Kinship nouns in American Norwegian: split possession in a heritage language

Kari Kinn (University of Oslo)

Recent studies have observed that Norwegian, as spoken in Norway (EurN), has a relatively rich system of split possession whereby certain syntactic constructions are reserved for certain kinship nouns denoting close family relations (Lødrup 2014, Johannessen et al. 2014).[1] This sets Norwegian apart from English, which does not have split possession to the same extent. The present paper presents novel data from American Norwegian (AmNo), a heritage language spoken by bilingual (mostly 3rd generation) Norwegian immigrants in North America. The main research question is to which extent split possession, a lexically restricted phenomenon is retained in AmNo; investigations into this topic can shed new light on syntactic change in situations of language contact and reduced L1 input. I will focus mainly on two possessive constructions reserved for certain kinship nouns in EurN: indefinite kinship nouns with i) a postposed possessive pronoun, and ii) a postposed PP. Both constructions are special in that the kinship noun may appear without the definiteness suffix, which is required for other nouns. Cf. 1–2:

(1)
mor mi
mother my
‘my mother’

(2)
mor til Mari
mother to Mari
‘Mari’s mother’

Drawing on CANS[2] and additional speech data, I will argue that split possession is not only retained but productively extended to new kinship nouns in AmNo.[3] Example 3a involves a noun borrowed from English; 3b involves a kinship noun that does not license the special constructions in EurN.[4]

(3)
a. ...var gift med auntie mi
...was married with auntie my (westby_WI_01gm)

b. det var hos e # kusine til han
it was with ee cousin to him
’It was at X’s cousin’s place’ (harmony_MN_02gk)

Other kinship nouns that, somewhat surprisingly, seem to license the special constructions in AmNo are e.g. barnebarn ‘grandchild’, kone ‘wife’, and jente ‘girlfriend’.[5] The absence of the definiteness suffix in these cases does not seem to indicate a general decline of the definiteness category in AmNo. However, kinship nouns stand out in terms of appearing in their bare form in certain contexts additional to the constructions discussed above; like in 4:

(4)
Når jeg får brev fra firmenning…
when I get letter from third cousin...
‘When I get a letter from my third cousin...’ (sunburg_MN_06gm)

This supports the idea that kinship nouns are treated as a natural class in AmNo, and that this class may be affected by the syntactic change. The retention of split possession may be related to frequency; we can reasonably assume that the relevant kinship nouns and constructions were robustly attested in the input when the AmNo speakers acquired their Norwegian. The extension to new contexts highlights that change in heritage languages is not always best understood in terms of loss; changes can be innovative in the same way as in non-heritage languages (Yager et al. 2015). Motivation for the change under discussion here might be found in the concept of cross-linguistic overcorrection (Kupisch 2014): bilingual speakers sometimes overstress what is different between their two languages, and as split possession is characteristic of Norwegian, but not English, this property was extended.

References

  • Johannessen, J. B., Julien, M., and Lødrup, H. (2014). Preposisjoner og eierskapsrelasjoner: et menneskesentrert hierarki. In Hagen, K. and Johannessen, J. B., editors, Språk i Norge og nabolanda., pages 65–97. Novus for-lag.
  • Julien, M. (2005). Nominal Phrases from a Scandinavian Perspective. John Benjamins.
  • Kupisch, T. (2014). Adjective placement in simultaneous bilinguals (German–Italian) and the concept of cross-linguistic overcorrection. Bilingualism: Language and Cognition, 17(1):222–233.
  • Lødrup, H. (2014). Split possession and the syntax of kinship nouns in Norwegian. Journal of Comparative Germanic Linguistics, 17(35–57).
  • Stolz, T., Kettler, S., Stroh, C., and Urdze, A. (2008). Split possession: an area-linguistic study of the alienability correlation and related phenomena in the languages of Europe. John Benjamins.
  • Westergaard, M. and Andersen, M. (2015). Word order variation in Norwegian possessive constructions: bilingual acquisition and attrition. In Johannessen, J. B. and Salmons, J. C., editors, Germanic Heritage Languages in North America. Acquisition, attrition and change, pages 21–45. John Benjamins.
  • Yager, L., Hellmold, N., Joo, H.-A., Putnam, M. T., Rossi, E., Stafford, C., and Salmons, J. C. (2015). New structural patterns in moribund grammar: case marking in Heritage German. Frontiers in Psychology, 6:1–9.

[1] To some extent, other relational nouns also exhibit special properties; they will not be discussed here.

[2] http://tekstlab.uio.no/glossa/html/?corpus=amerikanorsk

[3] Westergaard and Andersen (2015) note this for construction 1 but do not discuss split possession more generally.

[4] E.g. Julien (2005:92); I have also queried the Nordic Dialect Corpus.

[5] The wide range of kinship nouns involved in split possession makes AmNo resemble Icelandic (Stolz et al. 2008:119). One might ask whether AmNo exhibits diachronic continuity since Old Norse instead of extension. However, the Norwegian dialect literature and available EurN data do not appear to corroborate such a claim.