Talk:Ergative–absolutive language

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Linguistics (Rated B-class)
WikiProject icon This article is within the scope of WikiProject Linguistics, a collaborative effort to improve the coverage of linguistics on Wikipedia. If you would like to participate, please visit the project page, where you can join the discussion and see a list of open tasks.
B-Class article B  This article has been rated as B-Class on the project's quality scale.
 ???  This article has not yet received a rating on the project's importance scale.

-ee in English[edit]

Retiree? Standee? These examples seem a little convoluted - I've never heard or seen them. What about attendee? This, in comparison, is quite common. —Preceding unsigned comment added by (talk) 15:47, 8 April 2009 (UTC)

I believe the morpheme -ee is from French , which uses the morpheme with the verb to describe someone or something that is the object of the verb. Many verbs are transitive but unless the object is otherwise mentioned, they are frequently transitive directly to the subject of the sentence: John has retired (himself): John is a retiree. James is (has) divorced (himself from a previous partner): James is a divorcé. Fred is employed (by someone else's action): Fred is an employee.
I'm not sure of the origin of the use in attendee, which makes no sense to me in this fashion. I attend a meeting; that should make me an attender—the meeting or perhaps the host or speaker should be the attendee. See -- D. F. Schmidt (talk) 18:06, 14 August 2014 (UTC)

syntactic ergativity[edit]


Hmm. Needs info about syntactic ergativity? And perhaps expansion & more example? - Ish ishwar 05:24, 2005 Jan 6 (UTC)

Traces of ergativity in English[edit]

Can this be considered a consequence of the fact that in Latin the past participle is passive? I remember there were some cases in Lating where the past participle was used in an active meaning. So it seems the -ee form can express two different meanings: passive and/or completion. — Sebastian (T) 19:21, 2005 Apr 24 (UTC)

I am only an armchair linguist. But it seems like the behavior of many English verbs, which can be either transitive or intransitive, exhibit something that we might call a "trace of ergativity." These verbs, when transitive, are of the form s-intrans -- verb -- object, but when one noun element is deleted, they are intransitive verbs with a syntax form s-intrans -- verb, where the intransitive subject is the same as the object of the transitive verb. For example:

"I grow flowers" vs. "The flowers are growing."

"I hung a painting on my living room wall" vs. "A painting hangs on my living room wall."

"These changes improve my opinion of this painting" vs. "My opinion of this painting improves with age."

"I showed the apartment to the potential tenant" vs. "This apartment shows very well."

I could list a lot of other examples, but you get the picture. When one noun is deleted from a transitive sentence, with these kinds of verbs, an intransitive sentence is created where the former object of the transitive verb is the subject of the intransitive verb. Is this not a "trace of Ergativy" in English?

EDIT: I added a sentence referencing the article on ergative verbs.

Kdogg36 23:54, 16 November 2005 (UTC)

This has absolutely nothing to do with traces of ergativity. This is simply ambitransitivity -that is, verbs with dual transitivity possibility. You need to be careful in this matter because what you're talking about is a semantic difference which is not a syntactic difference. —Preceding unsigned comment added by (talk) 23:15, 22 January 2008 (UTC)

It seems to me that people speaking accusative languages loves to "find" ergative traces in their own language. It is, however, nothing but an oversimplification of the situation.

That fact that "grow" in "I grow flowers" vs. "The flowers are growing" behaves the way it does, does not make it all right to argue the ergativity case because if you replace the "flowers" with their respective pronouns you get "I grow them" vs. "They grow". Here you can clearly see that the intransitive subject and the transitive object are not treated alike as should be the case, if it were a matter of ergativity.

Thus, this is a semantic difference in the verb "grow", not a syntactic evidence/trace of ergativity.

esegel 0:23, 23 January 2008

They could also be considered causitives, as in 'I boil the spaghetti' or 'I walk the dog'. kwami (talk) 22:39, 29 January 2008 (UTC)

Isn't escape ditransitive? As in The man escaped the prison. Thus it seems to be a poor example, perhaps even invalidating the claim. Then again, I'm no expert. Zophar (talk) 07:03, 5 April 2008 (UTC)

Yes, causatives, indeed. Causatives are often cited as ergatives, and that's simply wrong. The reason, I think, as I wrote above, is because ergativity is an interesting phenomenon for us accusative-speaking people, and we thus like to find traces of it in our own languages. These instances, however, are not genuine instances of ergativity.

To Zophar1. As you know I don't like the ergativity analysis advocated by many, so of course I don't like th "escape" example either. "Escape" in "John escaped" could just as easily be argued to be a trasitive verb with the object omitted, just as "I eat" is transitive even though the actual eaten object has been omitted, as it often is. Regardless, I still don't think that the examples constitute ergativity.

esegel —Preceding unsigned comment added by (talk) 11:19, 16 July 2008 (UTC)

Active-stative alignment[edit]

The following raises a question:

However, there are some intransitive verbs in Georgian that behave like transitive verbs, and therefore employ the ergative case in the past tense. Consider:
Katsma daatsemina. "The man sneezed."
Although the verb sneeze is cleary intransitive, it is conjugated like any other transitive verbs. In Georgian there are a few verbs like these, and there has not been a clear-cut explanation as to why these verbs have evolved this way. One explanation is that verbs such as "sneeze" did use to have a direct object (the object being "nose" in the case of "sneeze") and over time lost these objects, and now used independently without stating the object.

Based on what little I know of Georgian, this is not an irregularity, and it is probably not explained by a former direct object now gone, and it is also not split ergativity, but instead it's simply an active-stative paradigm. I'd like to see what other intransitive verbs behave like this.

I found this very useful paper... Active languages, by Daniel Andréasson, and in the Georgian section it says that:

  • Georgian is always nominative-accusative with regards to verb agreement, and mostly also in case marking;
  • Georgian is almost always ergative-absolutive with regards to case marking in the aorist series (Series II);
  • Georgian is not ergative, but active-stative, with regards to case marking in the aorist series of Classes 2 and 3 (called "the intransitive verbs", even though some are not intransitive and there are some intransitives in the other classes). Class 2 verbs mark their subject as patientive (D. A. uses PAT), and Class 3 verbs mark them as agentive (AGT). That's a (restricted) split-S active alignment.

Now, how do we get all that into the article?

--Pablo D. Flores 10:50, 26 May 2005 (UTC)

Thanks for pointing out the article; I am going to work on expanding this article after I read the relevant sections. Though I have to say that, the theory of "former object having gone" is not something that is just made up to explain the ergative behavior of verbs such as sneeze. I have discussed this with two people who are native Georgian speakers and teach Georgian, and that is how they have explained it to me. This is also not an irregularity, either (I did not notice; when I wrote the explanation of "sneeze", I did not intend to explain this as an irregularity, is that how it sounds like?) Some other intransitive verbs that behave like sneeze are (I am just going to list a few that comes to my mind right now): Cough, stroll, jump, swim, roll, cry, dance,... (basically many class 3 verbs who just behave like class 1,(transitive verbs), so there are many intransitive verbs that use the ergative case in the aorist screeve, again, which is not an irregularity (I am assuming, you know the verb classes in Georgian(?)).

Many questions! What I said was that, as it was put in the article, it sounded like an irregularity plus an excuse for it. However, what I meant (based on what I read in the paper I referenced), is precisely that there is no irregularity. The intransitive verbs are grouped into Classes 2 and 3 based on an active-stative split (Class 3 being active, which in this case means performed by the subject, even if not volitionally as in sneeze). This is a recognized way of organizing intransitive verbs. However, if Georgian is described as being (or implied to be) purely nom-acc or purely erg-abs, then the treatment of verbs in Class 3 looks like an irregularity or a deviant behaviour (marking an intransitive subject with the ergative case!). This is what I think should be addressed carefully. I don't know much Georgian, except that its system is terribly complicated.
The theory of the "former object having gone" is not ridiculous or made up (it's not what I implied), but after what you told me, I think it may be an a posteriori justification. Native speakers and even teachers (who are not linguists) often rationalize facts about their language in this way. I know this because I'm learning a somewhat complicated language myself (Japanese) and I've heard all sorts of "explanations" for features of the language that are not based on linguistic research but on teacher's tradition.
I think I can write something in the article, but I'd like an actual speaker to double-check it, because my knowledge is limited to the trivia above.
--Pablo D. Flores 11:06, 27 May 2005 (UTC)

Then why don't you write what you wish, and I will double check it for you. Although I am not native speaker of Georgian, I have been studying it for a while, and definitely can be helpful.


The analysis of Tagalog as ergative-absolutive is not accepted unanimously. I've seen Tagalog analyzed as a trigger language, leaving the matter of nom-vs-erg completely aside; and as far as I've been able to research, the language does not fit well in any of the known paradigms. This is the conclusion to an online paper about topicalization in Tagalog:

At the beginning of this paper I noted that a controversy exists on the question of whether Tagalog is a nominative-accusative or ergative-absolutive language. This controversy has been predicated on the assumption that topicalization is case-driven movement; that is, that markers like ang and si mark nominals as bearing either nominative or absolutive case. I have tried to show here that this view is incorrect, and that movement to the topic position in Tagalog is more like movement to the pre-verbal slot in languages like Icelandic than it is like passive in English. If this conclusion is right, the question of Tagalog’s case system will have to be completely re-thought. [1]

Another reference:

In another class of languages, the Philippine-type languages, any element of the clause can have unmarked Case; the verb is marked with an affix designating one of its arguments as the nominative (referred to by Philippinists as the topic). [2]

If anything can be turned into a "subject" in Tagalog using voice operations, and no voice is more marked than the others, then it's a moot point to discuss ergativity vs accusativity. At best one should describe Tagalog's morphosyntactic alignment as mixed ergative-absolutive (but that doesn't say much). The first paper I mentioned says Tagalog is syntactically ergative but has some properties of accusative languages (wh-extraction and raising follow the "nominative" argument), and morphologically it is symmetric as to what can be turned into a subject (there's no true passive or antipassive voice).

--Pablo D. Flores 28 June 2005 11:16 (UTC)

Hi Pablo,
Retired linguistics professor and specialist in Austronesian and Philippine languages, Dr. Lawrence A. Reid of the University of Hawai'i, proofread my book In Bahasa Sug: An Introduction to Tausug back in 2002. In it he offered some constructive criticisms as to the way I presented information in my book. The three comments I culled from the Word document deal with ergativity, and they are below.
Tausug is a language that's related to Tagalog and they share many grammatical features.
[LAR14] I would suggest you avoid the use of the ambiguous term ‘object’. Since Tausug appears to be, like most Philippine languages morphologically ergative, you would run into problems, since Nominative typically means grammatical ‘subject’, not ‘object’. In these languages other linguists would label this case as Absolutive, but your use of Nominative fits neatly with what I believe is typologically more appropriate. Why not simply Genitive and Nominative, instead of actor and object?
[LAR43] There is much that I could comment on here. The description of Phil. languages as having a “focus” system, predated the recognition of Philippine languages as generally having an ergative actancy system, and failed to recognise the complete syntactic commonality of the all the non-actor focus constructions which are transitive, having and an Agent and a Patient, or actor and undergoer, from the actor focus constructions which are intransitive, but with no undergoer NP. “Focus” does not place “emphasis” on one of the NPs (emphasis is typically brought about the topicalizing mechanisms), but merely signals in the verb the kind of effect that the verb has on the Nominative NP. I suspect that you are probably not ready to give up the old “focus’ analysis, but I would encourage you to at least note the sytanctic equivalence of the non-actor focus types, and note that they are all transitive, and treat the actor focus types as intransitives, some of which may have more than one complememt, marked either locatively or genitively, but that the second is not a Patient, and is therefore not Accusative. This is not a split ergative system, as has been claimed by some linguists. I could go into arguments for this, but perhaps I am boring you!!
[LAR62] This typological statement is relevant only for accusative languages that have syntactic objects, or Accusative NPs. Tausug is clearly an ergative language and the grammatical subject is the Nominative NP, which in transitive sentences translates as an object. The Genitive NP is not a grammatical subject. Better to characerize the language in terms of the relative order of V and case-marked NPs such as Nom, Gen, Obl, and Locative, etc.
In the end, I did not follow Dr. Reid's advice 100% because I had not yet fully understood ergativity at the time. In the three years that has passed, I realized that I should have done so. I have also been learning some Basque, and I see the similarities between it and Tagalog's (my native language) ergativity system. Being a native speaker of Tagalog made learning Basque just a bit easier. --Chris 30 June 2005 01:59 (UTC)
Tagalog is not a good example of a prototypical ergative language. Stating this in the list without reservations is rather misleading. Many consider languages like Tagalog to a be distinct "Philippine type". I think that this article can mention Philippine langs as similar to ergative and then give arguments for inclusion and exclusion into this category. This should probably be in a separate section. I havent read so much about this, but I can provide references if desired. I think there are issues in determining subjecthood (like some sentences have 2 subjects).
peace – ishwar  (speak) 15:52, 10 October 2005 (UTC)
What do you make of the below? It its from: Starosta S., Pawley A. K. & Reid L. A., 1982, The evolution of focus in Austronesian, in S. A. Wurm & L. Carrington (eds), Papers from the Third International Conference on Austronesian Linguistics, vol. 2 : Tracking the travellers, Pacific Linguistics C-75, Canberra, Australian National University, p. 145-170. I'll just defer to you from now until I gain a better understanding of this. I think it would be wrong, based on what I know & understand so far, to totally do away with any references to Tagalog. Though, I wonder how Samoan differs from Tagalog in its treatment of ergativity. Do you know? --Chris S. 03:04, 11 October 2005 (UTC)
The claim that PAN was a mixed ergative language is based on the following considerations:
1) Within the lexicase framework, an ergative language is defined as one in which the grammatical subject is always in the Patient case relation. A mixed ergative language is one in which the unmarked subject choice is Patient, but which has one or more classes of derived verbs which choose their grammatical subjects according to Fillmore's (accusative) Subject Choice Hierarchy: Agent first, else Instrument or Correspondent, else Patient (using lexicase labels for the case relations).
2) A number of languages from different primary Austronesian subgroups, including Tongan, Samoan, Ilokano, Palauan, Chamorro, Toba Batak, Paiwan, Amis, and Tagalog (cf. DeGuzman 1978:199) are ergative or mixed ergative in the sense of 1) above.
3) In the mixed ergative languages, the ergative verbstems are often less marked than the accusative ones, and the completely unmarked `root stems' (DeGuzman 1978:199) are always ergative in languages such as Kagayanen (Harmon 1977:111, Table 6) and Toba Batak (van der Tuuk 1971:85,98), where `simple passives' consist of a bare stem, while `active' transitive verbs are derived (cf. Mulder and Schwarz 1979:23 on Kapampangan). That is, Toba Batak `simple passives' are grammatically ergative, since the unmarked subject is the Patient rather than the Agent.
I dont know about Samoan or many of the other languages mentioned above. My general impression is that Tagalog is similar to active-stative systems where what is marked are primarily semantic roles, but in addition, Tagalog has pragmatic marking through the topic/focus marker. I think what is being considered here is primarily morphological marking, i.e. the prepositional particles (or whatever they are usually called). How the syntax works is something (even more) outside of my knowledge. I think there are problems determining subjects where different syntactic tests point to different arguments having the status of subject.
Tagalog has been analysed by many people in many different ways. I dont know if there is really any consensus about which is a better analysis. It is probably good to consult general typological and specific language/language family works to get a good picture.
There is good reason to mention these languages here. But, perhaps, they really need their own article. An adequate coverage of the topic will probably be rather technical and definitely require research (at least for me). Also, Wikipedia will need to have better coverage of surrounding topics (like a more rigorous description of subject in its respective article, including tests determining subjecthood). – ishwar  (speak) 23:38, 11 October 2005 (UTC)

Ergativity in Nepali and other IE languages of South Asia[edit]

Nepali has an ergative/instrumental construction by adding -le after a noun. It can be used ergatively with a subject: "Ram-le dunga halyo." meaning "Ram threw the stone." It can also be used instrumentally with an object: "Ma ama-le pakayeko bhat kaye.", meaning "I ate the rice cooked by my mother." Apparently there are parallel constructs in other north indian IE languages.

Rethinking the Page[edit]

I am attempting to get a better grip on the exact function/import of ergativity, but this article is too complex for me to dig through. This is my fourth attempt (in over a month) at trying to figure out what the heck it means through this one Wikipedia article. Isn't there some simpler way to state its definition than "An ergative-absolutive language is one that treats the agent of transitive verbs distinctly from the subject of intransitive verbs and the object of transitive verbs"? I understand what every word of that sentence means. But it is at once thoroughly confusing and also unable to convey the deeper meaning of ergativity. I think it needs some work.

Furthermore, the representation of the ergative vs. accusative section does nothing to elucidate my confusion. I still understand what object/subject/agent mean. But what does the chart represent? Is it saying that the nominative forms of the object are different from those of the accusative in N-A languages? And still, what does it mean? There should be some more thorough examples included in this specific chart. It wouldn't be hard. Just enough to show someone like me what constitutes "different" or "same" once you have made those declarations.

Mind you, I understand a lot about comparative linguistics - though mostly through my own observations and experience - and I am still struggling with the meaning of ergativity, or rather ergativity in contrast with other modes of linguistic expression. Understand very well what transitivity means (i.e. the reason we say "John and I went to the store" instead of "John and me..."). But if someone of my patience and background has trouble understanding the article, doesn't it concern anyone here that other readers/users might have the same troubles? -z42

Well, it's a wiki for a reason; if you have thoughts about how to make the article clearer, you should do so. Appealing to others' concernedness makes no sense to me.
Personally, I think the article explains it rather clearly; whereas nominative-accusative languages (like English) mark the arguments of intransitive verbs (S) the same way as the agents of transitive verbs (A), ergative-absolutive languages mark them the same way as the patients of transitive verbs (O). Whereas English says "I looked around, and I saw him" (using I with both verbs), ergative English would say "Me looked around, and I saw him" (using me with the intransitive verb).
Ruakh 11:20, 26 June 2006 (UTC)
For what it's worth, I don't have any kind of formal linguistic training, and think that the article is clear as it is. The problem is, of course, that the topic is absolutely abstract and assumes that you've already acquired some familiarity with more basic related topics. This is to be expected, since no article can start from scratch. Maybe more examples would be a good idea, though. —Pablo D. Flores (Talk) 14:21, 26 June 2006 (UTC)
I dont't understand what you mean with the "John and me"-example. It doesn't really have much to do with transivity. That a verb is transitive means that it has a subject and an object. Eliae 16:41, 5 September 2006 (UTC)
Yes, it seemed that some different concepts were confused. See my note to z42: User talk:Zeppelin42#Ergative morphology. peace – ishwar  (speak) 19:39, 5 September 2006 (UTC)

I understand exactly why -z42 wrote what they wrote and it would be rather pushing it to advise, as Ruakh does, that someone as confused as -z42 admits to being about what ergativity is should suggest how to clarify the concept of ergativity on Wikipedia. Having just read the article for the first time, I must say that it didn't start well for me either. The first sentence may be a good wide-ranging definition but it is far too general for people like -z42 and me to make much sense of on a first read-through. What does it mean to "treat distinctly from"? The question arises in my mind, "in every respect, only in certain respects or merely in respect to morphology?" Dixon's wording, "treated in the same way", leaves the lay person none the wiser from the start but least he swiftly moves on to indicate that it was through the ergative case that such a phenomenon became a concern in linguistics. Notably this Wikipedia article deals primarily with matters of case in order to explicate the feature; would it therefore not be helpful to define ergativity by anchoring the opening definition within the example of cases as the most direct way of informing the lay linguist? Or is Wikipedia merely to be here for buffs? As this is English wikipedia, one can also validly use contrast with English grammar in order to convey the concept. Perhaps a good opening for the lay reader might read like the following.

""An ergative-absolutive language (or simply ergative) is one that treats the agent of transitive verbs distinctly from the patient of intransitive verbs and the object of transitive verbs. The feature is sometimes found in syntax but is chiefly observable in the morphology of nouns, ie, in an ergative-absolutive language, one might say, as it were, "HE loves HIM" but one would also say "HIM cares" instead of "HE cares".

Nominative-accusative languages such as English use one case for the patient of an intransitive verb and the agent of an transitive verb but another case for the object, eg, "HE cares" (intransitive) and "HE loves him" (transitive). Ergative-absolutive languages work differently by using one case for the patient of an intransitive verb and the object of a transitive verb but another case for the agent, eg, as it were, "HIM cares" (intransitive) and "He loves HIM" (transitive). In this context, HE would be the ergative case and HIM the absolutive case.""

I propose that these simple examples in English are a very immediate way of demonstrating this linguistic feature in terms of case and would be best placed right at the head of the article. The pictures would more usefully come in later as a secondary method of signifying the pattern for the reader's mind once they have already anchored their thinking in example sentences.

I also suggest either removing the word 'subject' and replacing it with the terms 'agent' and 'patient' or removing the words 'agent' and 'subject' and replacing them with the terms 'subject of transitive verb' and 'subject of intransitive verb'. Using the word 'agent' here without the complementary expression of 'patient' and in conjunction with the word 'subject' as terminology to cover both ergative-absolutive and nominative-accusative languages seems inconsistent. If the article is going to use the word 'subject' in relation to nominative-accusative language, it is perhaps inadvisable to create the impression for any length of time during reading that an 'agent' is different from a 'subject'. Likewise, either ergative-absolutive grammar has subjects or it doesn't, of course, and if the article is going to use a grammatical term from nominative-accusative grammar such as 'agent' INSTEAD of the word subject, it should also use the term 'patient' instead of the word subject. —Preceding unsigned comment added by (talk) 10:48, 10 June 2008 (UTC)

The article is not very clear. I think it would be best to start with Basq example sentences to explain the concept. Inasmuch as classical grammar terms were developed from non-ergative languages they should be used only with great cautions. For example, would transitive/intransitive have the same connotations for a Basq native speaker ? Even the term Subject-Object may need a refinement. For example, I would regard Ergativity as a bidirectional relationship A-B (A changes B but B also changes A) as opposed to a unidirectional one. Rather than entering into lengthy discussions into different degrees of ergavitity it would seem much better to simply add a table with ergative languages. Additional criteria (split/nonsplit) can then be added in one or more special columns. The restriction of ergativity to subsets can easily be refined that way (e.g. "aorist only)". Sumerian and Berber Languages should also be present with example sentences. It might be interesting to give some statistical data. For example: How many agglutinating languages (%) are ergative how many are not ? How many ergative languages are tonal / pitched ? It should be mentioned that Kurdish has some ergativity although it is considered to be Indoeuropean. Geographical distribution might be interesting. —Preceding unsigned comment added by Mdenk (talkcontribs) 17:02, 6 December 2008 (UTC)

I'm not sure one can take a Japanese sentence like that and say "in contrast" it is nominative-accusative - if one could perhaps supply another Japanese sentence that is equivalent. What is the relation of -a to -ak in Basque? For example.. I could argue that -ak might be equivalent to the marker of the passive agent "-ni" in Japanese... and make a slight alteration to the verb... then perhaps one could argue that Japanese is in this case also ergative, since the changes between the two sample sentences are equal. and since Japanese marks the absolutive case (-ga) on both the man in the first and the boy in the second - like Basque.

Ergative language
Sentence: Otoko ga tsuita.      Otoko ni kodomo ga mirareta.
Words: otoko ga tsuita      otoko ni kodomo ga mirareta
Gloss: man ABS arrived      man ERG child ABS saw
Function: S VERBintrans      A O VERBtrans
Translation: "The man arrived."      "The man saw the child."

Although I learned to translate the second sentence as the passive, "The boy was seen by the man". Zorgster (talk) 02:02, 6 August 2013 (UTC)

Languages used in Split ergativity section[edit]

In the Split ergativity section, two lines of devanagari Hindi are given, with what I assume is their Latin transliteration below them. But what are the phrases in the boxes; i.e. "Kur pirtûkekê dikire" and "Kurî pirtûkek kirî"? (ps thats is in kurdish/kurmanji " The boy is buying a book" and "The boy bought a book". This is confusing. 13:07, 7 August 2006 (UTC)

Good question. It was added without comment by an unregistered user — — and hasn't been touched since, probably because no one knew what it was supposed to be. I'll remove it, since right now it's not contributing anything at all to the article; if someone can explain what it is, that might make a difference. Ruakh 15:14, 7 August 2006 (UTC)
Never mind, Pablo-flores beat me to it. :-) Ruakh 15:18, 7 August 2006 (UTC)
Hadn't seen your comment. BTW I also took out the devanagari, which is distracting and useless in this context. --Pablo D. Flores (Talk) 15:21, 7 August 2006 (UTC)

Article needs editing![edit]

I, too, find this article unnecessarily obtuse. I'm not a linguist, just someone interested in languages. However, might it not be more informative to provide examples of ergativity (and accusitivity) from the standpoint of ENGLISH than obscure foreign languages? Since it's obvious that anyone on this page understands English, wouldn't it make the most sense to provide a framework for a concept based on a common foundation?

The only part of the article I truly understand is the '-ee' example. Since 'ergativity' must derive from the term for 'work' or 'doer' the concept must have something to do with marking who is doing what.

On the other hand, it would be also helpful if citations were used to provide further help. The accusative example from what little I know appears to be Japanese, it might be helpful to note that for others who might want to track down Japanese linguistic traits.

—The preceding unsigned comment was added by (talkcontribs) .

English isnt an ergative language. Neither is most of Europe. So, we need to show an example of a language that may be "obscure" to European language speakers (excepting Basque, of course). The concept is as explained in the article. It is easiest to show ergativity by comparing the different case marking patterns in an ergative language and an accusative language with the same word order. English also hardly has case marking either (only pronouns are marked for case). English is not suitable to show ergativity. Japanese and Basque conveniently have SOV word order and case markers that follow the noun.
The article does state that Japanese is the accusative language example. Sorry. – ishwar  (speak) 21:41, 17 August 2006 (UTC)
Hi! This being Wikipedia, you should feel free to make changes that you think are helpful. That being said, the problem is that English isn't an ergative-absolutive language. -ee is actually a really bad example of ergativity (it has some ideas in common, but is very different in key ways); the only reason it's included is that it's the closest thing English has. I don't think we can give any more English examples of ergativity. As for accusativity, I think you're right that we should use English rather than Japanese; would you like to start the undertaking of that change? Ruakh 21:45, 17 August 2006 (UTC)
Just thought I'd pitch in and offer a concise explanation of ergativity, based on my admittedly limited understanding of the article. In English and other nominative-accusative languages, the agent/subject is in the same case regardless of whether or not the sentence contains transitive or intransitive verbs. In Basque and other ergative-absolutive languages, the agent/subject is in a different case depending on whether or not the sentence contains transitive or intransitive verbs. Let us pretend for a moment that -ite and -doot are made-up ergative and absolutive case marker suffixes for English. We will use the two example sentences in the article: "The man has arrived" (Intransitive Verb "arrived") and "The man saw the boy" (Transitive Verb "saw"). English - a nominative-accusative language - usually reads like the following: "The mandoot has arrived" (The man is nominated as having arrived), "The mandoot saw the boyite" (The man is nominated as having seen the boy, who stands accused of being seen by the man). An ergative-absolutive English would case-mark the agent/subject separately depending on the presence of a transitive or intransitive verb, like so: "The mandoot has arrived" (The man is in the absolutive case, and is nominated as having arrived), "The manite saw the boydoot" (The boy is in the absolutive case, and is nominated as having been seen by the man in the ergative case, who stands accused of seeing the boy). Even if you change the word order, the marking of the agent and object remain to indicate which is which. "Mandoot has arrived", "Manite boydoot saw". So, that's basically how it works. Ergativity = nominating the subject differently depending on whether the sentence contains a transitive or intransitive verb, unlike a nominative-accusative language which nominates the subject the same way regardless. Either way, it hinges on the nature of the verbs contained in the sentence and how they're applied. Take from that what you will. – A Novice Conlanger (talk) 14:05, 2 March 2011 (UTC)

I suggest that the use of examples of English pronouns in order to demonstrate how English does NOT work shows very clearly to the English speaker how differently to English ergative-absolutive languages DO work. The didactic principle here is contrast, specifically between the morphology of nominative and accusative cases and in relation to agent, patient and object and transitive/intransitive verbs. The use of examples such as "HIM cares" and "HE loves him" etc can and do indicate this. One can then proceed to concrete examples in ergative-absolutive languages to show the precise morphological application in grammatical context in the pertinent languages. —Preceding unsigned comment added by (talk) 11:16, 10 June 2008 (UTC)

Anomolous Passives in English[edit]

Can someone please explain the sentence:

'I was given a book'

I wasn't given at all, the book was given. In Latin, Russian, German etc 'I' in this sentence would be in the dative case. Why don't we say in English:

'Me was given a book'?

Any ideas? Does this have anything to do with ergativity?


No, nothing to do with ergativity, but rather with animacy. Animate and human things tend to be more central to language than inanimate or non-human things, and it's very common among the world's languages for the sole object of verbs like 'give' to be the recipient rather than the patient. You can see in English that we can have it both ways: "he gave the book to me" (core object = book) --> "the book was given to me", as well as "he gave me the book" (core object = me) --> "I was given the book". A lot of languages are restricted to one or the other of those two options. kwami 22:53, 24 May 2007 (UTC)
There are two things going on here. Firstly, English's passive voice can promote direct objects (accusatives), indirect objects (datives), or objects of prepositions (ablatives/prepositional-accusatives); see English passive voice. Secondly, the object so promoted always ends up in the subjective (nominative) case, as in most languages. —RuakhTALK 00:57, 25 May 2007 (UTC)

Linked aorist[edit]

Note that that article needs help, if anyone here is buff on this term. MaxEnt (talk) 19:44, 14 December 2007 (UTC)


The following was in the article: ", and P is indeed more appropriate than is O, because O is biased towards accusative languages. Analogically, it is also defined as Agent of transitive verb below, not as subject. If accusative and ergative languages are to be compared, we need a non-biased terminology, completely omitting subject/object)"

Please refrain from stating what "we" need to do within the actual text of the article, as it is unencyclopedic.--Jeffro77 (talk) 15:41, 28 December 2008 (UTC)

Ergative English[edit]

Here is the example of hypothetical ergative English from the text:

   I (S) traveled; She (S) traveled.
   Me (A) invited she (O) to go with me; Her (A) invited I (O) to go with her.

I speak a split-ergative language myself, and this seems wrong somehow. I'm no linguist, but it seems to me that what we're saying is this:

   Me (S) sleep; Her (S) sleeps. (travel doesn't give ergative in my language)
   I (A) invited her (O) to go with me; She (A) invited me (O) to go with her.

The (S) in the intransitive sentence "feels" more like an object, and the (A) and (O) in the transitive seems reversed. Perhaps it's just my imagination, but could someone with a bit more linguistic knowledge check this, please? -- (talk) 21:23, 12 October 2009 (UTC)

Agree that the way you've phrased the example sounds clearer. Anyone have a third opinion over the next week before I throw it in? --Damian Yerrick (talk | stalk) 15:10, 25 January 2010 (UTC)

The explanation on this page uses too much nonsense jargon[edit]

This page is nearly unreadable without checking the stupid terminology used here. (talk) 00:58, 8 December 2009 (UTC)

Do you complain that you aren't familiar with the technical terminology of linguistics, or do you complain that the article misuses terminology? What precise terminology would you prefer that the article use? --Damian Yerrick (talk | stalk) 15:25, 25 January 2010 (UTC)
I have to agree with the unreadability, I'm a language enthusiast trying to get my head around this concept, and I find myself looking up every second word to find out what it means so that I can understand what's being said. This, in my opinion, is not the best way to explain a concept. Whoever is writing this article, I applaud your efforts and your knowledge in this area is obviously quite extensive, but it may help to assume that everyone reading this article has no linguistic experience at all and is reading this as a one-off. Simple, very clearly explained examples are the key here. Regards, Jimzip

Jimzip (talk) 03:48, 18 November 2010 (UTC)

Trace of ergativity in English, redux[edit]

It seems to me that there is a trace of, or shall we say something analogous to, ergativity in English past/passive participles used as adjectives. Specifically, when the participle of a transitive verb is used as an adjective, its noun is the patient -- the recipient of the action -- and the object of the sentence using the active verb :

He wounded the soldier (transitive) → There is a wounded soldier = a soldier who has been wounded (passive voice)

But when the participle of an intransitive verb is used as an adjective, its noun is also the patient, even though in this case it is the subject of the sentence that uses the active verb:

The leaves fell (intransitive) → Look at the fallen leaves = leaves that have fallen (present perfect aspect)

If this is right, there must be a reference for it, and it could go in the article as something that an English-only speaker might be able to relate to easily. Duoduoduo (talk) 21:11, 30 May 2011 (UTC)

Much too difficult to understand[edit]

This article gets off to a bad start by being ferociously difficult even in its most basic explanation. The first couple of sentences before the baffling diagram might be just about all right, but the part that "explains" the diagram is horrendous. Suddenly the terms "argument" and "core argument" are plucked from thin air and become part of the "explanation". This is an article that can only inform those who already know a lot about the subject, i.e. it is pretty pointless. Wikipedia articles need to be understandable to the general, moderately educated, reader. I cannot really understand it, and I write as someone who studied languages, including linguistics, at a famous old university. Judging by the response to a previous user who complained about this problem, those who are in the know may not have been very sympathetic up to now. Can someone fix it? 08:44, 15 January 2012 (UTC)~ — Preceding unsigned comment added by Pawebster (talkcontribs)

As you can see above, people have been trying to fix it for years. The trouble is that it is inherently difficult for people who speak only an accusative language such as English to get their head round it: there are no satisfactory examples of Ergativity in English.
The only way I can think of to make it clearer would be to start by explaining the relevant facts about English, with examples, and then hit the reader with "Now, all that is what ergativity isn't". That is certainly how I would go about teaching somebody the concept, but I'm very unsure about structuring an encyclopaedia article that way. What do other people think: would that be acceptable? --ColinFine (talk) 19:20, 15 January 2012 (UTC)
Since I began this section, the article has been much improved. I'm grateful to those who have worked on it. What's still lacking, though, is an explanation of the logic behind the constructions favoured in e-a languages. An expression equivalent to "Him sang" is hard to fathom, when the person involved clearly was the one wobbling the vocal chords, just the same as in "He sang a song". APW (talk) 08:07, 20 September 2012 (UTC)

Euskera Error[edit]

The morpheme {-a} does not mark the absolutive is a mark of grammatical number, the absolutive in Euskera is marked with ∅, and the ergative with-k, please clarify this.

Reference in Spanish:

I'm Lingüistic Student of Universidad Nacional de Colombia, thanks for your attention. — Preceding unsigned comment added by (talk) 10:36, 16 May 2012 (UTC)

Maybe we should. It's a question of precision vs. distracting detail. Maybe a footnote. — kwami (talk) 19:46, 16 May 2012 (UTC)

Kurdish is also ergatif language[edit]

I eat bread:

  • Kurmanji:

Min nan xwar. Ez nên (nanî) dixwim.

  • Zazaki:

Min nan werd Ez nanî wena --Alsace38 (talk) 12:16, 17 October 2012 (UTC)

Revision Number 9,001[edit]

Well, I tried to make it easier for English speakers to understand (for instance, replacing early instances of "argument" with "subject", since in English, the argument of an intransitive sentence IS the subject [and it took me 2 hours of MSPaint diagrams to figure that out]), but may have sacrificed complete accuracy in the proccess. The header certainly needed work; you had to look up "argument", and within that page, "predicate", and still didn't understand what you were reading, so I made the changes listed. I also defined the terms "intransitive", "transitive", "agent", and "patient" in the article itself for clarity, so that you won't have to have 5 or 6 tabs open.

I also reversed the "vs" part, just for continuity (so that the whole "A/S/O" thing would flow from the "ergativity" part), but that was just for aesthetics. If it's unsatisfactory, those were all I changed, so just revert my first edit (friggin' coding errors!).

In other words, I tried. :P

Edit: I have no idea what's wrong with the diagrams. I don't remember touching that part... (talk) 04:03, 4 December 2012 (UTC)

I agree the article is not very reader friendly. Unfortunately, your edit compounded the problem, by substituting difficulty with an error: There are no subjects in ergative languages. (Or, rather, some linguists claim there are, but they can't agree as to which case, ERG or ABS, marks the subject.) That just makes things more difficult down the road, when the reader will need to unlearn what you wrote. — kwami (talk) 04:22, 4 December 2012 (UTC)

Could we add an example to the lead, pointing out as we're making it that it's not exact? Such as: "As a rough, but not completely adequate, example, English pronouns align in an nominative-accusative way:

  • I watched
  • He watched
  • I watched him
  • He watched me

Here the subjective (nominative) forms "I/he" are the most basic, and the "me/him" objective (accusative) forms are different. An ergative-absolutive language might instead have:

  • Me watched
  • Him watched
  • I watched him
  • He watched me

Here the absolutive forms "me/him" are the most basic, and the "I/he" ergative forms are different. Ergativity can be expressed in other ways as well, such as which part a verb agrees with when conjugated." I'm also wondering if a crash-course in subject/object vs. agent/patient, probably using English passive voice to illustrate, would be genuinely helpful, or just further confusing and un-encyclopedic. Lsfreak (talk) 05:04, 5 December 2012 (UTC)

Maybe an example like that would work (though I'd put the transitive clauses first, and only list the intransitive ones under 'nom' and 'erg'). I'd avoid passives, however: that's a very different problem. — kwami (talk) 05:39, 5 December 2012 (UTC)

Is this correct ?[edit]

I couldn't follow the text in the original article (and I'm not a total linguistic incompetent)

An ip editor ( ) wrote this text above -

for the lay reader

""An ergative-absolutive language (or simply ergative) is one that treats the agent of transitive verbs distinctly from the patient of intransitive verbs and the object of transitive verbs. The feature is sometimes found in syntax but is chiefly observable in the morphology of nouns, ie, in an ergative-absolutive language, one might say, as it were, "HE loves HIM" but one would also say "HIM cares" instead of "HE cares".

Nominative-accusative languages such as English use one case for the patient of an intransitive verb and the agent of an transitive verb but another case for the object, eg, "HE cares" (intransitive) and "HE loves him" (transitive). Ergative-absolutive languages work differently by using one case for the patient of an intransitive verb and the object of a transitive verb but another case for the agent, eg, as it were, "HIM cares" (intransitive) and "He loves HIM" (transitive). In this context, HE would be the ergative case and HIM the absolutive case.""

I am quite inclined to insert that chunk of text into the article just after the following lines:

These different arguments are usually symbolized as follows:

   O = object of transitive verb (also symbolized as P for "patient")
   S = core argument of intransitive verb
   A = agent of transitive verb


If the examples in that quote are correct then I think I now understand what is meant by ergative-absolute language where the article itself was unclear. I don't know how watched this page is, but if I don't see any feedback in the next week I'll go ahead with that change. I'm still not 100% sure what the quote means by the words 'in this context' so expanding on that would be good. Cheers EdwardLane (talk) 08:49, 10 July 2013 (UTC)

Sentence diagramming[edit]

Perhaps a sentence diagram could help in comparing this type of language with English. I think we can all agree that the terminology is not easily understood by someone who hasn't otherwise studied the distinction between the language types. Fortunately, a sentence diagram might help communicate what terms mean and how to parse and write in languages other than what we're familiar with. Just because this isn't a "simple English" article doesn't mean that the typical reader of English should decidedly not understand it. Also or alternatively, an article covering the various types of languages should cover some distinctions with a sentence diagram. If there is already such an article, it might help to include an obvious link to that. (I, for one, haven't noticed it if it exists.) --D. F. Schmidt (talk) 15:08, 19 August 2014 (UTC)


"The Zazaki, Gorani, Sorani, and Kurmanji are the only ergative languages of the Iranian language family."

But isn't Pashto also an Iranian language (Eastern Iranian, to be exact) which is known to exhibit "split ergativity"? — Preceding unsigned comment added by 2602:304:CC94:6AC0:112:44D9:610D:C7AE (talk) 05:39, 6 October 2015 (UTC)


For instance, instead of saying "she moved" and "I moved her", speakers of an ergative language would say the equivalent of "her moved" and "I moved her".

This is one way of looking at it, but not the only way. You could equally well say:

instead of saying "she moved" and "I moved her", speakers of an ergative language would say the equivalent of "she moved" and "by me she moved".

Indeed this is the most obvious way of looking at languages such as Hindi (see Split ergativity) where you have sentences such as laṛke-ne kitāb xarīdī, which might be glossed as 'by boy book (was) bought' (the verb xarīdī here is a feminine participle, agreeing with kitāb 'book'); similarly in Basque (see Basque language) we are told that some verbs are loan-words derived from Latin passive participles ending in -tus, implying that the Basque-speakers who borrowed those participles felt they were the equivalent of verbs in their own language. So I think this formulation in the introduction is possibly rather misleading. It would be better if it were removed. Kanjuzi (talk) 04:37, 23 January 2016 (UTC)

I agree that these examples (with "she" and "her") are confusing, and anyone with sufficient linguistic understanding to work out what they mean would more easily understand a more symbolic explanation. Whereas anyone who could not understand a more abstract explanation, will just say "What is 'her moved'? It's just ungrammatical." Just looking at the beginning of the article now, I think that the first paragraph of the main body is easier to understand than the lead, except for being somewhat overloaded with terminology. Imaginatorium (talk) 07:08, 23 January 2016 (UTC)
I see that Kanjuzi has now changed the wording to speakers of an ergative language would say the equivalent of "she moved" and "by me moved she". However, that doesn't seem to me to agree with the definition given in the previous sentence at all: that definition says the single argument ("subject") of an intransitive verb behaves like the object of a transitive verb, but in Kanjuzi's example, the intransitive example (she moved) is left unchanged, with the subject playing the same role in what Kanjuzi gives as illustrating the ergative case as it does in English, while the example given of a transitive verb makes a change which does not seem to bear any relation whatever to what the opening sentence says. The earlier version, however, does agree with the definition, though it would make more sense to say "moved her", with "her" treated like an English object, rather than "her moved", which can be seen as an ungrammatical use of "her" as a subject. The editor who uses the pseudonym "JamesBWatson" (talk) 09:04, 15 March 2016 (UTC)


I can't see why the title 'Ergative–absolutive language' has a dash in it rather than a hyphen. This punctuation seems to me rather bizarre. But in any case surely the article would be better titled 'Ergative language'; since if you look up the two phrases on Google ngrams ergative language vs. ergative-absolutive language you will see that the phrase 'ergative language' is much more commonly used, in fact it is over 100 times more common. I propose therefore that it should be changed.

Another thing about this article, however, is that it deals only with ergative-absolutive languages and nominative-accusative languages, without mentioning except in one brief sentence that there are also tripartite languages in which all three participants (agent, object, and subject) are in a different case. The reader has to get quite a long way down the article before he or she comes across that sentence. Would it not be less confusing if the two articles 'Ergative-absolutive language' and 'Tripartite language' were merged? It would surely be much better if that rather useful SOA diagram near the beginning of the article were in three parts, one for each of the three possibilities. Kanjuzi (talk) 05:04, 23 January 2016 (UTC)

"Note that the word subject, as it is typically defined in grammars of nominative–accusative languages, is inapplicable when referring to ergative–absolutive languages, or when discussing morphosyntactic alignment in general."

There seems to be some potential for confusion here, first since the abbreviation 'S' for the 'core argument' of an intransitive verb implies that it is an abbreviation of the word 'subject', but also the writer goes on to refer to ergative languages as being 'SVO' etc. And of course, if it is not appropriate to use the word 'subject' when speaking of ergative languages, it must be equally inappropriate to use the word 'object' of the equivalent in an ergative language of a sentence such as 'John broke the window' (especially as, as I pointed out above, such a sentence could be equally well be glossed as 'By John the window (was) broken' or 'Because of John the window broke'; looked at from this point of view the word 'window' would be the equivalent of the subject, not the object). The words 'subject', 'agent', and 'patient' might be better, providing that it is made clear that 'subject' refers to sentences with intransitive verbs only. Again, I would urge the introduction to be rewritten making it clear that there are three participants, S, A, O, and that in Nominative-accusative languages S and A are treated the same, in Ergative-absolutive languages A and O are treated the same, while in tripartite languages all three are treated differently. Kanjuzi (talk) 05:39, 23 January 2016 (UTC)

I'll just comment on two points. Tripartite languages, together with the other possible alignments, are already in the Linguistic Typology navigation template at the top right. As for the concept of subject being inapplicable to ergative languages, I would disagree. In syntactically ergative languages like Dyirbal it's the absolutive argument that is a subject and in syntactically nominative ones like Hindustani (where case assignment and verb agreement follow a split-ergative pattern), the subject is understood to be the ergative argument (on grounds of anaphora binding, controlling of coreference in dependent clauses, or topic position). The sentence you cite hedges "subject" with "as it is typically defined in grammars of nominative–accusative languages". It might need expanding to make it clear what concepts are relevant for defining subjects. Uanfala (talk) 11:25, 23 January 2016 (UTC)
That's a good idea - mentioning the concept of anaphora binding (do you mean sentences such as 'John broke the window and ran away'? or if not then what?). The idea that 'subject' is an inapplicable concept isn't mine but the original writer's. I still think that the article would be better if it began first by discussing the difference between Nom-Acc, Erg-Abs, and tripartite languages before going into details about the second of these. And are tripartite languages included under the heading 'ergative languages'? This isn't clear. Kanjuzi (talk) 12:03, 23 January 2016 (UTC)
I think I've inadvertently used a generative term for the reflexives (John is unhappy with himself). But any kind of coreference with pronominals or omitted arguments in clauses (like your example) would be relevant, as far as I can tell. Agreed, the lead could do with a bit more context, at the very least a mention of, and link to, Morphosyntactic alignment. Uanfala (talk) 13:32, 23 January 2016 (UTC)