One of the most interesting words in Faroese is mann, the generic pronoun equivalent to English ‘one’. Officially it does not exist (it’s not in the dictionaries and the grammars) because it’s one of those horrible Danish words. Still, speakers use it all the time. An analysis of the use of mann shows the patterns of use are very different from those in Danish. Faroese may have borrowed the word, but it’s doing its own thing — and is very much not turning into Danish.
I’ve long been fascinated by the word mann, especially because it is a generic pronoun that strangely can be used to refer to oneself:
sjálvandi svimjihøllin bleiv bygd eisini tá ið mann var smádrongur
of course the swimming pool was built also when one was a little boy
It’s perfectly clear from the context that the ‘one’ here isn’t just any little boy, it’s the speaker of the utterance. In English it’s very strange to refer to yourself with a generic pronoun (in fact, only the spoof Twitter account of Queen Elizabeth II does this consistently) but in Faroese examples like this are a dime a dozen.
Time, therefore, to look into how Faroese speakers use mann. The starting point is to look at the word in its most common context: as a generic pronoun, a pronoun that refers to no one in particular. Apart from mann, there’s another pronoun that can do this, and that’s the second-person pronoun tú ‘you’. This is exactly like in English, where one/you can use both a generic or a second-person pronoun for this function; and like in Danish, where the options are man and du.
viss mann skal hava parkeringspláss so skal mann vera øgiliga heldigur
if one must have a parking space then one must be extremely lucky
tú fært ikki hondverkaraútbúgving viss tú gongur eitt ár bara
you don’t get a vocational diploma if you only go one year
Here it’s clear that the sentence doesn’t just apply to the speaker or the listener, but to pretty much anyone. Parking is a bitch for everyone, not just for me, and education is a multi-year commitment for everyone, not just for you.
So we have a pronoun mann that’s been borrowed from Danish, and that varies with tú, just like the Danish original man varies with du. The best thing about this is that we know exactly how this variation between man and du works in Danish, because it’s been studied in detail by Torben Juel Jensen and his colleagues at the University of Copenhagen. They looked at all the cases of generic man and du in an enormous collection of recordings of spoken Danish, and found out when speakers are more likely to use one or the other pronoun. Once we find out when speakers of Faroese are more likely to use mann or tú, we can compare this to the results from Danish, the language that mann is supposed to come from and that all Faroese people speak anyway.
The corpus of spoken Faroese that I looked at is much, much smaller than the Danish corpus. I only had just over 550 examples of generically used mann or tú, whereas the Danes had over 30,000. Still, 550 examples (74% mann and 26% tú) is more than enough to play with the statistics, and find out some of the rules for the use of mann and tú.
Women are much more likely to use mann than men are.
Why this should be is a mystery. Often we see that women use more of a feature when it’s the new, incoming variant in a language change in progress, but other than this gender effect, there doesn’t seem to be much evidence of a language change. In fact, older people use more mann than younger people, which would suggest a language change in the other direction. And the gender effect is there whichever age group you look at. (Funnily, the only two people with 100% use of mann are the oldest and the youngest person in the data.)
Interaction really matters.
The most challenging aspect of this variation was the role of the interaction between the two people in a conversation. The speakers were recorded in two interactional settings: an informal conversation with another speaker, and a more formal interview setting with one of two trained interviewers. People used mann more in interviews than in conversations. But there was another factor that threw a spanner in the works here: if the listener in the conversation was included in the referent of the pronoun (so if the stuff about parking spaces and vocational diplomas can reasonably be expected to apply to the listener as well) you’re more likely to use tú. And there were many more of such listener-included cases in conversations than in interviews. This makes sense to me, because in a conversation you’re probably trying to establish common ground with your conversation partner, and one way of doing this is by involving them in your thinking space; in interviews, you can just talk about yourself, and no one really cares about the interviewer. But this does mean that the effect of interactional setting becomes murkier.
With some fancy statistics, you can tease these two things apart. It turns out that in conversations, people tend to use mann a little bit less often than average, regardless of listener-inclusion. But in interviews, listener-inclusion does make a difference: if the listener is included, then people also use mann less often than average (and on a par with the conversations), but if the listener is not included, then mann use shoots up to ‘almost always’. That particular interactional setting clearly is the odd one out.
Nothing else matters.
There’s quite a few other factors that have been found to influence the choice of generic pronoun in other languages. These are mostly linguistic things, like whether the pronoun is used in a conditional construction (if…, then…) or whether they’re talking about past or present states and events. I also looked at how centrally located the speakers’ home town was, because that could give information on the spread of language change. But none of this has any effect at all. It’s gender and interactional setting, and nothing else.
What about the Danes?
Comparing the Faroese data to the Danish data wasn’t as straightforward as it could have been, mostly because the two corpora were so different in set-up, and because the Danish data had been picked apart with a different purpose and with super-fancy statistics. Still, a few things are relatively clear:
Speaker gender is not a thing in Danish, whereas it is in Faroese. The overall pattern of women using more man(n) than men do exists in Danish as well, but it’s not statistically significant.
Listener-inclusion is a bit of a thing in Danish, whereas it is a massive thing in Faroese interviews, but not a thing in Faroese conversations. The Danish data doesn’t distinguish between different settings, so it might well be that if we had Danish interviews and conversations, we would see the same pattern. But as it stands, the two languages don’t quite work in the same way.
Lots of other things are things in Danish, whereas they are not in Faroese. Again, this may be due to the differences in the corpora. The Danish corpus has recordings from the 1980s and the 2000s and can therefore track change in real time, but the Faroese corpus was recorded at one point in time. The Danish study also looks at social class, which is just not an obvious social factor in the Faroe Islands anyway. Other factors do occur in both studies, though: there’s a bit less man(n) in conditional constructions (but this is only significant in Danish) and the influence of location works in opposite ways in the two languages.
So what does this mean?
It’s difficult to say what this means for language change in the Faroe Islands. It’s clear that the pronoun mann, which isn’t officially recognised, is used quite a bit, and it doesn’t seem to be on its way out any time soon. Even though the pronoun originally comes from Danish, and Faroese speakers are all bilingual in Danish as well, they seem to be doing their own thing with the pronoun. They haven’t just copied what the Danes do and tweaked it a bit; they’ve completely re-written the rule book. Not that they’ve done so in a very imaginative way, though: listener-inclusion seems to be the clincher here, and that’s kind of obvious from an interactional perspective.
But Faroese mann is clearly different from Danish man, so maybe we can see it as an entity in its own right now, and stop pretending that it doesn’t exist…
Read the full article here (library subscription or payment needed).
Knooihuizen, Remco. 2015. Convergence in generic pronouns: Language contact and Faroese mann. Acta Linguistica Hafniensia 47(2).
Remco Knooihuizen 