Don’t believe in a paradigm that you haven’t manipulated yourself! – Evidentiality, speaker attitude, and admirativity in Ladakhi

Abstract

A speaker may conceptualise and represent a situation from three different 'perspectives': epistemic, evidential, and attitudinal. Languages differ in which of these concepts they profile and how a grammaticalised category may be extended to the other two. Modern Tibetic languages including the Ladakhi dialects are said to have grammaticalised evidentiality. However, their 'evidential' systems differ from the typologically more common systems, in that speaker attitude is co-grammaticalised and knowledge based on perception shares properties with knowledge based on inferences. The starting point for the development of this, as it seems, typologically rather uncommon 'evidential system' was a lexical marker for non-commitment (or admirativity): the auxiliary ḥdug.

Factual vs. evidential? The past tense forms of spoken Khalkha Mongolian

“…"Factuality" might thus indeed be linked to an epistemic value, but more in the eye of the beholder than for the speaker. [[70]] Areally, the evidentiality system of Khalkha Mongolian reflects Tibetan systems such as Ladakhi where assimilated, fully asserted authoritative knowledge expressed by the auxiliary yod contrasts with nonassimilated, seemingly non-authoritative knowledge within which immediate perception is expressed by ḥdug and inference is expressed by special inferential markers, while hearsay is mostly expressed by verba dicendi (Zeisler 2017) -like Khalkha ge-. The major difference is that authority and participation are much less significant in Khalkha, so that -laa can easily be used for events that the assertor undertook just recently and which are not yet well established mentally, and for decisions that the assertor is just taking, while ḥdug is not easily used in contexts that are controlled by the assertor.…”

A metalinguistic analysis of the terminology of evidential categories: experiential, conjecture or deduced?

“…The terms used in the descriptions of Ladakhi are more unusual: 'reportive' (Koshal 1979) is probably intended to mean 'neutrally reported i.e. presented (by the speaker)', rather than verbal report from other people (see Zeisler 2017: 301), and 'set 1' (Zeisler 2018) reflects a division between egophoric-evidential markers and evidentials indicating external sources like sense perception, collectively called 'set 2' markers.…”

Evidentiality

“…It is quite common, however, for languages to mark inferential processes (deducing the situation or event) and reported information (hearing about it from another person) as two (or more) different kinds of indirect experience. These categories, in turn, have their own twists and turns (regarding inference, see, e.g., Barnes 1984, de Haan 2001, Hill 2017, Zeisler 2017. Inferential evidentials often show especially complex and interesting interactions with time values (Fleck 2007;San Roque & Loughnane 2012a,b), relating to the different real-world evidential affordances for inferring that something has happened, as opposed to inferring that it is happening or will happen (in fact, evidential marking is usually more elaborate for past events in general; Aikhenvald 2004, Visser 2015.…”