skip to content

Interactional foundations of language universals and linguistic diversity

Language universals are tendencies that are shared across all or most of the world’s languages. Here are two examples:

  • Most languages show a preference for placing the agent before the patient in a sentence (Dryer 2013) .
  • Suffixes are preferred over prefixes in the languages of the world (Dryer 2013b)

There are various approaches to the question of how such universals can be explained. An approach that is central to our research works with the hypothesis that such universals are at least in part shaped by the infrastructure for social interaction shared by all humans – the human “interaction engine” (Levinson 2006). Nikolaus P. Himmelmann discusses some of aspects of this hypothesis in his Abralin ao vivo talk from January 2022.

Nikolaus Himmelmann

Despite universal tendencies like the above, the languages of the world show an astonishing diversity of structures and usage patterns (Evans & Levinson 2009). For instance, although all languages have linguistic means for giving feedback in a conversation, the use of these means and the means themselves exhibit remarkable differences across languages. Consider the examples below, which illustrate feedback in German and in Yurakaré (isolate, Bolivia). In the examples, a speaker requests a reconfirmation, i.e. a feedback signal that confirms what was just said. Speakers of German prefer conventionalized fixed formats for requesting such feedback, e.g., Echt? ‘really’. Speakers of Yurakaré in contrast, favor repeating part of the previous utterance (Gipper, König & Weber, in press).

German (Germanic, Germany)

D: ich wollte ja jetzt unbedingt karten für die drei fragezeichen haben
‘I was very much hoping to get tickets for “The three question marks”.’
D: die kommen nach münster also die originalsprecher
‘They’re coming to Münster, that is the actors.’
B: echt
‘Really?’
D: ja
‘Yes.’
D: aber es gibt schon für münster gibt’s gar nichts mehr alles ausverkauft
‘But for Münster, there are no tickets left. It’s already sold out.’
B: echt
‘Really?’
D: ja
‘Yes.’

Conventionalized request for reconfirmation

Confirmation


Yurakaré (isolate, Bolivia)

P: tiotta binta latij na winanij së
‘My hips hurt while I was walking.’
A: winanimti
‘When you were walking.’
P: otte
‘Yes.’
P: achamati mijankyaninaj werta lijadayucham wertay kuti mijumpadre
‘“I‘ll give you the panniers, you can carry it in the panniers,” I said then to your compadre.’
A: kutam tijumpadre latiji
‘You said then to my compadre.’
P: aja
‘Uhu.’

Repeat request for reconfirmation

Confirmation


In our work, we are exploring inter alia which feedback means are used cross-linguistically, which of these means co-occur, which do not, and what the general preference patterns within and across languages are. Ultimately we wish to explain, i.e. find the reasons for the preference patterns. To approach these questions, we investigate inventories of feedback signals in different languages, such as German, German Sign language, Spanish, and Yurakaré, among others.