Jim Bisso’s The Morphology of Peevology Facebook post says (I’ve added links):
Language Jones does a good job of explaining unaccusative verbs (as opposed to unergative ones). What’s that? Per the article in Wikipedia: “In linguistics, an unaccusative verb is an intransitive verb whose grammatical subject is not a semantic agent. In other words, the subject does not actively initiate, or is not actively responsible for, the action expressed by the verb. An unaccusative verb’s subject is semantically similar to the direct object of a transitive verb or to the subject of a verb in the passive voice.”
For example, in English “the tree fell”, “the window broke”, &c. It is related to the reason why some verbs in, e.g., French or German use “to be” as the auxiliary verb in past tenses, instead of “to have”: “je suis tombé” (I have fallen) versus “j’ai travaillé” (I have worked). (It was first described, in 1978, by David Perlmutter.)
Unaccusative verbs have been mentioned a few times at LH, at length in this 2012 comment by Wimbrel:
In linguistics this distinction is encapsulated in the contrast between unergative and unaccusative verbs, i.e., intransitive verbs whose subject is, from a semantic perspective, the doer (agent) or the experiencer (patient) of the action. In Romance and Germanic languages (like French, Italian, German, and Dutch) that have two different auxiliaries for forming the perfective past (preterite), unergative verbs take “to have” (avere/avoir/haben/etc.) as the auxiliary and unaccusative verbs take “to be” (essere/être/sein/etc.). Vestiges of the unergative/unaccusative distinction seem to have survived as late as Early Modern English (hence “the Lord is come”). Radford’s Syntax: A Minimalist Introduction gives some examples from Shakespeare, like “How chance thou art returned so soon?” (Comedy of Errors, I.ii) “She is fallen into a pit of ink.” (Much Ado About Nothing, IV.i)
The Wikipedia article provides the history of the concept and gives examples in English and other languages, notably Russian: “Unlike the subtle evidence for unaccusatives in English, Russian provides strong tests to determine unaccusativity.” I’m hoping making a separate post of it will help the concept (which was invented just as I was dropping out of grad school) stick in my head. Oh, and that Language Jones video is indeed good; it ends up showing how such verbs are reflected in brain scans, aphasia, and child speech.
The wiki article has this curious phrasing: “Further linguistic studies have shown that intransitive verbs will identify as either unergative or unaccusative, determined by the language. A recent study proposed by James Baker in 2019 suggests that intransitive verbs not only identify as either unaccusative or unergative, but with” etc etc. Curious because in current usage intransitive “identify”* is generally a verb implying human-like consciousness and volition on the part of the subject who “identifies” as such-and-such – an “unergative” one rather than an “unaccusative” one in this jargon, but one requiring a specific sort of agent.
Perhaps this is an ESLism, or bad machine translation from a non-English source?
*And this seems like it ought to be the fairly new sense glossed by wiktionary as “(reflexive, with as) To claim an identity; to describe oneself as a member of a group; to assert the use of a particular term to describe oneself.”
I think that the real difficulty with “unaccusative” and “unergative” is not so much the concepts themselves but the fact that the names are so badly chosen. Geoffrey Pullum (no less) is supposed to be to blame.
the fact that the names are so badly chosen
Yes, it’s as if those who use them have gotten mentally unstuck on the unside of Alice’s unmirror. Unaccusatives for unobjects or unsubjects, maybe ? It’s like Pig Latin minus the metathesis.
“Semantic agent” is contrasted with “grammatical subject”. What’s the diff between “semantic agent” and plain “agent” ? What happened to the “subject” ? Are taxpayer dollars funding this agency ?
Kusaal, like many West African languages, can “passivise” transitive verbs just by deleting the agent and putting the object in the subject position, but the resulting construction is limited in its aspectual possibilities, unlike the subject of verbs like (intransitive) “heat up” or “fall.” So “unaccusative” is exactly wrong: typical non-agentive subjects in Kusaal don’t behave like accusatives at all.
In a lot of languages, passives work in a whole other way from English passives, and lumping passives in with all other constructions with non-agentive subjects is bound to lead to tears before bedtime.
I feel vaguely like I may have first heard the label “unaccusative” way back in ’86 when I took a quite enjoyable seminar* called something like “Grammatical Relations” where we talked quite a bit about ergativity and whatnot. But I’m not sure what I took away from it and now I’m wondering if I’m instead remembering being introduced to the concept of the “antipassive,” which is perhaps likewise unhappily named.
*Maybe me and one other undergrad, three or four grad students, and the professor?
Yes, the names are terrible; they came up in the Difficult Words thread a couple of years ago when Y said:
teeming masses
#
„Keep, ancient lands, your storied pomp!“ cries she
With silent lips. „Give me your tired, your poor,
Your huddled masses yearning to breathe free,
The wretched refuse of your teeming shore.
Send these, the homeless, tempest-tossed to me:
I lift my lamp beside the golden door.“
#
Nowadays more likely Border Patrol searchlights.
I think “teeming masses” is a somewhat different cliche-phrase than “huddled masses” from a “teeming shore.” “Teeming masses” seems from a quick glance into the google books corpus to have been used as early as the 1840’s by authors as distinguished as W.E. Gladstone and James Fenimore Cooper.
Shore thing. I can’t imagine how to teem and huddle at the same time. Football teams huddle, though, and I think the same goes for rugby (which I know only from Heartstopper, where “giant moronic rugby lads” are deprecated).
It’s what masses do. They teem. Huddling, though, is climate-dependent. In many parts of the world, it’s quite usual to teem without huddling at all.
“Teeming ant-hill” = “wimmelnder Ameisenhaufen“. An expression that may come in handy one day.
The whole unergative-unaccusative-antiergative-antiaccusative-antipassive sequence reminds me quite belatedly that when I was young I would read old novels in which people’s parlors sometimes contained objects (of furniture? or something?) mysteriously called antimacassars, and I never knew what they were but was never quite curious enough to go look it up.* So now I have consulted wikipedia and am now rather underwhelmed that so prosaic an object should have had so recondite a name.
But maybe now the word can be repurposed as a bit of morphosyntactic jargon? Seems a pity to let its pentasyllabicity fall into desuetude.
*I probably at the time still had some not-yet-deceased great-aunts who would have known, had I been unable to find a suitable reference book.
Seems a pity to let its pentasyllabicity fall into desuetude.
Antianathema and antianaphora have never risen from desuetude. Both have hexasyllabicity, which has seven.
Now I too know what an antimacassar is. High time.
Unaccusatives in LH comments, here and here.
I knew what an antimacassar is, but had somehow managed never to realise why they’re actually called that. Duh.
Now I, too, am disappointed.
anti-macassar(n.)
also antimacassar, 1848, from anti- + macassar oil, supposedly imported from the district of Macassar on the Indonesian island of Sulawesi, which was commercially advertised from 1809 as a men’s hair tonic “infallible in promoting an abundant growth and in maintaining the early hue and lustre of the HAIR to the extent of human life” [1830]. The cloth was laid to protect chair and sofa fabric from men leaning their oily heads back against it.
source: https://www.etymonline.com/word/anti-macassar#13536
I knew the ye-olde-Dutch-East-Indies place name contained within antimacassar, because I was the sort of boy who browsed old atlases,* but couldn’t figure out the implied enmity.
*By the time I was in my atlas-browsing years, a dutifully up-to-date one would have used the new toponym Ujung Pandang that had been imposed by the Suharto regime. I am pleased to be belatedly advised by wikipedia that that never caught on with the locals and the post-Suharto government approved reversion to the old name in the late Nineties, although the current local spelling is Makassar, understandably following Dutch rather than English-or-Portuguese orthographic custom. Reportedly the doubled s is non-standard in Bahasa Indonesia orthography but they’ve let it slide.
“I declare to my antimacassar if you took up a straw from the bloody floor and if you said to Bloom: Look at, Bloom. Do you see that straw? That’s a straw. Declare to my aunt he’d talk about it for an hour so he would and talk steady.”
As we do here, right? “We are all Leopold Bloom now …”
I was told at age nine what and why an antimacassar is, by my Mum.
… not so much the concepts themselves but the fact that the names are so badly chosen
I join the chorus of assent. And I have my own preferred scheme for these, which is ignored by all (to say nothing of sundry).
i believe the implication of a pro-macassar camp* is false, but i would like to offer the idea to any alternative/speculative history writers present. i can imagine it being the occasion of much huddling (if somewhat less teeming) in steampunk drawing-rooms.
.
* whose maximalist position would presumably be inverted, with the head resting on the unprotected seat-cushions.
Football teams huddle, though, and I think the same goes for rugby …
Nope. Rugby teams (or a subset of each) ‘scrum’. Any players can form a ‘ruck’ or a ‘maul’. (The precise terminology depends on which code they’re playing, and a large number of recondite rules about how the bodies got into that particular pile.) I guess to the uninitiated, ‘huddle’ might be a cover-all term. Dangerously close to ‘cuddle’, though, which it certainly is not — if that’s what Stu was hoping. (“Hands in the ruck” are groping for nothing other than the ball.)
An antimacassar on their armchair will keep your hosts unaccusative.
I would like to be thought unaccusative, but I wouldn’t like to give the impression that it was just because I lacked agency.
Errghhh!
You should try to be a bit more unerrghhhative.
Hey, that was my joke!
The unergative/unaccusative thing is maddening. Once I learned the alternate terms “agent intransitive” and “patient intransitive” I never went back.
I think I first encountered antimacassar in “The Adventure of the Cardboard Box,” some years before higher brow Victorian novels. I had to go check, but Miss Cushing was working on one in her lap.
Antimacassar seems a bit of an odd usage. I think of technological words with anti- as terms for things that shield you from some undesirable externalities, like diseases or insects or the weather, but not for what protects you from generally desirable things. Thus antifreeze, because who likes freezing? but you wouldn’t call sunscreen “anti-sun”. Likewise, macassar oil was a desirable thing, just not on the furniture. So why the negativity?
I’m trying to think of unaccusative verbs in English that can be nominalized with -er. Keeper is not quite it: it can mean the subject, as in “keeper of the flame” and an implicit object, as in “it’s a keeper”, but the latter means ‘something worth keeping (by someone)’, not ‘something that doesn’t spoil’, which would be a true unaccusative meaning. But I feel like there must be some genuine examples out there.
Learners of Spanish who know some French and German get puzzled that Spanish always uses haber as auxiliary, even with verbs of motion. And Spanish is very good at idioms that seem peculiar to L1 English speakers, so se me cayó for I dropped it. This is all I suppose something to do with the (in)transitive/(un)accusative nexus, which the academic linguistics are good at pointing out in specific cases, but always fail to come up with general rules to aid the learner (end of peeve).
“the fact that the names are so badly chosen”
You can say that again. They were invented long before I went to grad school, and were well established in linguistic scholarship, and it still took me years of pretty regular exposure before I got them to stick for more than a few days at a time. If it’s been a while since I encountered them, I still usually have to take 30 seconds to think it through and get it straight in my mind.
‘something that doesn’t spoil’
That meaning is out there, but you generally have to say ‘good keeper’ (e.g. here) – you can also have poor keepers. I’m not sure if that still works for you – it does seem to be about its ability to be kept rather than its worth.
Huddles are nothing to do with scrums etc – they’re when you gather together in a standing group to decide what to do next, or just to encourage each other. They do seem to be restricted in rugby, though, in one of the Six Nations matches the referee was constantly breaking up illicit huddles!
I’m trying to think of unaccusative verbs in English that can be nominalized with -er.
English does that regularly with verbs that take experiencer (sic) subjects, even though they aren’t agents, but in the relevant senses those verbs aren’t intransitive, so they don’t qualify as “unaccusative.”
“Sleeper”, though. (You can sleep a good sleep, but that doesn’t make the verb properly transitive.) “Dreamer”, too.
Kusaal can make “agent nouns” (for want of a better word) from pretty much any verb that can be used in a direct command, including even the copula verb an “be something” (which is necessarily transitive, though, so not “unaccusative.” There is no agent noun from the intransitive verb bɛ “be somewhere, exist”, though, even though that verb can also be used in direct commands.) Including gbisid “sleeper” and zaans “dreamer.” It also happily makes agent nouns from stance verbs like zin’i “be sitting”, but those are construed as intransitive agentives in Kusaal, unlike many other languages.
You can even make an agent noun from zi’ “not know”, meaning “ignorant person.” However, even though that verb is often used without an expressed object (M zi’i “I don’t know”), I think it could be taken as belonging to the category of Kusaal actually-compulsorily-transitive verbs, where if there is no expressed object the meaning is actually always anaphoric: “I don’t know (sc. the answer to what you just asked.)”
With the verb kpɛɛm “be older than”, you probably could claim that the noun kpɛɛm “elder” actually was the corresponding agent noun, but again, the verb is actually compulsorily transitive, and so not “unaccusative.”
So you probably could say that Kusaal forms agent nouns from (nearly) all verbs, unless they are unaccusative. But I think this is more a question of pragmatics than anything else. The most clearcut unaccusative verbs in Kusaal are predicative-adjectival verbs like gim “be short”, and it’s hard to see the use of an agent noun “short-be-er” when you could just say “short person.”
I don’t think you can have *lut “faller” from lu “fall”, but you could use that same stem adjectivally, as in bʋnlutir “thing connected with falling”, and Kusaal “agent nouns” basically are deverbal adjectives of that type inflected as members of the “human” gender/noun-class-pairing. So there’s nothing wrong with a form lut in purely morphological terms. More that there is no call for it.
I have encountered contention; some say antimacassars can and should be crocheted, in which case they might be deemed doilies, while modern YouTubers insist they be made of fabric that matches the chair, and can be tucked in so you might not even know it has an over-layer.
When this false layer is washed for re-use, which is the whole point, the colors will fade and the subterfuge will be betrayed. For my part, I think antimacassars should be educational; videre licit a cloth map of Slovenia, or a comparison of the Elder and Younger Futharks. A conversation opener for us introverts.
“Remainer.”
@de
Can you sleep a good sleep? I can only sleep (for) [TIME], where [TIME] is only accusative in the sense that it is the object of the (omitted) preposition. I am not even comfortable with dreaming a good dream, it sounds like poetry or a slogan. In fact, I cannot offer a context where “I Xed a good X” is not marked, for any X (apart from X = the F word, where we may have a genuine unaccusative)
EDIT: for me, “I sang a good song” works.
This kind of figura etymologica is a rhetorical and poetic device, so I guess it is stylistically marked? Or does it feel ungrammatical / unidiomatic to you?
Isn’t “sleep” (as a noun) usually a mass noun not a count noun? So any use of it with an article (indefinite or definite) is going to feel marked.
“I Xed a good X” is a cognate object
For “X a good X,” consider also Yeats’ question “Who can tell the dancer from the dance?” (It’s a good “poetic” question because, come on, you can easily tell the difference unless you’re really really stoned.) You could easily swap in “singer from the song” or “dreamer from the dream” for much the same effect but if you try to swap in “sleeper from the sleep” it suddenly seems much weirder-to-nonsensical. Which I think is some evidence that “sleep” whether considered as verb or noun isn’t really quite the same sort of thing as those others.
@hans
Mostly style, but certain verbs like dream and sleep (also e.g., jump so “jump a fence” is missing a preposition over, unlike “climb a tree/mountain”, which for me have a slightly different meaning to “climbed up a tree/mountain”) do not take an object for me, because somehow implying change of state and not action on anything.
The “cognate object” thing mollymooly mentions seems kind of vacuous as a set. Sentences of the form “He experienced an experience” involve a notably different semantic relationship between agent and object than sentences of the form “He produced a product.”
“Who can tell the dancer from the dance?”
Actually “How can we know the dancer from the dance?”
And strange to tell, among that Earthen Lot
Some could articulate, while others not:
And suddenly one more impatient cried—
“Who is the Potter, pray, and who the Pot?”
Isn’t “sleep” (as a noun) usually a mass noun not a count noun?
CGEL includes sleep among the two dozen examples to make the case that nouns with both count and non-count uses are not, in fact, exceptional.
I think perhaps the former is a bit less common in American than British. Or maybe it’s in a different register.
Salishan languages actually have a specific grammatical category of “control”, marking the difference between e.g. “fall” and “drop down deliberately.” (Explained in e.g. section 20.45 of Thompson and Thompson’s The Thompson Language.)
Jen — “good keeper” works for me, at least for lack of anything else so far.
“Sleeper”, as in a railway car, is from transitive “sleep”, as in “it sleeps two people”.
The reference (in the 2012 comment by Wimbrel blockquoted in the OP) to Radford’s _Syntax: A Minimalist Introduction_ made me realize (if not on first reading …) that this must be the same Radford who had earlier written _Transformational Syntax: A Student’s Guide to Chomsky’s Extended Standard Theory_, which I was required to purchase a copy of for the “Syntax I” class I took in 1985. The more recent work presumably conceded that the content of the prior work was largely outmoded nonsense, but presumably with no refunds offered to prior generations of students who had purchased it.
Wikipedia advises that Prof. Radford died this past December, aged 79.
“Only n more sleeps till Christmas” is a usage I had not noticed until social media. It may originate from a December 1915 quote ascribed by a Christian missionary to the aptly named Indian Chief Turkey Legs.
agent is a semantic label, object a syntactic one. A syntactic object may be a semantic patient, instrument, product, etc. As I understand it, the point about a cognate object is that it allows a usually intransitive verb to become transitive. There is a difference between “he died tragically” and “he died a tragic death” — and between “he died” and “he died a death”.
@mollymooly: I amend my prior remark to substitute “A and O” for “agent and object,” per the “Dixon (1994)” terminology referenced here. https://en.wikipedia.org/wiki/Morphosyntactic_alignment The capital letters were apparently introduced as an abstraction precisely to sidestep difficulties about semantic roles v. syntactic ones.
If the point of the “cognate object” concept is to specifically exclude perfectly normal transitive sentences like “He painted a painting,” then it seems like yet another badly-named syntactic category/label. The potential synonym “internal accusative” offered by the wiki article seems less-bad.
I encourage anyone who clicks through mollymooly’s link about Chief Turkey Legs to scroll up to the immediately-preceding article, because whatever you may think of its substance its title (“Adequate Mission Architecture”) is one for the ages. “What do we want? Adequacy! When do we want it? As soon as reasonably convenient!” My Great-great-uncle Jack’s name appeared with some frequency back then in the pages of this particular churchly periodical, but apparently not in this issue.
@Y: Folder? Tumbler (drinking glass)? Cracker (either crackly food or British pyrotechnics)? Cutter (boat)? Twister (tornado)? Or am I misunderstanding?
In the Army we’d calculate the length of time till our release as “five days and a wake-up,” a more cheerful thought than “six days.”
“Thompson and Thompson’s The Thompson Language”
Remarkably, this is a real book.
dancer from the dance; potter/pot, dreamer/dream… it’s all* butterfly/zhuangzi to me.
in yiddish, we’ve got the structure “[מען, אַאַז] שלאָפֿן שלאָפֿט זיך” “shlofn shloft zikh [men, etc]”, which is more-or-less an abstracting/depersonalizing phrasing, that usually gets translated with things like “sleeping happens”**, or just as a continuous/habitual verb. i’m not gonna read back through the thread enough to try to label it an un-something or anti-anything, but it seems like one of them things.
.
* except if it’s in The Lathe of Heaven, He Who Shapes, or other exemplars of that subgenre, in which case it’s butterflies all the way down (getting scarier as they go).
** or “sleeping takes care of itself”, in contrast to prayer and study, in a famous ChaBaD nign.
@JF:
Folder — maybe? I don’t know that it has any living transparent connection to fold. Is a folder something that folds (i.e. “suffers folding”)? Or is it something that (en)folds papers?
Tumbler — again, it’s a fossilized construction. It’s attested in the OED since the 1600s, but its derivation is not clear to me. All the likely senses of tumble are intransitive.
Cracker — That works, I think!
Cutter — that’s a garden-variety subject of transitive cut.
Twister — I think that works, too.
I guess what I was thinking of were examples with the intransitive meaning of labile verbs, whereas unaccusatives are a type of intransitive verbs, not necessarily with a transitive counterpart.
Sparkler.
But as with “cracker”, and maybe even “sleeper”, I think you could make some sort of case that in these forms the verb is really being regarded as “agentive” within the meaning of the act.
This is a bit like Kusaal zin’id “sitter”, where in fact there is other evidence for an agential reading of the subject of zin’i “be sitting”: in Kusaal, being sitting is something that you are doing, not a state that you’re in (it’s “I’m sitting” rather than “je suis assis.”)
Similarly with sin “silent person”: the syntax of the verb sin “be silent” shows that “being silent” in Kusaal is regarded as something you do, rather than a way that you are.
(Both zin’i and sin can have progressive/continuous readings in appropriate syntactic contexts, which is something that verbs expressing states of being can’t do. This correlates neatly with the two verbs allowing for derivational processes not shared by state verbs.)
And as for “sleep”, it is well-known that colourless green ideas sleep furiously.
Of course, if verbs can tweak their meanings at will between less- and more-agentive readings (and they can) it suggests that any restrictions on which verbs can form which derivatives are more to do with semantics than syntax.
The Trinidadian Creole of Naipaul’s early books is full of expressions like “Is love he really love she. He looking for she.”
I agree with all here. I have never been able to get a grasp of the “unaccusative” and “unergative”. Still can’t figure ot what they mean.
“The house sold”.
“These hotcakes are selling well”.
Are these un…tives of some sort?
@Bathrobe: the intransitive sense of “to sell” in “The house sold” is unaccusative, although I suspect most of the historical focus has been on verbs where it’s not as easy to recast the sentence using the exact same verb transitively as e.g. “They sold the house.”
I guess what I was thinking of were examples with the intransitive meaning of labile verbs
There are hardly any agent nouns from such verbs in Kusaal, but I did record one, from naae “finish”, which can make naad “perseverant person.”
The absence of others may in fact be due to a morphological constraint: such verbs are mostly formed with a derivational suffix *g, and it seems only to be possible to add the agent-noun-deriving suffix after that if the *g has been previously absorbed by an internal sandhi change (as in fact it has been in naae.)
I have seen “alternation” as an umbrella term covering both changing the valency of verbs and permuting the arguments. The latter sometimes triggers peeves, eg evacuate location of people vs evacuate people from location.
permuting the arguments
Bantu languages typically have something called Locative Inversion
http://www.jzeller.de/pdf/Loc%20Inv.pdf
whereby a locative expression becomes the subject, the verb agrees with the locative subject via the wonders of the Bantu noun-class system (which generally has at least one locative “gender”), and the logical subject just follows the verb, twiddling its thumbs and looking vaguely helpful.
Some Bantu languages can do this with adverbials other than locatives as well.
Kusaal verbs don’t agree with anything, so Kusaal can’t do anything quite so dramatic, but on the other hand “therefore” can be the subject* of a clause:
dinzugʋ kɛ ka ti sʋnya kpɛ’ɛŋ
therefore.LINKER cause and our hearts strengthen
“So we are comforted.”
(The structure here is that dinzug “therefore” is the subject of the verb kɛ “let, cause”; the LINKER here is working as a focus particle, in the standard construction for focusing a clause subject.)
* Agentive? Non-agentive? These are deep waters …
Sticker!
That’s what I was looking for. Either an adhesive label or a thorny plant bit.
@Y: Note that the adhesive-label sort of “sticker” has two possible relationships to the verb. Either a) it’s the thingie that intransitively sticks to e.g. a car’s windshield; or b) it’s the thingie that someone transitively stuck onto the windshield. You would think “stickee” might be more precise for the latter concept, but “He stuck the sticker onto the windshield” seems like a perfectly idiomatic sentence.
You can’t scare me, I’m sticking to the union.
You can’t scare me, I’m sticking to the union.
I stick at naught.
I once saw a sticker that said
Jesus Saves
At the First National
Why is it that I have to add “once” when I want to be indefinite about a past experience? “I once saw” is not the same as “I saw”. It’s as if “I saw” by itself implied recency, and/or were inviting the hearer to ask “when?” Depends on the timeline of whatever is being discussed. I don’t feel there’s an invite of that kind with “I once saw”, which sorta says “don’t bother to ask exactly when”. Or am I imagining things ?
@Stu: I think that’s right. Using the preterite/Imperfekt often comes with some explicit statement of when the event occurred: I saw X [this morning / yesterday / last week /etc.] Failing to specify a time seems to carry some implicature of the fairly recent past. “I saw _Star Wars_” w/o some further specification (or clarity provided by context) seems misleading if what is true is that I saw it during its original first-release run in 1977. OTOH “I’ve seen _Star Wars_” is perfectly consistent with having seen it either last week or in 1977 or at some point in between. “Once I saw _Star Wars_” negates the “comparatively recently” implicature that saying nothing would yield, but also sort of implies that “exactly when” is not particularly relevant so the addressee is not really being invited to ask for more detail on when. I.e. the time it “specifies” is more or less “once upon a time” or “a while back” or something like that.
So maybe the answer to your question is that “once” enables you to be indefinite because it is one of the vaguest available ways to meet the grammar’s demand to include some specification of definiteness as to time.
In Kusaal, most dynamic verbs in the indicative perfective aspect have past reference by default. Outside specific contexts (notably within past narratives), if you just use the perfective, the meaning is like an English present perfect, viz past, but at no particular time, and thus potentially having present relevance as well:
Saa niya.
sky rain
“It has rained.” (At some point. Maybe I just mean “it’s not a desert here”, but the fact that I’m saying it at all is more likely to mean that we’re talking about why the grass is wet now, or why the washing has not dried on the clothesline.)
If you want something like an English simple past, you have to put a specific past-tense marker before the verb:
Saa pa’ niya.
“It rained.” (Earlier today.)
Saa sa niya.
“It rained.” (Yesterday.)
Saa daa ni.
“It rained.” (Two days ago or more.)
Saa da niya.
“It rained.” (Quite a long time ago.)
So that kind of past needs to be explicitly marked as past.
[Kusaal additionally has a specifically not-current-relevant past, a kind of antiperfect, but I’m keeping it simple here …]
More on this in some non-SAE languages:
In Siouan languages, agentive and non-agentive verbs belong to different conjugations: thus, in Lakota, “come”, “seek”, “hold” conjugate differently from “die”; but unlike Kusaal, in which stance verbs behave as agentives, in Lakota, “sit” etc conjugate like “die.” Lakota is the the “je suis assis” camp.
Muskogean verbs are sensitive to this distinction, too, though it seems that only Koasati and Alabama preserve the proto-Muskogean fully-distinct conjugation of agentive and non-agentive verbs. But in Koasati it’s not simply lexical: the same verb can be conjugated either way, for e.g. “I slept” (agentive) versus “I fell asleep” (non-agentive.) I previously mentioned how Koasati cardinal numbers, which are verbs, can conjugate either way, with a corresponding meaning difference. Kimball says that the morphology generally correlates well with the degree of control exercised by the subject, but not invariably: “sneeze”, for example, is agentive …
It’s interesting to see that while, yes, this is a very common distinction cross-linguistically, it actually plays out very differently in different languages. It may be (almost) a lexical feature of individual verbs, in which case, which verbs go in which category still varies a lot between languages; or it may belong more to the morphology, as in the Salishan languages or in Koasati.
Just thinking about the absence of agent nouns from labile verbs used in intransitive senses in Kusaal (forms that would mean things like “person who gets lost” (from bɔdig “lose/get lost”) or “person who gets hot” (from tʋlig “heat/get get hot.”)
I think this is actually due to the semantic content “human” of the a/ba noun-class pairing (“gender”), which all agent nouns belong to. All nouns in that “gender” in Kusaal refer to people, and I think it’s that fact that tips forms made from labile verbs like naae “finish/come to an end” into having the transitive sense: humans are the archetypal semantic agents, after all.
When the same stem is used adjectivally, in other “genders”, you do get the intransitive senses appearing: bʋntʋligir “thing to do with heating”, which could be either a heater or something that just gets hot.
Mooré, unlike Kusaal, does have some nouns in the a/ba class pairing which don’t refer to people, and they too always have agentive senses, but I think they are all modern calques, like wilgda “demonstrative” (the grammatical category) from wilgi “show.” (I don’t think the word for “demonstrative pronoun” can be reconstructed to proto-Western Oti-Volta.)
(I don’t think the word for “demonstrative pronoun” can be reconstructed to proto-Western Oti-Volta.)
Are you implying they didn’t have grammar back then?
Maybe they had grammar but just didn’t want to talk about it. Like me most of the time. idk it makes me feel watched.
Grammar is an imperialist colonialist tool for suppression of the aspirations of the oppressed.
I thought the aspirations were suppressed by phonology.
Indeed.
When we immanentize the eschaton, Cockneys will be free to pronounce their aitches once again. Oh yes.
And French h will always be aspiré.
Ça ira!
Workers of the world unite! You have nothing to lose but your psilosis!
I find that psilosis also means a medical condition in which you lose your hair. For the phenomenon in question here, ‘alitosis would be more specific.
Ah, like il m’est tombé and es ist mir heruntergefallen, where, instead of the agent, I’m the (negative) beneficiary? “It dropped on me”, with “on me” like “you switched on me”…
Unfortunately, English “It fell on me” is already taken.*
Some people seem to make a big deal of English “I forgot” versus Spanish “Se me olvidó”, but English has “It slipped my mind” and Spanish has “Olvidé.” Is the point that “I forgot” is more common in English than “Olvidé” is in Spanish?
*On an episode of Milligna (or Your Favourite Spike):
https://en.wikipedia.org/wiki/Milligna_(or_Your_Favourite_Spike)
I don’t get the joke…
It means nothing. Weird Milligan “humor”..
https://blogger.googleusercontent.com/img/b/R29vZ2xl/AVvXsEiP8yzngcSLK5xjYaQXYRKvcUH-PDLTXg7C4mQIOJbITgLUXAc_k-tJvKSJREQCfXAX0cDKrnq70ja8n1ajP1-SmTlc7tUh3XVgoCX5ts6VsFTSyk9oQObvVNHhsxJdOKfoIVb3WBnT8W75/s1600/rCrumb+Mr+Natural+3.jpg
@Hans: I thought it was a wacky reference to circumcision, but this comment is the only thing known to Google that makes that suggestion. The Forward said it’s reminiscent of the saying “Shver tsu zayn a yid,” which possibly nobody here needs translated as “It’s hard to be a Jew.” Otherwise people seem to agree with Stu (not necessarily including the quotation marks) and David.
Linguistics anomaly: The Forward article puts an asterisk before the Yiddish line. I don’t think they mean it’s unattested. Maybe that’s their signal for Yiddish?
@Jerry Friedman: at least in the dialects I’m fluent in, transitive olvidar is dated or literary. Most people will have it passively, but I don’t think I know anyone who actively uses it in spontaneous speech.
The colloquial equivalent is olvidarse de.
Thank you, Alon. Just out of curiosity, does Me olvidé de la cita sound less evasive about agency (as they say at Language Log) than Se me olvidó la cita?
Just noticed this, reading through Jacques Nicole’s grammar of Nawdm (section 3.3.1):
The Nawdm formation he’s talking about corresponds to the Kusaal/WOV agent noun formation with the derivational suffix -d. In Kusaal, the same stem also appears as s deverbal adjective, and can then have active or passive meaning, as in e.g. sumdʋgʋda “cooked groundnuts (summa)”, from dʋg “cook”; but when that same stem is used as an independent noun in the “human gender” it’s never passive. But evidently this is not always so in Nawdm. The suffix corresponds not only to English -er, but also (though much less often) to English -ee.
On the one hand, this supports my view that the WOV agent-noun stem and deverbal adjective stem are fundamentally the same formation; on the other, I think it also supports what I was arguing above, that the specifically “agentive” sense of what I have always called deverbal “agent” nouns in WOV is actually a sort of side-effect of them being marked as human-reference by their noun-class membership.
here’s the bit from the Forward:
In one typical exchange reflecting the old Yiddish expression “It’s hard to be a Jew” (*Shver tsu zayn a yid), Milligan would ask Bluthal, “Are you Jewish?” and receive the reply: “No, a tree fell on me!”
the asterisk looks to me like someone was trying to format it as a footnote in a medium that doesn’t include them. and i for one don’t see any reflection of “shver tsu zayn a yid”. damn near anything (but most likely a really dumb swollen-nose gag) would seem far closer to what’s happening there.
Didn’t think of that. I’ll stick with don’t mean sheeit. It covers all bases.
I love me a universally apposite dismissal.
On nasi, I commend Crumb’s admonition: “You can pick your friends, and you can pick your nose, but you can’t pick your friends’ noses”.
I just noticed multivalence in “pick” there, because plastic surgery. How did I miss that all these years ??
@rozele: Thanks, an attempt at a footnote makes sense.