Edited by Ad Foolen, Helen de Hoop and Gijs Mulder
[Human Cognitive Processing 61] 2018
► pp. 173–198
First, this chapter discusses some problems with a ‘narrow’ view of evidentiality that imposes structural requirements, both in its analysis of evidentiality as a whole and in its analysis of the reportive vs. quotative distinction in particular. It argues instead for a ‘broad’, exclusively functional view where evidentiality is seen as relating to justification for knowledge. On this basis, reportive evidentials are distinguished from quotatives, which are non-evidential forms that attribute information to a source. Second, the chapter substantiates its claims with a case study of specific uses of German sollen ‘shall’ that are analyzed as non-quotative, non-epistemic modal, reportive evidential constructions. Finally, sollen is related to ‘referral’, a newly-coined concept that captures the functional relation between quotatives and reportives.