Final conjunctions
« previous post | next post »
In response to my posting about the language of three imprisoned children in Linz, Austria, several people have commented on the report that the children finished all their sentences with the word "but" (which I took to mean German aber). Two noted that many German speakers use sentence-final conjunctions (oder 'or', but also aber) as discourse particles, and one supplied a possible parallel from colloquial Italian (with ma 'but'). Then I recalled a discussion on the newsgroup sci.lang in 2006 about sentence-final but in varieties of English and in Hawaiian Pidgin. No doubt there are parallels in other languages; the semantic development is not surprising (more on this below).
So it's likely that the children were just using a feature of colloquial German speech, which was then over-reported by an observer and treated as strange and new by journalists. Certainly we've seen many such cases here on Language Log.
Steve Harris noted sentence-final oder as a tag inviting a (confirmatory or non-confirmatory) response, and Fred added aber as another tag. There's probably literature on the phenomenon. It's pretty much guaranteed that these sentence-final conjunctions will have a variety of discourse functions, arising from the many different reasons people have for suspending a coordinate sentence after the coordinator or appending a coordinator as an afterthought following a sentence.
What starts as a strategic use of suspension — to indicate that you're pausing but will have more to say, to indicate that you're waiting for a response from your interlocutor(s), to indicate that you're unsure of what to say next, to indicate that there is an alternative (for or and its equivalents in other languages) or a contrast (for but and its equivalents in other languages) that your interlocutors can supply for themselves, and so on — can come to serve as a conventional signal of discourse function. And what starts as the appending of an element that could have occurred earlier in the sentence but didn't — in particular, a contrastive conjunction (like but) that wasn't produced sentence-initially — can come to serve as a conventional sentence-final signal of that element's discourse function. In either case, the conjunction can end up serving as a discourse particle.
English sentence-final but (reported for Scottish, Australian, and New Zealand English in the sci.lang discussion in November 2006) apparently arose in several different ways, to judge from the uses reported for it. For Scottish, Sylvie Hancil (posting as "sylh") identified the following uses:
contrastive meaning ('though')
intensive meaning ('really')
in most of the cases, a particle used for interactive reasons to show the other speaker he/she can speak.
(Hancil has since given a conference paper on the grammaticalization of but.) Some of her examples from British corpora:
I went and looked in the dustbin to see if he was back in there but.
He would have stayed and while I did it but.
Er, in those days women didn't work, but.
it is classified as being part of Glasgow, but.
I find these hard to interpret out of context.
John Atkinson reported on Australian:
… sentence final "but" is common in some varieties and registers in Australian. Usually it means the same thing as it would if used at the beginning of the same sentence — i.e., roughly the same as "however". Just like final "however", in some circumstances it has a pause (comma) before it, sometimes not. For some speakers, who make a lot of use of it, it seems to have very little real meaning other than as a sign of the end of the sentence — just like sentence final "eh" in the speech of rural Queenslanders, and, I've been told, some Canadians.
Note that this a report of impressions, not the result of a systematic study. Like all such reports, it has to be viewed skeptically. But it can serve as a starting point for research. (There is, by the way, considerable literature on "Canadian eh".)
Finally, Hawaiian Pidgin, about which Marc Adler posted:
I don't know if they have a corpus for Hawaiian pidgin (if they don't, just hang out in Waianae for a while), but this kind of sentence-final usage is very common there. I'm guessing that it comes from the Japanese component of Pidgin, because that's exactly how they use these constructions in Japanese.
In Pidgin "He's going but" (with no pause but a drop in tone between "going" and "but") means the same as "But he's going." The latter is a marked form in Pidgin, because it sounds very "haole" (=the way white people talk).
So: not just in German, but in English too.
Ryan Denzer-King said,
May 3, 2008 @ 2:44 pm
I very often append "so…" to the end of my sentences, usually signifying a following clause which can be easily inferred by my addressee. E.g.:
"How have you been?"
"Well, I'm trying to finish up a paper, so… [I've been really busy]."
I do the same thing with "but," though not as often. I use a lot of discourse markers in my everyday speech, so I'm not sure if these usages are somewhat standard or if they're idiosyncratic. I'd like to do some sort of systematic study, but….
panne said,
May 3, 2008 @ 3:00 pm
In Norwegian, there is an idiomatic phrase "men men" ("but but"), which means something like "oh well" or "that's just the way it is". Many people (myself included) use it if they want to signal that a particular topic of discussion has come to an end. Something like "oh well, nothing to to about that, and nothing more to say either".
Jeremy Hawker said,
May 3, 2008 @ 3:07 pm
Norwegians do the "or?" at the end of the sentence too:
kommer du, eller?
I think of it as, "are you coming, or what?", and as a result I've started to use that construction in English (esp. with my daughter).
Alex Price said,
May 3, 2008 @ 5:02 pm
It sounds to me like sentence-final but (or the equivalent in German or Norwegian) and the Canadian "eh" are similar to, if not exactly like, the Japanese particle "ne". A quick search on the Web reveals that there is a blog post from 2005 on this topic, "Canadian eh' and Japanese 'ne'", on a certain Language Log: http://158.130.17.5/~myl/languagelog/archives/002118.html
Breffni said,
May 3, 2008 @ 5:08 pm
Sentence-final unstressed "but" occurs in working-class Dublin English too. It's at least roughly equivalent to "though", much like German sentence-final "aber". It gets a mention in R. Hickey, 2004, Legacies of Colonial English, p.610 (available in Google Books), though Hickey doesn't limit it to Dublin.
Michael Roberts said,
May 3, 2008 @ 5:14 pm
Sentence-final "aber" in German is more or less equivalent to sentence-final "though" in English. That seems pretty normal for English-speaking kids.
Breffni said,
May 3, 2008 @ 5:19 pm
Correction – that should be Hickey (ed.) 2004, though he is responsible for the passage I mentioned.
pasunejen said,
May 3, 2008 @ 5:44 pm
Like Ryan above, I frequently append "but" or "so" (more often so) to the end of my sentences. Take for example the question, "Are you coming tonight?"
To reply in the negative, I could answer:
(a) I was thinking about it, but I've been really busy, so.
(b) I was thinking about it, but.
In the positive:
(a) I thought I was going to be busy but my schedule's cleared up, so.
I can't think of a way to make "but" work for a positive answer. It could be done, but it would sound kind of awkward.
pasunejen said,
May 3, 2008 @ 5:45 pm
Should be "affirmative" above, not "positive."
Jeremy Hawker said,
May 3, 2008 @ 5:50 pm
Germans sometimes use "Oder?" (i.e. or) as a complete sentence, meaning, "Don't you agree?" Some speakers of English similarly put a "no?" on the end of a sentence. It's as if they were adding an Inspector Clouseau-like, "Non?".
Arnold Zwicky said,
May 3, 2008 @ 6:11 pm
Just a reminder: sentence-final discourse particles and mood markers can arise from many different sources, by many different routes. Sentence-final negatives and question markers, in particular, have many sources. But I was posting specifically about ways *conjunctions* can end up as discourse particles.
About the Belfast connection: not surprising, given other connections between Scottish and (Northern) Irish varieties of English.
I could speculate that Scottish/Irish English was the source of, or at least a contributor to, antipodal final "but", but I don't know squat about the history.
Marivic said,
May 3, 2008 @ 7:21 pm
I've been transcribing some sociolinguistic interviews, and I'm seeing this all over the place with 'but', 'so', and 'or'. I was wondering if it had something to do with my field partner, the interviewees, and I all being in our 20s, but I guess it's more widespread than I thought. I didn't assume it was region because I'm not from the same area as they are, central Ohio (although my dad's from Akron, so I do say 'needs washed'). Some examples from one of the interviewees:
They do trade schools and end up going straight to work . so.
and then like the city schools are always considered like the inner city, and everyone thinks that's like kind of ghetto, or – I don't think it is but.
(asked about travelling and studying abroad)
And then, um, Canada, a couple of times, just to go to Niagra Falls and Windsor and you know, just for fun. So, never anything like studying, but.
Um, but there's a carrousel and we have a pretty big library down there, and there's a few restaurants and bars but . Um, I know they've talked about doing a renovation project,
The spaced out periods like this . indicate a pause, longer than a comma. In the sound file you can hear that he's not just trailing off when he says 'but' at the end like that, and it's clear that I'm not interrupting his speech in these in any way.
David Marjanović said,
May 3, 2008 @ 7:58 pm
Linz? I thought Amstetten?
Being from Linz, however, I can confirm that sentence-final aber is no more common there than elsewhere, and I can confirm the equivalence to English sentence-final "though". I'd be surprised if that were different in Amstetten, which is, like, 50 km east of Linz.
David Marjanović said,
May 3, 2008 @ 8:00 pm
When sentence-final or constituting an entire sentence, oder is the same as an English question tag, isn't it.
Leslie said,
May 3, 2008 @ 8:03 pm
Czech does the same thing. You can end a sentence with že (that), which is short for the tags že jo or že ne.
Ty jsi z Prahy, že (jo)?– You're from Prague, right/aren't you?
Now that I think about it I'm not sure if že ne can be shortened. I moved away 5 years ago. Some native speaker can come along and help out.
Maureen said,
May 3, 2008 @ 8:11 pm
I'd like to know what tone of voice is used.
Also, when I (as an Ohioan on the other side of the state) use this construction, it tends to be accompanied by shrugging, sighing, making a clicking noise….
David Marjanović said,
May 3, 2008 @ 8:18 pm
Oops, you aren't talking about the Amstetten case in the first place… sorry, I hadn't seen that post yet.
David Eddyshaw said,
May 3, 2008 @ 10:40 pm
It's a very usual way of asking questions in Hausa:
Kaa ji koo? (You-perfective hear/understand or?)
and other languages of West Africa, e.g. Kusaal
Fu wum ya koo? (Also "Do you understand?", with koo loaned from Hausa).
I've heard the same construction in Bukina Faso French:
Tu comprends ou?
john riemann soong said,
May 4, 2008 @ 4:55 am
"I’d like to know what tone of voice is used."
In Singlish, "what" often becomes a particle of assertion; it does ask a rhetorical question "I don't see a point to your statementt?" but its tone is very markedly different from say, a normal interrogative.
(where ! denotes a particle)
"!Aiyah, I don't get why Ah Lian has to eat all the congee !one, !hor."
"She's sick with the flu !what. Let her eat."
Martyn Cornell said,
May 4, 2008 @ 5:50 am
As a south of England/London speaker, I had seen end-of-sentence "but" as strongly typical of "Geordie" or north-east of England English, but not of Scottish or any dialect of Irish. The phenomenon is described on this website about Geordie English, which offers the following examples:
Note also the use of "might could"!
john riemann soong said,
May 4, 2008 @ 5:59 am
"I’ve heard the same construction in Bukina Faso French:
Tu comprends ou?"
Ah that reminds me! It seems that French almost has it naturally anyway, like, "donne-le-lui alors!"
Felix said,
May 4, 2008 @ 8:03 am
Using "or" as a question marker seems common. There is for example the Basque question particle "al" from the word for "or": "ala".
Japanese has the question particle "ka", and the word for "or" is also "ka".
Papua New Guinea English also has this, as in "Did he come or?".
Peter said,
May 4, 2008 @ 11:00 am
Anecdotally, as an Australian, I would place the usage of a final "but" more commonly among Queenslanders than among other Australians. As the majority of British settlers to Queensland in its first century of European settlement were Protestant rather than Catholic (ie, from England and Scotland, rather than from Ireland), this usage may have been carried from Scotland.
The meaning is usually in the sense of "however", presenting an alternative argument for, or a counter-argument to, to some claim just made. It is not usually expressed with an intention of concluding discussion, in my opinion. In other words, its use is indicative that the speaker is joined with the hearer in a joint (and sincerely-joint) dialogue to decide some matter, rather than the speaker trying to persuade the hearer to endorse some conclusion the speaker already holds — ie, it is indicative of an Inquiry or a Deliberation dialog, rather than a Persuasion or a Negotiation dialog, in the Walton & Krabbe typology of human dialogs.
Kate said,
May 4, 2008 @ 5:45 pm
As an American who's been living in Hamburg for the past few years, I can confirm that Michael Roberts has got it: "Das nervt mich aber" is the same thing as "That annoys me though."
David, I suspect that this use of "aber" is more common in the southern parts of the German-speaking world generally. Using the small sample of "people I know", it seems to me that a couple of friends from around Karlsruhe and Freiburg use it more than people from northern Germany. A friend of mine from Hungary who lived in Bavaria for a few years also picked up on it, whereas I don't think I've ever integrated it into my German vocabulary.
Alex, "aber" and "oder" aren't really filler words in German like Canadian "eh" is. That honor goes to a syllable that sounds like "nuh" or "na", which is used exactly like "eh":
"Das ist deine Schwester, na"
"That's your sister, eh"
btw I love the comment function on here!
Ingrid Jakobsen said,
May 5, 2008 @ 6:35 am
Another Australian, and I agree final "but" is more common in Queensland, and it is often used to introduce conflicting information without being argumentative or setting up conflict. It's a very natural and comfortable part of my language. It bothers some people, but.
Peter said,
May 5, 2008 @ 7:53 am
Ingrid — Your statement about the final "but" in Australia being used to introduce conflicting information without being argumentative supports my reference to the Walton & Krabbe dialog typology — Inquiry and Deliberation dialogs are generally viewed by argumentation theorists as more co-operative dialogs than other types (or at least, they begin that way).
Kate G said,
May 5, 2008 @ 8:34 am
For plenty of examples of sentence-final though (exaggerated for comic effect) search for "Am I Bovvered?". The characters in these sketches often exchange long volleys of conversation in which each "turn" is one or two words followed by though ("Shame, though." "Innit, though?" "Yeah, though" etc).
David Marjanović said,
May 8, 2008 @ 8:15 pm
I didn't mention that, when moved towards the end of the sentence, it's not intrinsically word-final in German. Rather, it comes… hmm… second behind the verb or something: das nervt mich aber — das würde mich aber wundern "that'd surprise me, though".
David Marjanović said,
May 8, 2008 @ 8:18 pm
Does this mean "hey, look, she's… like… sick with the flu"? If so, it's, like, identical to French sentence-final quoi, quoi.