Ásgrímur Angantýsson (University of Iceland)

Résumé du séminaire

The goal of this talk is to shed light on how individuals change their pronunciation through the lifespan and the extent to which conscious and subconscious language attitudes play a role in explaining such real-time linguistic changes, using the uniquely documented development of local phonological variation in Iceland as a test case.  Selected results from the first data collection phase of the research project Regional pronunciation, attitudes and real-time change (https://pronunciation.hi.is/) will be presented. The presentation focuses on four regional phonological variables:

(1)       North-East
            a. Hard speech, i.e. post-aspiration of /p, t, k/ after long vowels: tapa [tʰaːpʰa] (‘lose’),
            líka, [liːkʰa] (‘like’), bíta, [piːtʰa] (‘bite’) as opposed to lack of such post-aspiration.
            b. Voiced pronunciation (“raddaður framburður”) with voiced sonorants before an
            aspirated stop: hempa, [hɛmpʰa] (‘cassock’), mennta, [mɛntʰa] (‘educate’), hjálpa,
            [çaulpʰa] (‘help’) as opposed to voiceless sonorants in this position.

(2)       South-East
            a. hv-pronunciation in words like hvalur, [xaːlʏr̥] (‘whale’) has a voiceless velar
             fricative [x] in initial position, as opposed to the general kv-pronunciation (“kv-
            framburður”) with [kʰv] in this position.
            b. monophthongal pronunciation (“skaftfellskur einhljóðaframburður”): bogi,
            [pɔːjɪ] (‘bow’), magi, [maːjɪ] (‘stomach’), as opposed to the more common
            diphthongal pronunciation (“tvíhljóðaframburður”).

In the so-called RÍN project in the 1980s (Thráinsson and Árnason 1992), the northern voiced pronunciation and the southern hv-pronunciation showed very clear correlation with age, in the sense that the younger age-groups were less likely to use the variants in question than the older generation, while the northern hard speech and the southern monophthongal pronunciation had relatively equal distribution with respect to age. However, the overall use of the four regional varieties had declined from Guðfinnsson’s (1946) study in the 1940s, to a varying degree, with the hard speech being the least endangered variety.    Our results indicate that both the voiced pronunciation and the hv-pronunciation are fading out, while the monophthongal pronunciation and the hard speech are maintained in their respective core areas. As was expected, the older participants generally maintain their original pronunciation, and this pronunciation does, on average, show much higher scores for the regional variants than those found amongst the youngest age group. It is also noteworthy that the scores appear to be highest amongst those who live in the core areas of the regional variants, while those who live in neighbouring areas or have moved away from the home communities altogether use the variants to a lesser degree.   
            The informants under discussion here, from two different age groups and three different regions (north Iceland, south Iceland and a so-called neutral area), did a pronunciation test, listened to and gave their attitudinal response to recordings of speakers of dialects other than their own, and answered a questionnaire on their attitudes towards their own dialect. With regard to the attitudes, it appears that, in general, participants consider speech that contains northern features to be clearer than speech that does not. In similar fashion, participants appear to find these northern features to be more easily recognizable, at the same time as they seem to be more closely linked to speakers’ self-identity than other regional features.
            As is often the case, the general trends come with some interesting exceptions or deviations that should not be overlooked. It turns out, for instance, that while northern voicing can be maintained or even increased, on the basis of a conscious and attitude-driven decision after moving away from its core area, participants from the south-east tend not to make conscious decisions on whether or not to maintain their original pronunciation variants, even though they may not want to stand out through their use of regional features outside their home area. This may in turn conform nicely to northern speech generally being seen as clearer or more “correct” than speech elsewhere in the country.