Nominal compound acquisition in 10 languages:Psycholinguistic evidence from lexical typology

• Main Authors: Dressler, Wolfgang U., Kilani-Schoch, Marianne, Ketrez, Nihan, Argus, Reili, Basbøll, Hans, Dabašinskiene, Ineta, Johansen Ijäs, Johanna, Kamandulyte-Merfeldiene, Laura, V. Kazakovskaya, Victoria, Kjærbæk, Laila, Korecky-Kröll, Katharina, Laalo, Klaus, Sommer-Lolei, Sabine, Thomadaki, Evangelia, Stephany, Ursula
• Format: Conference Object
• Language: English
• Published: 2017
• Subjects: saami
• Online Access: https://portal.findresearcher.sdu.dk/da/publications/b1d4b0ab-0e34-4f60-8f48-fe8b9742d932; https://findresearcher.sdu.dk/ws/files/140510094/SLE_Abstract.pdf

We compare typologically (cf. [1]) emergence and early development of nominal compounding in longitudinal spontaneous speech corpora of 10 languages (up to age 3;0). We argue that wealth and productivity of compounding, as reflected in child-directed speech (CDS), morphological decomposition, pattern selection, productivity (specifically profitability) [2] of compounds in child speech (CS) of specific languages are better typological predictors of early emergence than general morphological typology, which characterises, to varying degrees, the investigated languages Turkish, Finnish, Saami as agglutinating, Estonian as agglutinating and inflecting, Lithuanian, Russian and Greek as richly inflecting, German, Danish and French as weakly inflecting. Statistical analysis of interaction effects in a Linear Effects Model confirms that compound richness in CDS of the languages investigated is the best predictor (cf. [4]). For this purpose we apply to first-language acquisition studies the theory and methodology of Lexical Typology [3] in quantifying the percentage of nominal compounds among nouns and studying onomasiologically their rivalry with multilexical words (phrases), derivations (particularly suffixations) and simplex words for each language and relating CS to CDS. The percentage of compound noun tokens ranges from about 1% in Lithuanian, Russian and French CS and CDS to 17% in German, whereas the percentage of compound noun lemmas ranges from more than 2% in Russian and 3.8% in French to 38% in German. These frequencies neither correspond to the morphological type of e.g. French and German (both weakly inflecting) nor are they related to the general wealth and productivity of compounds in adult speech found in written genres and described in grammars. However, rather close frequency relations exist between CS and CDS. The main result of our study is the importance of morphological richness in CDS (see above). Moreover, morphosemantic and morphotactic transparency is greater in CS than in adult speech and, as ...