Ergative, accusative and Austronesian do not exist?

[TehranHamburger]

As a separate category I mean. The difference of Accusative and Ergative languages mainly hinges on the passive and antipassive voices supposedly being 'intransitive', but they aren't intransitive at all. 'I am greated by John', the marker 'by' in this case marks an argument of the verb. The only reason why English is not considered ergative is because the passive isn't the 'main' voice. If the active by some magical accident died out and we always said 'By John I am greeted' then English was an SOV language and 'By John' was the ergative marker and what is now called the nominative was the absolutive. This is in fact exactly what the 'split ergative' of many Aryan languages in the perfective aspect is, the use of a passive construct to express perfectiveness. Typically an historical ablative or instrumental case is used as an 'ergative'.

Consider Russian and many slavic languages which basically exhibit perfect Austronesian alignment with an actual case. The instrumental case in this way can be said to be the ergative, the nominative the direct case. The 'passive' voice in this case is the patient-trigger and the active voice the agent-trigger, the only real difference is that in Slavic languages, the agent-trigger is the default voice whereas in Austronesian alignment, the patient-trigger is.

Obviously, in Ergative languages, the same logic applies, it can be said that the 'normal' transitive voice of ergative languages functions like the passive in accusative languages whereas the antipassive functions like the normal transitive voice of accusative languages, with whatever construct they use to itnroduce a semantic patient being the 'accusative'.

This reduction however hinches on the passive and antipassive voices of accusative and ergative langauges actually being transitive. And they are transitive in English, Russian, Latin, Dutch etc. If they aren't transitive however they cannot be reduced to Austronesian languages.

Therefore, theoretically. English, Russian etc are all Austronesian type where the agent-trigger is default. Aboriginal languages are Austronesian type where the patient-trigger is default.

But a hypothetical accusative langauge where the passive voice is actually intransitive and that completely lacks a way to introduce an agent in that case and the nominative argument becomes a patient and the patient must be unspecified. That would be an actual accusative language whch cannot be reduced to Austronesian alignment. The same with ergative langauges and a true intransitive antipassive voice. I'm not sure of the existence of any such language however.

Another weird case can be said to be Finnic languages which do not have a passive voice proper. They have technically a zeroth person which is used to communicate the same idea. Consider 'minut tapetaan' for 'I am being killed', literally '[something unspecified] kills me', note that 'minut' is still in the accusative case as in 'sinä tapat minut', (you kill me) or 'minut tapat sinä' (I am being killed by you). However they do have a passive participle and you can say 'minä olen tapettava sinusta', (I am being killed out of you) but this construction is rare and is not idiomatic as far as I know.

[Vardelm]

I think that with this issue, and issues in other threads, you are glossing over or completely ignoring distinctions that do exist and do make a difference in languages. Yes, you can draw comparisons between different structures of different languages, but that doesn't mean they are the same.

[TehranHamburger]

Feel free to point out a morpholosynthactic difference if you think it exists.

I'm seeing Austronesian alignment being defined having two transitive voices, one of whom aligns the agentive argument with the intransitive argument (direct), the other aligns the patientive and there existing two seperate unaligned arguments (ergative/accusative) to fill the other in both. This exactly matches English and Russian where we can say that the nominative argument fills the spot of the direct, the 'by ...' or instrumental filling the spot of the ergative and the accusative that of the accusative. If you think English in some way does not fit this definition or the definition is wrong, feel feel to point it out.

Edit: To better phrase it, I'm not as much saying that there are similarties between English and Austronesian alignment, I am saying that English and Russian meet every single criterion from the definition of Austronesian alignment.

[Hallow XIII]

Congratulations, you have discovered that underlying roles exist.

[Jipí]

Suggested reading: Van Valin, Robert D., Jr. An Introduction to Syntax. 2001. Cambridge: CUP, 2003. 21–80. Print.

[TehranHamburger]

Just quote the relevant part if you think I'm wrong or better yet point it out. At this point I'm not seeing how English and Russian do not exhaust every criterion for being an Austronesian-type aligned language.

[Vardelm]

To follow up on Sir Gwalchafad's post, I would put it this way:

What you are seeing are situations that are (approximately) the same semantically. However, they are morphologically and syntactically different. They are morphologically different in that one language might use a primary, syntactic noun case for something that another language uses an oblique, prepositional phrase for. They are syntactically different in that different noun cases or prepositional phrases or verb voicing.... or whatever... are required by that language in that situation. That's why it's called morphosyntactic alignment. That is, how the language aligned morphologically and syntactically in order to represent the semantics (underlying roles) of a given situation.

The structures you point to can have certain parallels or points of comparison, and in some cases one can change into the other over time. For instance, the intransitive, passive voice in a nominative/accusative language can be reinterpreted as a basic, transitive voice as the language changes. As a result, it ends up making the language become ergative/absolutive. The semantics may stay about the same. However, that does not mean that the two are morphosyntactically equivalent. The morphology & syntax are different, and it simply changed over time as it was reinterpreted.

In short, you are seeing semantic equivalencies and morphosyntactic parallelisms that do exist and have been noted by linguists (and other members of this board), but you are drawing conclusions about the morphology & syntax that just aren't true.

[Miekko]

TehranHamburger, I suggest you learn how the terminology you use is used by the people who you clearly are misunderstanding.

[Legion]

As a separate category I mean. The difference of Accusative and Ergative languages mainly hinges on the passive and antipassive voices supposedly being 'intransitive

Learn what Ergative actually means, and come back.



Some tips:

1) Accusative/Ergative has to do with how a language marks the semantic roles of agent, patient, and experiencer in a prototypical sentence, and only that.
2) Passive/antipassive voice are operations aimed at removing obligatory arguments of a verb and thus are not directly related to semantic roles.
3) Not all languages have obligatory verbal arguments.
4) Not all accusative languages have a passive voice.
5) Not all ergative languages have an antipassive voice.
6) Some ergative languages have a passive voice.

[TehranHamburger]

To follow up on Sir Gwalchafad's post, I would put it this way:

What you are seeing are situations that are (approximately) the same semantically. However, they are morphologically and syntactically different. They are morphologically different in that one language might use a primary, syntactic noun case for something that another language uses an oblique, prepositional phrase for. They are syntactically different in that different noun cases or prepositional phrases or verb voicing.... or whatever... are required by that language in that situation. That's why it's called morphosyntactic alignment. That is, how the language aligned morphologically and syntactically in order to represent the semantics (underlying roles) of a given situation.

The structures you point to can have certain parallels or points of comparison, and in some cases one can change into the other over time. For instance, the intransitive, passive voice in a nominative/accusative language can be reinterpreted as a basic, transitive voice as the language changes. As a result, it ends up making the language become ergative/absolutive. The semantics may stay about the same. However, that does not mean that the two are morphosyntactically equivalent. The morphology & syntax are different, and it simply changed over time as it was reinterpreted.

In short, you are seeing semantic equivalencies and morphosyntactic parallelisms that do exist and have been noted by linguists (and other members of this board), but you are drawing conclusions about the morphology & syntax that just aren't true.

I feel to see how I'm using a 'semantic' thing, I'm leaving semantics out of, also note how in Russian it is entirely morphologically identical, we have three cases, nominative, accusative and instrumental, on a morphological level these all exist on the same axis. You could argue that 'by ...' in English is distinct because it is the only one marked by an adpositional but the 'ergative' case of many Aryan languages is also marked in a very destinct way from the nominative and accusative and note that many language do not mark the nominative or absolutive at all.

However, if you feel this is wrong, then give me a definition of 'Austronesian alignment' and demonstrate why English does not fall into it. As far as I know. The definition is this:

- 3 verbal arguments, for sake of argument called A, B and C.
- A is always used for the intransitive clause.
- A is used for the patientive argument with B as the agentive argument in a transitive clause in the patient-trigger
- A is used for the agentive argument with C as the patientive in the agent-trigger.

If you feel this definition is wrong, please correct it, if you feel it is correct but English does not meet it by saying that 'by ...' is B, nominative is A and accusative is C, please provide an example where it doesn't.

[Legion]

Also: a prototypical passive sentence is not "the mouse was eaten by a cat"

It's "the mouse was eaten".

[Vardelm]

Your argument boils down to prepositions = noun case, which isn't true.

Also, you've been told by every respondent that you are incorrect, in so many words. I think the onus is on you to go do some homework. It's not the job of forum members to teach you linguistics. I'm sure they'll be happy to make suggestions on where to learn about specific topics or help answer specific questions. Debunking bullshit theories isn't something most are going to feel like spending time on.

[TehranHamburger]

Also: a prototypical passive sentence is not "the mouse was eaten by a cat"

It's "the mouse was eaten".

That's my point, if there was no way to introduce an agent then the passive voice in English would be intransitive and it would fit the definition of an accusative language. But there is a way to introduce an agent and this is grammatical. 'mouse was eaten by a cat' construct with 'by' is completely grammaticalized at this point. It's no longer semantic, it can't be replaced with a word with a similar meaning like 'the mouse was eaten through a cat' or 'with a cat' which are also used to mark instruments alongside 'by'.

There is no possible argument to be made for the passive voice in English to be intransitive and therefore it's not a true passive voice by the letter of the definition. Just because the patientive argument is marked by word order and the agentive by an adpositional changes nothing, the very same thing happens in Aryan languages and they call that an ergative as well.

Your argument boils down to prepositions = noun case, which isn't true.

I have at no point even talked about noun cases. Arguments in alignment can be marked by word order, cases, adpositionals, tone, you name it. The relevant point is that it is marked. The 'nominative' in English isn't a case either, it's marked by word order, not by morphology.

[Radius Solis]

Tehran, when you find definitions and descriptions of things and how they work and feel that they are pointless or that they don't exist as described or otherwise don't make sense, there are two things you might conclude: a) that the experts in this field are therefore stupider than you, or b) that you have not yet learned enough to make proper sense of them. It should take little reflection, of course, to know that the answer will virtually always be B.

Of course, one way to learn more is to present and defend your objections and see what arguments you get back from people who know more about the subject, and that can be an okay strategy so long as you don't make yourself insufferable. I hope that's what's going on here. But you look like you have instead concluded A, and if that is the case, suffice it to say that you will not last very long here.

[TehranHamburger]

Tehran, when you find definitions and descriptions of things and how they work and feel that they are pointless or that they don't exist as described or otherwise don't make sense, there are two things you might conclude: a) that the experts in this field are therefore stupider than you, or b) that you have not yet learned enough to make proper sense of them. It should take little reflection, of course, to know that the answer will virtually always be B.

Of course, one way to learn more is to present and defend your objections and see what arguments you get back from people who know more about the subject, and that can be an okay strategy so long as you don't make yourself insufferable. I hope that's what's going on here. But you look like you have instead concluded A, and if that is the case, suffice it to say that you will not last very long here.

Okay, let's assume I'm wrong here. Let's assume I don't know what I'm talking about. Then tell my why I'm wrong instead of telling me to read books I've already read. There has been only one person who had made any serious attempt and that person hasn't really answered me. Like I said, there are only two possible cases. A: My definition of Austronesian is wrong or B: English does not meet the definition. Just tell me why I'm wrong. That's all I'm asking, and not something like 'Go read this' without even telling the page I should read.

Also note that this position is not unusual at all after some googling. It's even raised on the wikipedia talk page about the article which has this to say back:

No, the difference is that in English active voice is basic and the passive is derived: "Stacy kicked the ball" vs. "the ball was kicked". In Austronesian-alignment languages the two equivalents are parallel: the valency is the same, and one might argue that one is basic because it's more frequent (this would be the one analogous to the passive in English), there's not much else to make such a judgement on.

[Hallow XIII]

This is somewhat difficult when your argument is a jumbled mess of nonsense.

The best guess I can hazard is that you have not yet understood what valency is, which I suggest you go read up on. I hope this will at least somewhat alleviate problems of the sort of

As a separate category I mean. The difference of Accusative and Ergative languages mainly hinges on the passive and antipassive voices supposedly being 'intransitive', but they aren't intransitive at all. 'I am greated by John', the marker 'by' in this case marks an argument of the verb. The only reason why English is not considered ergative is because the passive isn't the 'main' voice.

[TehranHamburger]

By the way, it's not super hard to point out a difference, I finally found someone who could. He argued that the major thing that marked the English active voice as primary to the passive secondary is the existence of actual transitive, not ambitransitive active verbs. You can't say 'I raise' or 'I wear', it absolutely needs an object. Whereas passive verbs are always ambitransitive at best. Thereby he argued that the primary voice should be construed as the one voice wherein true transitivity occurs and that you can define Austronesian alignment as being any language lacking a primary voice via this criterion, either by both, or none of both, exhibiting true transitivity in its members.

See, was that that hard to point out the criterion why English is not Austronesian?

[Xephyr]

There is no possible argument to be made for the passive voice in English to be intransitive and therefore it's not a true passive voice by the letter of the definition.

What is a "true passive voice"?

[Hallow XIII]

By the way, it's not super hard to point out a difference, I finally found someone who could. He argued that the major thing that marked the English active voice as primary to the passive secondary is the existence of actual transitive, not ambitransitive active verbs. You can't say 'I raise' or 'I wear', it absolutely needs an object. Whereas passive verbs are always ambitransitive at best. Thereby he argued that the primary voice should be construed as the one voice wherein true transitivity occurs and that you can define Austronesian alignment as being any language lacking a primary voice via this criterion, either by both, or none of both, exhibiting true transitivity in its members.

See, was that that hard to point out the criterion why English is not Austronesian?

Yes, as it requires lengthy argumenting and I hate doing that because I can't communicate. Thank goodness it is not necessary anymore now.

[linguoboy]

You can't say 'I raise' or 'I wear', it absolutely needs an object.

I guess you've never played poker.

[Radius Solis]

Just tell me why I'm wrong. That's all I'm asking, and not something like 'Go read this' without even telling the page I should read.

A full and solid answer is a big job - there's a lot of side-stuff that plays into it and has to be addressed. I don't think I can do it in less than an hour and a full page of text. Not to mention that most of us have more pressing things to do than be remedial tutors any time someone shows up with broad informational or contextual deficits, which is quite often. There are times I will do it anyway, but this morning isn't one of them. Maybe later.

[TehranHamburger]

There is no possible argument to be made for the passive voice in English to be intransitive and therefore it's not a true passive voice by the letter of the definition.

What is a "true passive voice"?

A true passive voice is intransitive. The English passive voice technically is ambitransitive. The point is that someone I've spoken to was able to demonstrate a formal difference between accusative, ergative and Austronesian-type. Essentially he argued that you can formally define the 'primary' voice as the voice which inhibits true transitive verbs which absolutely require two arguments contrasting the secondary voice as the one that is only inhibited by ambitransitive forms.

You can then define ergative languages as those who align the argument of an intransitive clause with the patientive argument of their primary voice and accusative as those who align it with the agentive of their primary voice and Austronesian languages as those who lack a primary voice altogether. Either not allowing true transitivity in none of the voices, or both.

This excludes English from being Austronesian because the active voice in English has a couple of verbs which exhibit true transitivity while the passive voice has no such. But this necessarily makes accusative languages with two voices which are pro drop Austronesian aligned. But two voice in pro drop languages is rare since the same logic as the passive is often simply expressed by dropping the nominative argument altogether. Finnish might be argued to be Austronesian-aligned though, I believe in transitive clauses you can always drop the accusative argument and the nominative can be dropped with the zeroth person. Finnish is also capable of forming a passive construct but I'm not sure if the 'ergative' in such a situation which uses the illative case is actually grammatical and idiomatic.

A full and solid answer is a big job - there's a lot of side-stuff that plays into it and has to be addressed. I don't think I can do it in less than an hour and a full page of text. Not to mention that most of us have more pressing things to do than be remedial tutors any time someone shows up with broad informational or contextual deficits, which is quite often. There are times I will do it anyway, but this morning isn't one of them. Maybe later.

Like I said, I beg to differ, I was told by someone on IRC in two sentences what the formal difference was. It basically came down to 'The English active voice exhibits true transitivity, the passive voice is always ambitransitive, therefore the active is the primary.'.

It's really not that hard, that establishes a formal difference between the active and passive in English.

[Whimemsz]

It's still wrong to call English passives "ambitransitive." English passives are intransitive, and meet any reasonable criteria for prototypical passives. "By"-phrases are not verbal arguments, they are obliques. This can be seen, for instance, by the fact that they take non-nominative pronouns, while the surface subject of the passive does take nominative pronouns ("I was seen by him"). Thus the only argument of English passives is the surface subject -- the semantic agent is optionally expressed with an oblique NP or else omitted entirely.

[Drydic]

The big question is why you didn't realize that the fact that the majority of English clauses are in the active voice was the biggest argument against what you claimed yourself. Also what Whim said. By the dog in the man was bitten by the dog is not, grammatically, a verbal argument. It is a semantic argument, yes, but those do not have to be the same thing.

[TehranHamburger]

It's still wrong to call English passives "ambitransitive." English passives are intransitive, and meet any reasonable criteria for prototypical passives. "By"-phrases are not verbal arguments, they are obliques. This can be seen, for instance, by the fact that they take non-nominative pronouns, while the surface subject of the passive does take nominative pronouns ("I was seen by him"). Thus the only argument of English passives is the surface subject -- the semantic agent is optionally expressed with an oblique NP or else omitted entirely.

And why is 'I saw him' a verbal argument then?

I seriously don't get why a verbal argument cannot consist of an adpositional and an oblique case. The ergative in Khariboli also is exactly that, the enclitic adpositional -ne which comes at the end of an entire noun phrase and the oblique case being governed by it, just as 'by' comes in front of the entire noun phrase with the oblique case being governed by it. Arguments of verbs can be marked in a variety of ways, cases, word order, adpositionals, what defines a morphosyntactic alignment is how these markings align across different types of clauses. There is no argument against calling the English passive ambitransitive with one argument being marked by coming in front of the verb and the second argument being marked by the propositional 'by'. There are many languages in the world which do not mark the nominative/absolutive at all for instance but mark the accusative or ergative with an adpositional.

The big question is why you didn't realize that the fact that the majority of English clauses are in the active voice was the biggest argument against what you claimed yourself.

Because it's not an argument? Frequency of use is not a formal grammatical argument? Frequency of use is completely irrelevant. Both voices are grammatical.

Also what Whim said. By the dog in the man was bitten by the dog is not, grammatically, a verbal argument. It is a semantic argument, yes, but those do not have to be the same thing.

There is no argument to ever say that 'by ...' is semantic rather than grammatical. The fact that any (non copulative) transitive clause can be transformed into a passive with 'by' regardless of the semantics proves that this is a grammatical argument. It can even be used in such cases where there is clearly no verbal action whatsoever going on such as 'x equals y' and 'y is equalled by x'. 'By' at this point in English is purely grammatical, it cannot be replaced by a synonym like 'through' or 'with' and makes no sense from a semantic perspective. There are cases where you can use an actual semantic adposition, consider 'The train is ran on steam and coal'. But with 'by', not really.