In: J. Hoeksema & D. Gilbers (Eds). Black Book. A Festschrift in honor of Frans Zwarts. University of Groningen, (2014). 1 Experimental investigations of licensing environments for NPIs in English Jennifer Spenader, 1 Frank Richter 2 and Janina Radó 2 Abstract: In an experimental investigation of NPIs and their licensing conditions in German, Richter & Radó (2013) showed that Weak Licensors can partially license Strong NPIs. In two experiments we investigate these results for English. In Experiment 1 we empirically identify 16 Strong and 16 Weak NPIs by examining their acceptability when licensed by sentential negation and weak adverb licensors and find results similar to German. In Experiment 2 we compare the licensing abilities of two different types of conditionals, normal, causal conditionals and rhetorical conditionals, such as `If you win, I’ll eat my hat’. We did not find statistically significant evidence for our prediction that rhetorical conditionals might be able to license strong NPIs better than normal conditionals, but we discuss possible changes in the experimental design that might lead to different results. Experiment 2 thus serves as a pilot study for future investigations. 1. Introduction The defining feature of Negative Polarity Items (NPIs) is their restriction to certain environments. Sentential negation has the ability to license all NPIs, hence the name, but the actual licensing properties are much more complex entailment properties. The discovery that the entailment properties of semantic operators have linguistic effects is one of the major triumphs of formal semantics, and explains why research on NPIs is a fundamental part of both formal semantics and linguistics proper. This research area has a rich history and an active present, with new questions and extended theories continually joining the discussion (see e.g. Eckhardt, 2012 and Csipak et al. 2013). The seminal work of Franz Zwarts (1981) with his proposal for embedded licensing contexts 1 Corresponding author. j.spenader@gmail.com, Institute for Artificial Intelligence, University of Groningen. 2 Institut für England- & Amerikastudien. Goethe Universität Frankfurt am Main frankmrichter@gmail.com, rado@em.uni-frankfurt.de. The second and third author’s main contributions were extensive advice on the experimental design and general discussion.