The term ‘Negative Sensitive Items’ (NSIs) describes a category of words and expressions that interact with negation in various ways. This interaction takes place in at least three different ways. First, negation is (usually) required in simple declarative sentences containing an NSI. Second, certain NSIs have a negative interpretation when used as fragment answers (the Fragment Answer Test), despite the absence of overt negation. Third, certain negative markers are in complementary distribution with certain NSIs. This chapter introduces the key contrasts between two major NSI categories and shows the explanatory adequacy of a multi-locus analysis of negation to account for the mutual exclusivity between certain NSIs on the one hand and bipartite negation in JA and the negative marker maa in SA on the other hand. The chapter starts by defining the major NSI types along with their lexical categories. It then lays out the theoretical background of NSI licensing as this is necessary to develop an analysis for NSIs based on the multi-locus analysis of negation. This chapter extends this analysis to the Qatari and SA data giving further evidence that the different positions of negation have different effects on the licensing of Negative Polarity Items (NPIs). This chapter also discusses the interaction between higher negation and determiner NCIs (wala-NP) as well as the interaction between maa and NPIs in SA.

Types of NSIs and their lexical categories

At least two different categories of words fall under the term NSI. They differ from each other with respect to their sensitivity to negation. NPIs form the first category. These words always require negation in simple declarative sentences and may be in complementary distribution with certain negative markers, but cannot stand alone as fragment answers without negation.

The NPIs ʔaħadun ‘one’ and ʔayyu ʔaħadin ‘anyone’ in Standard Arabic (SA), ħada ‘one’ and ʔayy ħada in Jordanian Arabic (JA), and ʕumr ‘ever’ in JA all require a negative marker (see (1)a, (2)a, and (3)a) and cannot PASS the Fragment Answer Test (see (1)b, (2)b, and (3)b). The NPI ʕumr is in complementary distribution with the enclitic negative marker -š in JA (Alqassas 2015), in Moroccan Arabic (Ouhalla 2002; Benmamoun 2006), and in Egyptian Arabic (Soltan 2012; Brustad 2000).