Linguistic Miscellany Thread

[Qwynegold]

Spanish verbs like parecer 'seem'? (Though you could argue that in deep structure "parece que X" derives from "X parece".)

How is this verb used? I don't know Spanish.

It's very similar to English "seem", but without the dummy pronoun when it's used with a complement clause.

X parece necesario.
X seems necessary.

Parece que X es necesario.
It seems that X is necessary.

All right, thanks! Now that I think about, it is used similarly in Swedish (with a dummy pronoun) and in Finnish (without).

[Qwynegold]

I was looking for a way to classify some of my verbs, and found this:

Especially in some languages, it makes sense to classify intransitive verbs as:

• unaccusative when the subject is not an agent; that is, it does not actively initiate the action of the verb (e.g. "die", "fall").
•• Unaccusative verbs are typically used to show action or movement.
••• Examples:
•••• I arrived at the party around 8 o'clock.
•••• Do you know what time the plane departed?
•••• How did the disease spread to this town so rapidly?
•••• I sat on the train.
•••• I was in a car accident and the other person appeared out of no where. [3]
• unergative when they have an agent subject.
•• Examples:
••• I am going to resign from my position at the bank.
••• I have to run six miles in the morning.
••• Will you talk to your child about sex before or after they are a teenager?
••• The icecube tray froze solid.[4]

This distinction may in some cases be reflected in the grammar, where for instance different auxiliary verbs may be used for the two categories.

But from what I understood from the explanations; depart, sat and appear should be unergative because they are actions that you actively initiate. Arrive I think could be either or. If you arrived by train, you are in no control of the arriving, but if you arrived by foot you are. And I'm thinking that froze should be unaccusative, because surely icecube tray is not an agent?

But when I read the articles Unaccusative verb and Unergative verb, I again thought I understood these concepts.

In other words, it does not actively initiate, or is not actively responsible for, the action of the verb.

Could you then say that the action in unergative verbs require volition? But then why would the first article say "Unaccusative verbs are typically used to show action or movement"? Don't movement verbs like walk, run, jump, drive typically take an agent subject?

[Estav]

I don't think the relationship between the semantics of a verb and its grammatical status as "unaccusative" or "unergative" is necessarily that transparent. Compare the situation with the temporal structure of the action described by a verb versus the verb's grammatical aspect: even though there is usually some connection, there are a fair number of cases where different languages use different aspects for verbs with similar meanings.

In English, one thing that can be viewed as an indicator that an intransitive verb is "unaccusative" is when the past participle can be used to describe the subject of the verb; e.g. "The newly arrived residents".

That quoted passage has errors in it. Intransitive "freeze" is unaccusative (which the cited source actually says; the editor must have misread the source or accidentally written "freeze" in the wrong list.)

[Whimemsz]

.

[bradrn]

But from what I understood from the explanations; depart, sat and appear should be unergative because they are actions that you actively initiate. Arrive I think could be either or. If you arrived by train, you are in no control of the arriving, but if you arrived by foot you are. And I'm thinking that froze should be unaccusative, because surely icecube tray is not an agent?

But when I read the articles Unaccusative verb and Unergative verb, I again thought I understood these concepts.

In other words, it does not actively initiate, or is not actively responsible for, the action of the verb.

Could you then say that the action in unergative verbs require volition? But then why would the first article say "Unaccusative verbs are typically used to show action or movement"? Don't movement verbs like walk, run, jump, drive typically take an agent subject?

You have just re-invented active-stative split ergativity! See my article on the subject for a fuller description.

But, unfortunately, Wikipedia happens to be wrong here. The real definition is this: an unergative verb is one in which the intransitive subject = the transitive subject, and an unaccusative verb is one in which the intransitive subject = the transitive object. In the aforementioned active-stative languages, the distinction does tend to be one of volition or agentivity, but usually it’s just a syntactic phenomenon. For instance, see in English is unergative, while break in English is unaccusative:

I see you vs I see
I break the window vs The window breaks

There is nothing inherently more volitional here about see vs break; it just so happens that the former is unergative but the latter is unaccusative.

(Also, the distinction only makes sense for ambitransitive verbs. A non-ambitransitive verb is neither unergative nor unaccusative.)

[Kuchigakatai]

Not sure I follow. There's quite a lot of syncretism here eg the 1st person sing pres of rahé turns out to be the same as the 2nd person sing of the past, and the 2nd person sing pres is the same as the 2nd person future and 3rd person future masculine. Normally I would say "So what? Just add the pronoun and a bit of context and it's clear what's happening."

I just wanted to add that you don't need subject persons (1st, 2nd, 3rd) to be clearly stated all the time. You can perfectly have pro-drop languages where speakers are often supposed to understand the subject from context. For example, in Mandarin, what is literally "want do-finish soon" can mean 'I want to finish it soon', 'I want you to finish it soon', 'I want him/her/them to finish it soon', 'if people want to finish it soon', 'if you want to finish it soon', etc., depending on the context. A typical chat conversation in Mandarin can easily go like:

- "already have December monthly pass? can give phone call to receive help [image]" ('Do you already have your monthly pass for December? You can call them for help. [image of the people's contact info]')
- "no. thank information~." ('No. Thank you for the information!')

- "this is today's route. very forest! [image]" ('This is what today's route was like. It was very forested [image of tree branches/leaves narrowing the hiking path].')
- "true! should wear long sleeves." ('True! You (or anybody who goes there) should really wear long sleeves for that.')
- "yeah! also tired/tiring~" ('Yeah! It's also tiring.')

Conlangers who're mostly familiar with European languages tend to assume that you need subject agreement to have a pro-drop language, but this isn't true in much of Southeast and East Asia.


The converse is also true. Russian has lost most verbal finite conjugation (having been largely replaced by participles now), but it retains very clear person agreement in the present tense and in various auxiliary verbs like that of the future construction. For example, for the verb жить [ʒɨtʲ] 'to live [somewhere]', the classic European six forms of agreement (singular and plural) would be:

- present: [ʒɨˈvu ʒɨˈvʲɵʃ ʒɨˈvʲɵt ʒɨˈvʲɵm ʒɨˈvʲɵtʲe ʒɨˈvut]
- future: [budu budʲɪʃ budʲɪt budʲɪm budʲɪtʲe budut] + [ˈʒɨtʲ]

However, subject pronouns are still often used even with these unambiguous verbs in places where pro-drop languages with finite conjugations like Spanish or Italian wouldn't. For example, the phrase "I love you" is famously [ˌja lʲuˈblʲu tʲɪˈbʲe], with the 1SG pronoun я [ja], even though [lʲuˈblʲu] is very unambiguous as it is. Russian does tend to drop the subject pronoun in a number of particular contexts though, like purpose clauses, or when you're answering a question about yourself while repeating the verb of the question, or when the focus of the sentence is a long phrase afterwards.

[Kuchigakatai]

But from what I understood from the explanations; depart, sat and appear should be unergative because they are actions that you actively initiate. Arrive I think could be either or. If you arrived by train, you are in no control of the arriving, but if you arrived by foot you are. And I'm thinking that froze should be unaccusative, because surely icecube tray is not an agent?

But when I read the articles Unaccusative verb and Unergative verb, I again thought I understood these concepts.

In other words, it does not actively initiate, or is not actively responsible for, the action of the verb.

Could you then say that the action in unergative verbs require volition? But then why would the first article say "Unaccusative verbs are typically used to show action or movement"? Don't movement verbs like walk, run, jump, drive typically take an agent subject?

A lot of the motivation behind including verbs of movement is syntactic behaviour observed in European languages. Estav gave you the example of "the newly arrived students". A lot of the discussion about unergative and unaccusative verbs has also been about Romance languages like French, Italian and Old Spanish, which have the same thing with participle modifiers ("newly arrived residents") and also the use of 'to be' as the auxiliary for compound TAMs (where otherwise they use 'to have' instead). Syntactically speaking, there is not much of a different between French il est mort ("he is dead") for 'he died' on the one hand, and on the other il est allé ("he is gone") for 'he went [away]' or il est monté ("he is gone-up") for 'he went up'. French otherwise uses 'to have': il a ri 'he laughed', il a disparu 'he disappeared'.

And this besides the inherent confusion of the terms themselves. In all these years I've never stopped seeing conlangers mixing up "unaccusative" and "unergative" with each other.

I posted a rant about the terms "unaccusative/unergative" a while ago:

Part of the problem is that much of the terminology has been built piece-by-piece, often family by family (see IE vs. Semitic), so there's an awful lot of stuff which hasn't had the chance to be standardised because there hasn't been work which would warrant the terminology to be coined, or alternatively there might have been works that have worked on the same thing but coined different terms.

Yes. Distinguishing the terms used for the adverbial-y "converbs" of Altaic languages from the preposition-like "coverbs" of Chinese more clearly would be a good start... (Hungarian linguistics could also stop using the term "coverbs" for derivational verbal prefixes.)

The awful, awful terms "unaccusative" and "unergative" could also be replaced by the Americanists' much more sensible "patientive" and "agentive", since these are closely-related semantic concepts (an unaccusative verb is a verb with a patientive subject). People in Romance linguistics would also need to be told to stop using "unaccusative/unergative" for syntactic phenomena (French aller 'to go' typically has an agentive subject, stop calling it an unaccusative verb!).

A common definition for "dynamic verb" could also be imposed, so that the struggle between those who use it for continuous actions and those who use it as a synonym for "inchoative verb" would finally stop.

I should've probably said that dynamic verbs aren't just continuous actions though, but also semelfactives like "to hit", and really anything that implies a change of state whether in the subject or the object.

My point was that way too many linguists use "dynamic verb" as a synonym of "inchoative verb", forgetting its opposition to the term "stative verb".

You have just re-invented active-stative split ergativity! See my article on the subject for a fuller description.

But, unfortunately, Wikipedia happens to be wrong here. The real definition is this: an unergative verb is one in which the intransitive subject = the transitive subject, and an unaccusative verb is one in which the intransitive subject = the transitive object. In the aforementioned active-stative languages, the distinction does tend to be one of volition or agentivity, but usually it’s just a syntactic phenomenon. For instance, see in English is unergative, while break in English is unaccusative:

I see you vs I see
I break the window vs The window breaks

There is nothing inherently more volitional here about see vs break; it just so happens that the former is unergative but the latter is unaccusative.

(Also, the distinction only makes sense for ambitransitive verbs. A non-ambitransitive verb is neither unergative nor unaccusative.)

I think you're talking about the occasionally used terms "accusative verbs" and "ergative verbs" there instead? I'm not familiar with a use of "unaccusative" and "unergative" that way. I think Wikipedia is basically right.

[bradrn]

You have just re-invented active-stative split ergativity! See my article on the subject for a fuller description.

But, unfortunately, Wikipedia happens to be wrong here. The real definition is this: an unergative verb is one in which the intransitive subject = the transitive subject, and an unaccusative verb is one in which the intransitive subject = the transitive object. In the aforementioned active-stative languages, the distinction does tend to be one of volition or agentivity, but usually it’s just a syntactic phenomenon. For instance, see in English is unergative, while break in English is unaccusative:

I see you vs I see
I break the window vs The window breaks

There is nothing inherently more volitional here about see vs break; it just so happens that the former is unergative but the latter is unaccusative.

(Also, the distinction only makes sense for ambitransitive verbs. A non-ambitransitive verb is neither unergative nor unaccusative.)

I think you're talking about the occasionally used terms "accusative verbs" and "ergative verbs" there instead? I'm not familiar with a use of "unaccusative" and "unergative" that way. I think Wikipedia is basically right.

Funny, that… I’ve never heard of ‘unaccusative’ being used for anything other than S=O ambitransitive verbs, and I’ve never heard of ‘unergative’ being used for anything other than S=A ambitransitive verbs. If you want a source, as per usual you can find it in Dixon’s Ergativity:

… the S=O type is said to be ‘unaccusative’ and the S=A type ‘unergative’ …

(Although admittedly he spends the rest of the paragraph talking about how many different meanings these terms have, and this particular definition is given as a parenthetical comment.)

[Kuchigakatai]

Funny, that… I’ve never heard of ‘unaccusative’ being used for anything other than S=O ambitransitive verbs, and I’ve never heard of ‘unergative’ being used for anything other than S=A ambitransitive verbs. If you want a source, as per usual you can find it in Dixon’s Ergativity:

… the S=O type is said to be ‘unaccusative’ and the S=A type ‘unergative’ …

(Although admittedly he spends the rest of the paragraph talking about how many different meanings these terms have, and this particular definition is given as a parenthetical comment.)

I honestly think Dixon is using the terms very weirdly there, arguably misunderstanding them, even if he is kind of approaching the definition.

Discussions about unaccusativity/unergativity are most often carried out regarding verbs that are mostly or only intransitive (for unaccusatives, verbs of change of subject state like 'be born / die', monovalent intransitive motion verbs like 'to go', stative verbs like 'to exist'; for unergatives, dynamic verbs like 'to work' or 'to swim'), not ambitransitive verbs where the semantic role of the subject may change, like "to break; to break sth" and "to see sth". Try reading the literature about these terms, much of which is about Romance languages.

Here is a relevant quote of one attempt to define the (still necessarily vague) terms, from Michela Cennamo's "Late Latin Pleonastic Reflexives and the Unaccusative Hypothesis" (1999, Transactions of the Philological Society 97.1). Notice how unaccusative ambitransitives (like Italian aumentare 'to become larger; make sth larger') are talked about as most peripheral, in contrast to core andare 'to go'. Notice, too, that the terms 'Theme' and 'Patient' are used for verbs that can have a lot of agency, but an agency constructed as less-than-stereotypical because the subject undergoes a change of state. (I do find it funny she talks about 'to sleep' as a verb with an agentive subject, but oh well.)

Central to this approach is the idea (hinted at in Van Valin (1993:97)), that Unaccusativity/Split Intransitivity is a gradable phenomenon, determined by the interplay of three ‘dimensions’ referring to the internal structure of the situations described by the verbs, namely the features Dynamic/Static, Concrete/Abstract, Telic/Atelic, with the Agent-like, Theme⁴-like nature of the subject of intransitive verbs. The interaction among these parameters allows to set up a Hierarchy of Unaccusativity/Unergativity, brought out by her studies on the acquisition of morphosyntactic properties of intransitive verbs in Italian L2, illustrated in figure 2 (cf. Sorace 1993a; 1993b; 1995).

Footnote 4: Sorace's notion of Theme differs from the one adopted in this paper, in that it comprises also Participants undergoing a change of state (Patients in our terminology).

In this framework verbs denoting inherently directed change of location (e.g., Italian andare ‘go’) instantiate Core Unaccusativity, in that they have a Theme subject and are the most telic, concrete, dynamic. State verbs denoting existence of a state (condition in Sorace's terminology) (e.g., Italian esistere ‘exist’) lie at the periphery of the Unaccusativity Hierarchy, since they denote static, atelic, abstract situations, though having a Theme subject. Intransitive change of state verbs having a transitive counterpart (paired (dyadic) verbs with a transitive alternant in her terminology) (so-called anticausatives in the typological literature) (e.g., Italian aumentare ‘increase’) and change of location verbs having an atelic counterpart (i.e., a non-directional counterpart, like Italian correre ‘run’), with telicity added compositionally by means of a directional phrase, appear to be most peripheral along the Unaccusative Hierarchy.

The Unergativity Hierarchy consists of three subtypes. Core unergatives have an Agentive subject and denote non motional activity, i.e., static, atelic, concrete situations, as Italian dormire (‘sleep’), with more peripheral unergatives instantiated by atelic verbs denoting change of location, having a telic counterpart and intermediate positions realized by motion activity verbs such as the Italian nuotare (‘swim’).

As we shall see, our data speak for slightly different Unaccusativity/Unergativity hierarchies, with core Unaccusatives realized by verbs denoting change of state, i.e., concrete, dynamic, telic situations with a Patient subject (e.g., Italian perire ‘perish’), and core Unergatives instantiated by concrete, dynamic, atelic situations with an Agentive subject, like the Italian lavorare (‘work’). [...]

Again, I would like to stress that the Romance scholar here is using the terms in typical Romance linguistics ways, talking about (possibly very voluntary!) changes of state in the subject as the crucial thing to distinguish "unaccusative" from "unergative", using Romance syntax as the proof of relevancy for these semantic categories. But the terms were originally used for intransitive verbs with "accusative-case-like subjects" (= semantically patientive subjects, so "unaccusative") and "ergative-case-like subjects" (= semantically agentive subjects, so "unergative"), again using syntax as the proof of relevancy (as with Estav's example "the newly arrived students").

Believe me, I do say my rant against both those terms and Romance scholars' twisted version of them for good reasons.

[Kuchigakatai]

My proposal is that the terms "unaccusative" and "unergative", in their original definition (shown on Wikipedia), should be replaced by "verbs with a patientive subject" and "verbs with an agentive subject" respectively. Or if you need a shorter noun phrase, "pat-subj verbs" (= unaccusative) and "agt-subj verbs" (= unergative).

Meanwhile, Romance scholars should be told to stop using those terms too, and switch to, uhhh, "self-mutative monovalents" (= unaccusative) and "stable-state (or stable-subject) action monovalents" (= unergative). They could further specify that they usually talk about self-mutative monovalents in the context of telic, perfective-aspect notions in particular.

[akam chinjir]

I take it that if you do have an S=O ambitransitive, the fact that it's S=O is a pretty good indication that it's unaccusative in its intransitive use. And for good reason, an unaccusative verb is one whose single argument is like the object of a transitive verb. Literally, in relational grammar, where the idea comes from, and also in Chomskyan linguistics (where you say "internal argument" instead of "initial 2," to draw about the same distinction). (For the origins, Perlmutter's Impersonal Passives and the Unaccusative Hypothesis is still fun.)

Ser, do you know of a language that clearly treats the basic motion verbs as unergative? Like, an active/stative language where their subjects are ergative? Because I don't read a lot about Romance languages, and I've definitely acquired the idea that you expect them to be unaccusative. (Though in the linked paper Perlmutter says "they typically involve ambiguities and the possibilities for alternative analyses similar to those observed with slide." ---He's just contrasted "Joe slid into third base" with "Joe slid on the ice" and "the wheels slid on the ice.")

[Whimemsz]

The paragraph from Dixon that bradrn cited is actually referring to Perlmutter's original paper; Dixon seems to be referencing "unaccusative" and "unergative" as descriptors for ambitransitives and split ergative phenomena because one of the things Perlmutter briefly mentions is that Boas and Deloria's old description of Dakota (which is active/stative) looks to him like an example of a language clearly distinguishing unaccusative and unergative verbs -- so Dixon includes Relational Grammar as one of the theoretical models which has attempted to deal with ergativity (which is the broader context of the quotation). But most of Perlmutter's paper is not about that, it's about impersonal passives (mostly in Dutch), and most of the subsequent discussion in the literature has been about issues that Ser has mentioned.

I would also caution against an over-reliance on Dixon or thinking he always represents the """mainstream""" on some given point, since he has some idiosyncratic views, although this case isn't so much an instance of that, as he's just mentioning the cases where "unaccusative" and "unergative" have relevance to his model of ergativity, and he notes that the terms are also used in reference to other phenomena.

[Kuchigakatai]

I take it that if you do have an S=O ambitransitive, the fact that it's S=O is a pretty good indication that it's unaccusative in its intransitive use. And for good reason, an unaccusative verb is one whose single argument is like the object of a transitive verb. Literally, in relational grammar, where the idea comes from, and also in Chomskyan linguistics (where you say "internal argument" instead of "initial 2," to draw about the same distinction). (For the origins, Perlmutter's Impersonal Passives and the Unaccusative Hypothesis is still fun.)

I don't have any objections to the concept, just the terminology, because I think people reasonably tend to find it morphologically confusing in terms of the intended metaphor. And regardless of what Perlmutter, Postal and Pullum intended, Romance scholars tend to focus on the syntax of unaccusative verbs that are just intransitive/monovalent (in some contexts of completed change of state), like nacer 'to be born', morir 'to die', venir 'to come', ir / andar 'to go' and llegar / arribar 'to arrive', which are not used as transitives with a direct object in Romance. As in my quote from a paper above, the likes of 'to break; break sth' or aumentar 'to increase; increase sth' are considered more peripheral unaccusatives in that field, even though they're semantically unaccusatives too.

BTW, bradrn, I do encourage you to read that paper akam chinjir just linked to... Note the basic example of an unaccusative verb given (in the 4th page, or page 160) is "gorillas exist", not an ambitransitive verb...

Ser, do you know of a language that clearly treats the basic motion verbs as unergative? Like, an active/stative language where their subjects are ergative? Because I don't read a lot about Romance languages, and I've definitely acquired the idea that you expect them to be unaccusative. (Though in the linked paper Perlmutter says "they typically involve ambiguities and the possibilities for alternative analyses similar to those observed with slide." ---He's just contrasted "Joe slid into third base" with "Joe slid on the ice" and "the wheels slid on the ice.")

Man, I am familiar with only a few languages... I don't know about active/stative languages, but would the restrictions on motion verbs in Latin suffice? Latin doesn't allow motion verbs to be in their passive participle form unless a compound TAM of the impersonal use is involved, as in ventum est 'people in general came by' (literally "it is come"). Unlike English and modern Romance, you can't say the likes of "the recently-arrived students" or "the guys gone shopping".

But this is just syntax. In terms of semantic unaccusative/unergative, Latin would not distinguish "Joe slid into third base" and "Joe slid on the ice", no, so I don't know.



BTW, a fun sentence I just happened to find in the meantime, from Ennius' Iphigenia (the fragment quoted in Cicero), with three verbs used impersonally (in bold). Hey, this is the miscellany thread after all.

Imus huc, hinc illuc. Cum illuc ventum est, ire illinc lubet. / Incerte errat animus; praeter propter vitam vivitur.
(literally) "We go hither, hence thither. When it is come thither, it pleases to go thence. / The spirit roams uncertainly; it is lived beyond for life."
(not literally) 'We move from elsewhere to here, and from here to there. And once we arrive there, we feel like leaving. Our spirit roams, always uncertain, and we keep living for the sake of living.'

[bradrn]

Believe me, I do say my rant against both those terms and Romance scholars' twisted version of them for good reasons.

My proposal is that the terms "unaccusative" and "unergative", in their original definition (shown on Wikipedia), should be replaced by "verbs with a patientive subject" and "verbs with an agentive subject" respectively. Or if you need a shorter noun phrase, "pat-subj verbs" (= unaccusative) and "agt-subj verbs" (= unergative).

Meanwhile, Romance scholars should be told to stop using those terms too, and switch to, uhhh, "self-mutative monovalents" (= unaccusative) and "stable-state (or stable-subject) action monovalents" (= unergative). They could further specify that they usually talk about self-mutative monovalents in the context of telic, perfective-aspect notions in particular.

I agree that those are bad terms. There are a couple of others which I have complained about before (infinitive, adverb, participle), all of which are also vaguely defined terms originally from Romance or Latin.

I would also caution against an over-reliance on Dixon or thinking he always represents the """mainstream""" on some given point, since he has some idiosyncratic views, although this case isn't so much an instance of that, as he's just mentioning the cases where "unaccusative" and "unergative" have relevance to his model of ergativity, and he notes that the terms are also used in reference to other phenomena.

Could you elaborate on that? Since his book Ergativity by far the most comprehensive and detailed book on the subject I have found, I’ve been using it as the main source for my series on ergativity. If there are any problems with it, they would be very useful to know.

BTW, bradrn, I do encourage you to read that paper akam chinjir just linked to... Note the basic example of an unaccusative verb given (in the 4th page, or page 160) is "gorillas exist", not an ambitransitive verb...

I will!

EDIT: I’ve tried reading it, but I couldn’t make it past the second page — they seem to be using some formal theory which I can’t understand.

BTW, a fun sentence I just happened to find in the meantime, from Ennius' Iphigenia (the fragment quoted in Cicero), with three verbs used impersonally (in bold). Hey, this is the miscellany thread after all.

Imus huc, hinc illuc. Cum illuc ventum est, ire illinc lubet. / Incerte errat animus; praeter propter vitam vivitur.
(literally) "We go hither, hence thither. When it is come thither, it pleases to go thence. / The spirit roams uncertainly; it is lived beyond for life."
(not literally) 'We move from elsewhere to here, and from here to there. And once we arrive there, we feel like leaving. Our spirit roams, always uncertain, and we keep living for the sake of living.'

I wouldn’t have a clue about the sentence itself (I can’t speak Latin), but your translation is beautiful.

[Qwynegold]

In English, one thing that can be viewed as an indicator that an intransitive verb is "unaccusative" is when the past participle can be used to describe the subject of the verb; e.g. "The newly arrived residents".

Hmm, let me test this on those example verbs on WP.

Unaccusatives (grammaticality expected):
The recently arrived train has malfunctioned.
?The departed train has malfunctioned.
The spread disease is starting to become a huge problem.
?The pie sat on the windowsill is still hot.
?The mysteriously appeared figure freaked everyone out.

Unergatives (ungrammaticality expected):
The resigned CEO started a new company.
*The run dog is exhausted.
*The talked parrot wants a cracker.

This seems very inconclusive. Am I wrong with these sentences?

Either way, this all depends on the given language. In Finnish I think anything can take a past participle.

[Qwynegold]

You have just re-invented active-stative split ergativity!

Not again!

See my article on the subject for a fuller description.

Ah, I've been meaning to read that for a long time now. I'll just post my replies on this thread first, or otherwise I'll forget what my thoughts were.

The real definition is this: an unergative verb is one in which the intransitive subject = the transitive subject, and an unaccusative verb is one in which the intransitive subject = the transitive object.

[...]

(Also, the distinction only makes sense for ambitransitive verbs. A non-ambitransitive verb is neither unergative nor unaccusative.)

That makes a lot of sense to me. But in the conlang I'm working with there are no ambitransitive verbs. You need to use some kind of valency changing operator.

Anyhow, people seem are disagreeing a lot about what these terms really mean. Now I'm thinking they might not be appropriate for describing the different kinds of verbs in my conlang.

[bradrn]

See my article on the subject for a fuller description.

Ah, I've been meaning to read that for a long time now. I'll just post my replies on this thread first, or otherwise I'll forget what my thoughts were.

I agree that it’s probably a good idea to reply here first. As my ergativity thread, unfortunately it’s not done yet, so at the moment you can only read about half of it. But I’m working on it (admittedly very slowly…), and hopefully the next part should be posted soon!

The real definition is this: an unergative verb is one in which the intransitive subject = the transitive subject, and an unaccusative verb is one in which the intransitive subject = the transitive object.

[...]

(Also, the distinction only makes sense for ambitransitive verbs. A non-ambitransitive verb is neither unergative nor unaccusative.)

That makes a lot of sense to me. But in the conlang I'm working with there are no ambitransitive verbs. You need to use some kind of valency changing operator.

Well, I would say that if you have no ambitransitive verbs, then unaccusative/unergative doesn’t make much sense. But as the discussion above showed, clearly I have no idea what I’m talking about here, so it’s probably best to disregard what I’m saying.

Anyhow, people seem are disagreeing a lot about what these terms really mean. Now I'm thinking they might not be appropriate for describing the different kinds of verbs in my conlang.

What ‘different kinds of verbs’ are those? If you give a description, there may well be some well-known and less ambiguous terms that you could use for them.

[zompist]

Hmm, let me test this on those example verbs on WP.

These would be my grammaticality judgments:

Unaccusatives (grammaticality expected):
The recently arrived train has malfunctioned.
*The departed train has malfunctioned.
*The spread disease is starting to become a huge problem.
*The pie sat on the windowsill is still hot.
*The mysteriously appeared figure freaked everyone out.

Unergatives (ungrammaticality expected):
*The resigned CEO started a new company.
*The run dog is exhausted.
*The talked parrot wants a cracker

OK, this is weird.
*The arrived train has malfunctioned.
The recently departed train has malfunctioned.
The recently spread disease is starting to become a huge problem.

So an adverb helps, but not for "Sat" or "appeared".

[bradrn]

For me, all the unaccusatives are grammatical, and all the unergatives are ungrammatical (except for ‘the arrived passengers are all sick’, which is possibly grammatical).

[Linguoboy]

So an adverb helps, but not for "Sat" or "appeared".

I'm 100% in agreement with you except for "sat", which I'd accept but which marks the variety as BE (thus I wouldn't use it except in conscious imitation of a BE-speaker).