This paper describes a specific non-standard negation strategy in Iquito, a moribund Zaparoan language spoken in northern Peruvian Amazonia. This strategy is used in finite subordinate clauses (namely adverbial dependent clauses and relative clauses), as well as information questions, and it utilizes two negative markers: a negative particle which is also found in standard negation, and a verbal affix which does not function as a negator in any other context. Using existing typological characterizations of subordinate clause negation, we see that Iquito exhibits the following attested traits: it uses the standard negator in a different position, it also utilizes a distinct negator, and it employs more negators in the subordinate clause than in the main clause. But unlike the languages presented in the literature, Iquito utilizes these parameters simultaneously. Additionally, the position of the standard negator changes within the subordinate clause, depending on the reality status of the clause. Using Iquito as a case study, I propose a set of parameters for comparing subordinate clause and interrogative negation strategies to standard negation strategies, which include the type of negator used, its position, the overall number of negators, the potential for interaction with other grammatical categories, such as reality status, and the resulting word order of the clause. This set of parameters expands the initial typological characterizations of subordinate clause negation strategies.
Iquito, a Zaparoan language of Peruvian Amazonia, marks a binary distinction between realis and irrealis clauses solely by means of a word order alternation. Realis clauses exhibit a construction in which no element intervenes between the subject and verb, while in irrealis clauses a phrasal constituent appears between the subject and verb. No free or bound morphology otherwise indicates whether an Iquito clause is realis or irrealis. Based on these facts and partially similar phenomena in other languages, this article argues that typologies of inflectional exponence should be expanded to include word order as an inflectional formative.