Principal investigators: Prof. Sol Lago, Prof. Esther Rinke & Prof. Petra Schulz

A key goal of the CRC is to identify how morpho-syntactic and semantic properties contribute to the makeup of negation—Neg-Plus/Neg-Only hypotheses—and whether this contribution is universal or language-specific. Our project investigates the assumption that cross-linguistic differences in the morphosyntactic expression of negation can lead to differences in how negation is processed and acquired in typologically different languages. By focusing on the aspect of cross-linguistic variation we will contribute to narrowing the hypothesis space, which will bring us closer to the overarching goal of the CRC, namely to decide between the Neg-Plus- and the Neg-Zero hypotheses.

From a psycholinguistic perspective, the question arises of how negation is processed and acquired in languages with different patterns of negation. Specifically, we address the following research questions:

QC.2 How is negation processed and acquired in languages with different morphosyntactic types?

QC.3 What does the processing and acquisition of negation tell us about the grammatical representation and/or interpretation of negation?

Previous processing research on adults has focused on the question of how the mental representation of negated sentences is built. Earlier studies argued for a 2-stage-model (e.g., Clark & Chase 1972, Kaup et al. 2005), with the representation of the affirmative sentence preceding the representation of the factual state of affairs. But recent studies have also provided evidence for an incremental account of negation, according to which the negator can be immediately integrated into an ongoing sentence interpretation (e.g., Tian et al. 2016). Notably, online comprehension studies on negation have often used German and English as the language of investigation, and little attention has been paid to the potential role of language-specific differences. Previous acquisition research on negation has provided evidence that children use negation early, before age 2, but that adult-like understanding is acquired much later, between the ages of 3 and 6. This variation results from the finding that children’s comprehension ability is influenced by various contextual factors including truth/falsity of the negatives (Wason 1972; Wojtecka et al. 2011, 2013). Cross-linguistic differences regarding negator position have to date not been explicitly considered in terms of their role for acquisition pace and developmental patterns.

In our project, we investigate whether cross-linguistic differences regarding the position of negation affect adult comprehension and child acquisition by comparing German and Spanish. German and Spanish provide ideal cross-linguistic test beds for examining the role of the negator position: in Spanish, the surface position of the negator is early in the sentence, before the finite verb. In German, the negator appears late and after the finite verb. Moreover, sentential negation in Spanish is restricted to one, preverbal, position, whereas the sentential negator in German may appear in different surface positions.

Concerning the underlying structural position of the negator, it is disputed whether it equals the surface position (i.e., high NegP in Spanish, low NegP in German, see Laka 1990, Zanuttini 1989) or whether both languages share a low NegP. In the latter case, the differences between the two languages on the surface are attributed to cliticization of the negator to the finite verb and subsequent verb(+neg)-movement in Spanish versus an invariable position in German (Meisel 1997, Zeijlstra 2004). We take the latter assumption of a shared base position as our starting point and we derive specific empirical expectations for the comprehension of negation in adults and children, which we will test in offline and online experiments with both groups.

Comparing adult comprehension of negation in Spanish and German, we address two main research questions. The first question is whether the earlier, fixed surface position of sentential negation in Spanish and its later, more variable surface position in German affect comprehension in adults. Specifically, we ask to what extent cognitive processes (e.g., revision and/or inhibition) play a prominent role in building the mental representation of negated sentences in the two languages. That is, does the earlier surface position of the negator in Spanish facilitate its comprehension compared to comprehension of the negator in German? Alternatively, processing may need to involve the low base position of negation in German and in Spanish, resulting in no difference between speakers of German and Spanish or even a processing advantage of German speakers.

The second question is whether differences in earliness and variability of the surface position of negation modulate the time course of acquisition in German compared to Spanish children. Based on the conviction that adult processing and child language acquisition are closely related (e.g., Friedmann et al 2009), we expect that processing difficulty in adult speakers of Spanish (or German) corresponds to particular developmental delays in target-like understanding of the relevant structures by children acquiring Spanish (or German). Spanish speaking children may show an advantage over German-speaking children because the earlier surface position of the negator in Spanish facilitate its comprehension, leading to mastery at a younger age. Alternatively, if the low base position of negation, shared by both languages, is crucially involved in discovering and building the correct representation of negated sentences, Spanish children should have no advantage over children acquiring German or should even be disadvantaged because of the dissociation between the base and surface position of the negator.

In the first phase of the CRC, we will investigate these research questions by investigating monolingual German- and Spanish-speaking adults and children. In the subsequent phases, we plan to extend our cross-linguistic perspective in two dimensions. In phase 2, we wish to substantiate our findings and broaden our typological perspective by including languages that seem to pattern with Spanish (e.g., Portuguese, Polish, Basque) or German (e.g., Dutch) or that exhibit yet a different pattern (e.g., Finnish, French). In phase 3, we intend to apply our cross-linguistic perspective to different acquisition types in order to uncover potential cross-linguistic effects and their direction in the comprehension of negation. Depending on the outcomes of phases 1 and 2, these may include heritage bilinguals of Spanish in Germany or L1 German–L2 Spanish and L1 Spanish–L2 German adults and children.