Negation is one of the few grammatical features observed in all languages. While typically thought of as a property of predicates, it can be manifested in a wide range of structural positions associated with verbs (typically, V, I or ̂I or as a verbal adjunct, represented as NEG), but is also observed on other parts of speech (e.g. D/ N, C, P and CONJ) and is sometimes expressed across two or more nodes within c-structure (e.g. Butt et al. 1999, Alsharif & Sadler 2009, Laczkó 2014, Bond 2016, Alruwaili & Sadler 2018).

In the most straightforward cases there is one representation of negation at f- structure, with a binary feature indicating the presence or absence of this value. However, distributional differences between superficially similar negators, and evidence from structures with multiple negative forms within a single clause, suggest that more than one feature may be necessary to account for the syntactic and semantic effects observed in negative contexts. For instance, when a negation scopes over a sub-constituent in c-structure (so-called constituent negation or cneg) which is part of a finite syntactic structure which is also negated (known as eventuality negation or eneg) two representations of negation appear to be required within the same f-structure (Przepiórkowski & Patejuk 2015). The distribution of Negative Concord Items (NCIs), Negative Polarity Items (NPIs) and case-forms licenced by negation also suggests that multiple features must also play an important role in accounting for restrictions on the occurrence of certain forms in antiveridical contexts (Sells 2000, Camilleri & Sadler 2017).