Expressing knowledge through grammar is crucial for understanding how language works and what it is good for. Evidentiality – the linguistic marking of information source – has been the focus of linguistic investigations for quite a while now. Expressing how one knows things – especially in those languages with grammatical evidentiality where one is forced to do so – has distinct bearings on human communication, cognition, categorization of types of knowledge, and societal conventions. Evidentiality lies at the heart of articulating and communicating knowledge. It is a feature of many languages. But perhaps the most fascinating insights come from those languages which have large obligatory evidential systems – including Tariana and its neighbours from north-west Amazonia. Some ‘linguists’ with limited knowledge of the languages of the world and a penchant to work just with translations into some familiar European languages – have obscured the notion and the category of evidentiality. Some have confused it with modality – probability and possibility, others diluted it with other concepts – including ‘truth’, ‘evidence’, and the like. Evidentiality is not about any of those. It stands apart from other categories, and at the same time interrelates with them.
Throughout my work on evidentiality (started in 2003, as shown in the Commentary to this essay), I have been demonstrating its special status in its many guises. Over the years, there have been numerous studies of evidentials, and their correlations with other means of expressing knowledge. The most valuable ones come from the discovery of new systems and the insights into the immense potential a human language has. This essay comes as a logical step in the progression of our understanding of the phenomena involved and documented to date. Its major focus is on the ways in which evidentials are special, and how they correlate with other categories within language, be it synchronically or through language history.
On 22 May 2019, Junwei Bai, then a senior PhD scholar at the Language and Culture Research Centre, gave a talk entitled ‘Egophoricity in Munya?’. The presentation was about the differences between egophoric marking, evidentiality, mirativity and other categories in a previously undescribed Tibeto-Burman language, the topic of Junwei Bai’s PhD (since then successfully approved and awarded a rare summa-cum-laude appreciation). During this talk, it all came together for me – the division of labour between various knowledge-related categories and their interactions (summarised in Diagram 1 in this essay, p. 19). This was the foundation for a shorter version of this essay, originally prepared for a workshop ‘Evidentiality and modality: at the cross-roads of grammar and lexicon’, scheduled to take place at the Université de Montpellier-3 Paul Valéry in June 2020, and then cancelled, as so many events were in that parlous year.
The focus of this essay is the ways in which many grammars of minority languages deal with how one knows things, setting forth a consistent exposure of issues, based on individual grammars and my own fieldwork, and tested through many years of investigation.