Good, good, that was pretty good. But why not try explaining what those terms mean, just in case that idiot con quesa doesn't understand what they mean? [/i]

I really like the reciprocal voice...

Con Quesa: I couldn't tell you some of these, like anti-passive. However:
1. Medio-passive/Middle passive refers to a PIE voice echoed in modern Romance languages as reflexive, where the subject acts as the direct or indirect object, somehow gaining or benefitting from the action. In Greek, there were verbs which were found only in the middle voice, and Latin had verbs (deponent) which were passive in form but active in meaning.

2. The co-reciprocal is where two or more subjects conduct the action unto each other, such as 'they punched each other in the throat at the same time!

3. I would also throw impersonal verbs, which have no agent or patient.

Antipassive is a bit tricky to explain; it's like passive the other way round. In a passive sentence what would be the object (I read the book) becomes the subject (the book was read by me); with antipassive the ergative noun phrase becomes absolutive.

Hmm. This makes me wonder: is there any weird extra voice you can get if your language has a tripartite system?

Despite being indo-european, Swedish has got some interesting voice phenomena.

Just like Latin there are deponent verbs (passive in form but active in meaning). Also, there are reciprocal verbs, which if put into passive mode translate as "each other". An example: Att se, to see, in 1st person plural becomes vi ser. However, to say "we'll see each other" it's perfectly legal to say vi ses, which also means "we are being seen".

Third, there are two ways to turn verbs into passive voice. You can do it the English way: "we were being seen", vi blev sedda, or the aforementioned way, vi s?gs. Only the second way can be reciprocal.

Nifty, huh?

Waldkater wrote:And Medium/ Mediopassive, for example in Greek.

And Icelandic, Sanskrit and German I think.

I guess in a tripartite system they'll be three types of passive. But only one would probably be used, as there's little sense putting an intransitive into the ergative or such like. But i could be wrong.

One of the conlangs I'm working on makes major use of ditransitives, which will have interesting voice implications I guess.


Also, trigger languages can be considered systems with elaborate voicings - eg, not just a subject/object passive but a subject/locative as well.

The medium voice in Greek is, as described, used for verbs that benefit the doer. But I don't think that there are any exclusively mediopassive verbs - I'm pretty sure that any of them can be expressed in the passive voice too. But that being said, some do not appear in Active voice at all.

Drakefyre wrote:The medium voice in Greek is, as described, used for verbs that benefit the doer. But I don't think that there are any exclusively mediopassive verbs - I'm pretty sure that any of them can be expressed in the passive voice too. But that being said, some do not appear in Active voice at all.

Welcome to the board! beware of the pickle monster!

Indeed, welcome!

Salmoneus: what are ditransitives?

Verbs with two objects. In germanic languages, the second object (indirect object) is usually marked by dative case. In English, the two objects are simply placed adjecently.
Eg I call it yellow. I gave him the aardvark. I fed the omniverous mushroom a spatula.

This conlang has far more ditransitive verbs than English does - for instance, the usual verb for "to eat" is ditransitive. It would be better translated "feed".
This gives me an easy way to express intent:
I eat (intransitive) vs. I feed me (ditransitive, missing direct object) vs. I caused it to be eaten (missing indirect object).

I think I might use this to show the aorist, as well: I eat (intransitive, aorist) vs I eat [sth.] (transitive with missing object).

[To clrify: each verb is either transitive, ditransitive or intransive, but affixes can change which it is. Eg. H?nh, to speak, vs. S?unh, to say sth. to sb.]

Ahribar wrote:Antipassive is a bit tricky to explain; it's like passive the other way round. In a passive sentence what would be the object (I read the book) becomes the subject (the book was read by me); with antipassive the ergative noun phrase becomes absolutive.

Hmm. This makes me wonder: is there any weird extra voice you can get if your language has a tripartite system?

Isn't "antipassive" a term used by early linguists encountering ergative langs to explain away its system?

Or something like that.

I might just be talking out of my ass.

Ahribar wrote:Antipassive is a bit tricky to explain; it's like passive the other way round. In a passive sentence what would be the object (I read the book) becomes the subject (the book was read by me); with antipassive the ergative noun phrase becomes absolutive.

Hmm. This makes me wonder: is there any weird extra voice you can get if your language has a tripartite system?

In a word: no. Remember that the third case is the subject of an intransitive verb, which by definition has no object.

Space Dracula wrote: Isn't "antipassive" a term used by early linguists encountering ergative langs to explain away its system?

Or something like that.

I might just be talking out of my ass.

Not quite sure what you mean here. There are many ergative languages that don't make use of an antipassive - just as many nominative ones don't have a passive. Those ones that do have the system though, do require a distinct label.

What the passive and the antipassive both have in common is that they're essentially a system of detransitivising. This means in a nominative system, that the subject loses its special status, and in an ergative system the patient loses its special status. Both can only appear, after all, with fully transitive verbs.

finlay wrote:Drakey's the head admin/mod guy on another board, so it's no wonder he immediately assumed there might be yelling, given that he does a lot of yelling himself .

Does anyone have an example of an antipassive sentence and how it would be translated?

Give me ten minutes and I'll dig out my Yukulta grammar . . .

Oh, and I meant to say "voice", not "mood", above.

(. . .)

Keen (appearing in Dixon and Blake, 1983) wrote:
"The anti-passive transformation is an important syntatic process in Yukulta which 'reduces' the surface transitivity of a sentence by reassigning the A NO with intransitive subject (S) function and the O NP with intransitive (indirect) object function. Compare:

(146)
kungul-i-?-kanta pa:tya
mosquito-ERG-him(ACC)-TR+PAST bite(Vtr)+IND
"The mosquito bit him"

(147)
kungul-ta-thu-yingka pa:tya
mosquito-ABS-me(OBL)-PAST bite(Vtr)+IND"
"The mosquito bit me"

The reason that the second sentence and not the first appears in the antipassive is because in Yukulta, certain subject/object combinations can only expressed in a detransitivised verb (like, as in here, 3rd person subject with first person object). Since Yukulta word order is free, the antipassive does not so much mark focus on a nominal constituent as marking of an "unusual" construction - such as a negative or conditional sentence, or one with a non-prototypical subject-object combination. Anyway, here are some sentences which demonstrate it a bit more clearly:

(148a)
kurritya-ngarri ngumpan-ta miyarl-rta
see+IND-TR+PRES your-ABS spear-ABS
"I see your spear"

(148b)
walirra-kati kurrityarri ngumpan-inytya miyarl-inytya
NEG-I+PRES see+IND+NEG your-DAT spear-DAT
"I can't see your spear"

(149a)
kapatha-nganti ngumpan-ta miyarl-rta
find+IND-TR+FUT your-ABS spear-ABS
"I will find your spear"

(149b)
walirrra-thayi kapatharri ngumpan-inytya miyarl-inytya
NEG-I+FUT fund+IND+NEG your-DAT spear-DAT
"I won't find your spear"

Erk. I can feel that explanation wasn't exactly clear. . .

vegfarandi wrote:

Waldkater wrote:And Medium/ Mediopassive, for example in Greek.

And Icelandic, Sanskrit and German I think.

Faroese for sure, and probably Danish too.

Hm.. would the different intransitive constructions in fluid-S systems be considered seperate "voices"? What would they be called then, I wonder...

Btw, I don't think anyone's mentioned the Inverse voice, which is like the passive only "better"-- instead of demoting the agent to an oblique role, it makes it into the object.

Glenn wrote:My Kazakh grammar (by Bekturov & Bekturova), in addition to active and passive, lists the causative ("impelling"), reflexive, and co-reciprocal as voices (the latter can refer either to two actors peforming an action to each other, or helping each other do something), but I don't think the latter three are described as "voices" in most languages. I suspect that Bekturov and Bekturova call them voices because they, and the passive voice, are all formed by verb affixes in a similar fashion.

p@,
Glenn

I was interested in the question "are reciprocal and reflexive persons or voices?"

As I understand it, they count as voices when applied to verbs, but as persons when applied to pronouns.

------

BTW: While in classical languages the "middle voice" applied mostly to situations in which the agent was a secondary or indirect patient -- that is, significantly (from the speaker's point of view) the agent's own interests were somehow affected by the action -- I have a paper somewhere that lists eleven situations that might be described as "middle diathesis" in various languages (most of which, of course, are not classical).

One of them is "actions which are inherently reciprocal", such as "embrace".

Another group would be any action in which the subject is the agent of part of the predicate but the patient of a different part of the predicate. The prototypical example is when the subject is the patient of the main verb but the agent of the adverb "easily"; as in "this cloth tears easily". A less-prototypical example is when the subject is the patient of the main verb but the agent of an auxiliary, such as "I'm going to get screwed, blewed, and tattooed". (Presumably there may be examples where the subject is the agent of the main verb but the patient of some other part of the predicate.)

Whether or not a language actually marks "I bathe him" (active) vs "I bathe" (reflexive? or middle?) etc. with a different voice, depends on the language; the paper merely listed eleven kinds of semantics that were often (not always) marked as "middle" somehow.

------
Does that help?

-----
Tom H.C. in MI

con quesa wrote:I've heard somewhere that there are, but I can't remember where. Nor can I think of any other things that can be catagorized under voice.

Correct me if I just haven't read this thread closely enough; but I don't think anyone has mentioned the various Applicative Voices.

In some languages, several African languages among them, a participant other than the Subject or the Direct Object can be "promoted" to the grammatical position of "Direct Object". The erstwhile Direct Object, if there was one, is either rendered implicit, or "demoted" to an oblique, adjunct "position".

(If, in a dative language, what gets promoted to D.O. is what was the Indirect Object, this is called "Dative Movement".)

----

There are also transformations called "Dative Applicative", in which a participant other than the Subject or Direct Object or Indirect Object gets "promoted" the to position of "Indirect Object". The former I.O., if there was one, either becomes implicit, or is "demoted" to an oblique, adjunct "position".

----

Just as, for Accusative vs Ergative, there are differences between Passive and Anti-Passive; so, for Dative vs Dechticaetiative, there are differences between Dative Movement and Anti-Dative Movement.

----

Interestingly there is a big overlap between languages with Applicative Voices and languages with Dechticaetiative systems.

This is partly because many languages with Applicative voices are African languages, and many Dechticaetiative languages are also African languages.

A "Dative Language", or "Indirective/Directive Language", is one in which the theme or patient of a ditransitive verb (like give or show or tell) -- the gift, the exhibit, or the tale -- is treated like the patient of a monotransitive verb, while the other -- the recipient, the viewer, or the listener -- is treated in some "new" way. In such languages the gift or exhibit or tale is called "the Direct Object", and the recipient or viewer or listener is called "the Indirect Object". If the cases are different, usually the Indirect Object's case is called "Dative".

A "Dechticaetiative Language" or "Secundative/Primative Language" is one in which the recipient or viewer or listener of a ditransitive verb is treated the same way the patient of a monotransitive verb is treated, while the theme or patient of the ditransitive verb -- the gift or exhibit or tale -- is treated in some new way. In such languages the recipient or viewer or listener is called "the Primary Object", and the gift or exhibit or tale is called "the Secondary Object". If the cases are different, usually the Secondary Object's case is called "Secundative".

-----
Tom H.C. in MI

Xephyr wrote: Hm.. would the different intransitive constructions in fluid-S systems be considered seperate "voices"? What would they be called then, I wonder...

"Active" and "stative", usually. Most intransitive verbs would usually be just one or the other; but some could be either active or stative. Also there are languages in which some non-canonically marked clauses actually have a third form; the only noun is in the dative.

Xephyr wrote: Btw, I don't think anyone's mentioned the Inverse voice, which is like the passive only "better"-- instead of demoting the agent to an oblique role, it makes it into the object.

Sorry for making three posts.

That's not precisely how I'd explain what happens in inverse voice.

Inverse voice applies to the Hierarchical morphosyntactic alignment.

In Hierarchical morphosyntactic alignment, the Direct voice -- the unmarked voice -- signals that whichever participant is higher on the hierarchy is the agent, and the lower participant is the patient.

In Hierarchical morphosyntactic alignment, the Inverse voice -- the marked voice -- signals that whichever participant is higher on the hierarchy is the patient, and the lower participant is the agent.

Usually the verb agrees with the higher participant and does not agree with the lower participant.

So what you said is something like what happens; but I think it could be misleading to not state that it couldn't apply to accusative nor to ergative nor to active/stative nor to tripartite morphosyntactic alignment systems.

-----
Tom H.C. in MI

TomHChappell wrote: I was interested in the question "are reciprocal and reflexive persons or voices?"

As I understand it, they count as voices when applied to verbs, but as persons when applied to pronouns.

I dislike the confusion about what is actually a voice. It seems to me that voice, depending on the linguistic, can mean anything from a limited group of {passive, active, middle} to a large number of valency or argument structure adjusting and/or pragmatic role marking verb operations, either morphological or done via verbal constructions.
I know you've read Klaiman, Tom, since you suggested the book to me. I find it odd that he (she?) accepts things like applicatives as voice, but not causatives? Under his definition "derived voice" and "basic voice" seems to fall every other kind of valency adjustment or argument rearrangement operation it's possible to mark by verbal operations except a few odd cases that he (she?) excludes like causatives.
I believe, I think, that the distinction between voice and non-voice operations is, at the moment, ill founded. It seems to me (as to Klaiman, except with a few bizarre exclusions) that voice operations are marked on or around the verb and do at least one of the following:

i) alters valency or argument structure

ii) mark pragmatic roles (topic, focus)

I propose that we either call everything that performs one of those two functions and associates with the verb a "voice" operation, or discard the term completely in favour of specifying more exactly what we mean when we talk about "voice".
By that definition, reciprocals and reflexives are indeed voice if they are marked on the verb, or via a verb construction of some kind (rather than by a reflexive pronoun say).

chris_notts wrote: I dislike the confusion about what is actually a voice.

Me too, and "voice" isn't the only term I feel that way about.

chris_notts wrote: It seems to me that voice, depending on the linguist, can mean anything from a limited group of {passive, active, middle}

This is the "perspective" or "empathic focus" or "empathic subject" marking; it is the super-system Klaiman calls "Basic Voice". Or, at least, so I understand it. In this system "middle voice" means the subject is the agent, but it and/or its interests is also to some degree affected by the action.

chris_notts wrote: to a large number of valency or argument structure adjusting

This, AIUI, is the "Derived Voice" super-system of Klaiman.

chris_notts wrote: and/or pragmatic role marking verb operations,

This, AIUI, is the "Information-Salience Voice" super-system of Klaiman.

chris_notts wrote: either morphological or done via verbal constructions.

AIUI unless it's marked on the verb somehow it doesn't count as "voice". Often the same information is instead marked on the nouns and counts as "case".

chris_notts wrote: I know you've read Klaiman, Tom, since you suggested the book to me. I find it odd that he (she?)

Because his(?) dedication seemed to be to a woman(? "Madan Madhave Sathe"), I just assumed he(?) was a man.
See
http://assets.cambridge.org/052136/0013 ... klaiman%22

I can't find any biographical data about him(?) on Google.

(There's a much younger MD/LLD with the same name involved in Humanism. That could be the M.H. Klaiman referred to by Wikipedia.)

I tried to look at book reviews to see which pronoun any reviewers used to refer to this author; without success.

chris_notts wrote: accepts things like applicatives as voice, but not causatives?

I'm afraid I neither recall him rejecting causatives nor accepting applicatives. I don't have a copy of that book; did he do either?

In any case clearly both transformations fit into a somewhat-broader idea with what we usually mean when we say "voice" about a language with one of Klaiman's "Derived Voice" systems; they move participants around among the various grammatical relations. Languages without grammatical relations in the first place wouldn't have them; nor would languages whose grammatical relations were strictly determined by semantic/thematic roles (and/or vice-versa) allow them. Whether one wants to call such voice-like transformations "voice" or not could depend -- I don't know what it would depend on, exactly, and different linguists might make different decisions. But some people do indeed call them "voices".

chris_notts wrote: Under his definition "derived voice" and "basic voice" seems to fall every other kind of valency adjustment or argument rearrangement operation it's possible to mark by verbal operations except a few odd cases that he (she?) excludes like causatives.

1. IIRC all of them are "derived voice", not "basic voice".

Some languages with "basic voice systems" do have some "derived voices"; for instance, passive voice occurs in Latin, Greek, Fula, Sanskrit, and (IIRC) Tamil (if that's a "basic voice" language). But that doesn't re-classify those languages, anymore than the fact that passivization occurs in those languages re-classifies it as a "basic" rather than a "derived voice".

2. Did he really exclude causatives? I don't remember that.

He may have declined to talk about it in that book, as opposed to declaring that it didn't count as a "voice"; could you check? I can't.

chris_notts wrote: I believe, I think, that the distinction between voice and non-voice operations is, at the moment, ill founded.

Yep.

chris_notts wrote: It seems to me (as to Klaiman, except with a few bizarre exclusions) that voice operations are marked on or around the verb and do at least one of the following:

i) alters valency or argument structure

"Derived voice".

chris_notts wrote: ii) mark pragmatic roles (topic, focus)

"Information-Salience voice".

--------

You've left out the "Basic Voice"; which marks the empathic or perspectival subject, and/or the empathic or perspectival object.
Klaiman talks about "the grammar of control"; he divides control into two kinds, "agenda control" and "outcome control"; and says control can be partial or shared (or total or absent). Some voices encode such information; "causative" could be one of these -- it encodes that the agent-of-effect (the "causee") may not have been the agent-of-cause (the "instigator").

In that book, as I recall, he doesn't talk about "the grammar of affectedness"; but somebody (maybe him?) does talk about it somewhere (else?). It sometimes happens that one participant -- the direct object, for instance -- is mainly affected, but some other participant is also somewhat affected; and it is the affect on this other participant which mainly concerns the speaker. Some voices encode such information; the "middle voice" is one of those voices, at least in "Basic Voice" systems -- it encodes that the agent (or its interests) is also, somewhat, affected by the action.

chris_notts wrote: I propose that we either call everything that performs one of those two functions and associates with the verb a "voice" operation,

I gather the prevailing habit -- not necessarily a conscious decision -- is to do as you propose.

Klaiman may have set limits; but I think it was so his book would come to an end before he ran out of trees to print it on, just as Siewierska's book doesn't include "obviative" nor "long-distance anaphora" nor any other "fourth persons". She left them out so her book would meet the deadline, not because she's committed to the idea that "fourth persons" aren't really persons. She may have done a better job than Klaiman in saying so, but I think his motives were the same -- even explicitly the same. (I could be wrong.)

chris_notts wrote: or discard the term completely in favour of specifying more exactly what we mean when we talk about "voice".

I believe it would be a good thing to always specify exactly what we mean when we talk about "voice". I don't think this pre-requires discarding the term completely.

chris_notts wrote: By that definition, reciprocals and reflexives are indeed voice if they are marked on the verb, or via a verb construction of some kind (rather than by a reflexive pronoun say).

Apparently so.

But aren't they persons, as well?

Tradition seems to say "no, they're voices but not persons". But I am not sure yet that this is principled.

--------
Thanks for writing!

-----
Tom H.C. in MI

My increasing preference these days is to reserve the term "voice" for clear cases of nearly prototypical passives, middles, reflexives, or antipassives that operate in both the arenas of valency and pragmatics. Everything else - such as causatives, applicatives, "dative shift", argument omission, inverses, triggers, and their ilk - I would lump together under "configuration-mapping operations".


Of course, "configuration mapping" isn't itself a real word. It's pretty much my own personal name for something that doesn't, to my knowledge, have any good, clear, established single term. What I mean by it is the stew of morphosyntactic operations and built-in assumptions in a grammar that together act to 'map' theta roles to morphosyntactic markings. Existing terms like "morphosyntactic alignment" and "valence" refer to things that are part of configuration mapping but can't rightly be said to encompass all of it.