Polar questions in Dutch Sign Language (NGT): A production experiment.

Polar questions play a fundamental role in communication and have been investigated across a wide range of spoken languages, from which we know that polar question forms may vary considerably across varying contexts. Much less is known about polar question forms in different contexts in sign languages. This paper concentrates on one particular sign language, namely Dutch Sign Language (NGT). It investigates (i) which polar question forms exist in NGT, (ii) in which types of context these different forms are used, and (iii) what the semantic contribution is of specific elements that may occur in polar questions. We present a production experiment in which two contextual factors are manipulated through role play between the participant and two confederates: (i) the prior expectations of the person asking the question with respect to the truth of the proposition that the question is about, and (ii) the evidence available in the immediate context of utterance with respect to the truth of this proposition. We identify a broad range of polar question forms, differing in terms of polarity marking (positive, negative, null), the presence or absence of question tags, and use of non-manual markers. We observe that each question form has a distinct distribution across the different contexts of use under consideration. With regard to non-manuals, a notable finding is that, contrary to expectation given the previous literature on polar questions in sign languages, raised eyebrows are not consistently used in the polar questions in our dataset; in fact, brow lowering is also rather common. This leads us to suggest that a head or body forward position - not raised eyebrows - is the main polar question marker in NGT. We suggest that raised and lowered eyebrows mark a broad range of other functions that may be relevant in (non-canonical) polar questions.