Indology

International Workshop April 25-26, 2024

Ego-Evidentiality and the right(s) to know (better)

Wilhelmstraße 21 (Bonatzbau), Schulungsraum B004

 

Short meeting description  –  Programm  –  Workshop report and comments

Short meeting description

The modern Tibetic languages are known to have developed quite a particular type of ‘evidential’ marking, their basic principles having been described for quite a few of them. One of the key features is the subjective involvement of the epistemic origo (the speaker in statements, the addressee in questions, and the original speaker in reported speech) in the events relayed. The ‘system’ is thus also known under the key terms of ‘egophoricity’ and ‘conjunct/disjunct’, both concepts often mistaken for a somewhat weird syntactic person category (ego vs. non-ego).

However, at a closer look, the ‘system’ is extremely flexible, allowing, in principle, most if not all forms for all persons, albeit in different frequencies and for different motivations. It further does not only deal with the source of information (firsthand vs. second-hand) or the access channels (self-centred knowledge, perception, and inferences), but also or even predominantly with the subjective assessment of the situation and/or socio-pragmatic factors. These factors appear to be related to a speaker’s rights to treat a particular piece of knowledge as belonging to his or her ‘territory of information’; this also means that speaker-hearer (a)symmetries may play a crucial role. Apart from ‘epistemic rights’, other key words may be ‘empathy hierarchy’ and ‘engagement’.

The workshop aims at discussing the ‘unsystematic’ aspects of ‘ego’-evidentiality or participatory knowledge marking.

Program

Thursday, 25.04.2024 – Tibetic languages

09:30-10:00
Address of welcome and introduction
10:00-11:30
Nicola Tournadre & Dickey Tsang Tsering Wangdue

The grammatical expressions of access to personal knowledge and personal experience. An illustration of the flexibility of the Tibetic evidential systems.

 

break

 
11:45-12:30
Hiroyuki Suzuki,

Towards shaping the egophoric category in Tibetic langua­ges: Suggestions from the factual evidential expression emerging in Japanese

12:30-13:15
Rtamgrin Lhamo

Egophoricity of Minyag

 
lunch break
 
15:15-16:00
Wang Jiahong

Distribution of egophoricity in Golog: An investigation of flexibility and inflexibility

16:00-16:45
Juha Yiliniemi

Flexibility of personal and neutral forms in Denjongke

 
break
 
17:00-18:00
Discussion
 

Friday, 26.04.2024 – Tibetic and beyond

09:45-10:00
Introduction to the second part
10:00-11:30
Ilana Mushin

Managing knowledge asymmetry in grammar and interaction

 
break
 
11:45-12:30
Henrik Bergqvist

Revisiting the origo: a view from the Andes

12:30-13:15
Christian Huber

Egophoricity, evidentiality and modality in Shumcho/ Humcho

 
lunch break
 
15:15-16:00
Zoe Tribur, Sangsrgyas Tshering & Rtamdrin Lhamo

A preliminary account of “non-canonical” occurrences of epistemic markers in spontaneous speech data of Amdo Tibetan

16:00-16:45
Bettina Zeisler

A paradigm of pragmatic flexibility: the case of the Ladakhi dialects

 
break
 
17:00-18:00
Discussion

 

Workshop report and comments

Report and comments on the workshop

Egophoric-evidentiality and the right(s) to know (better)
Tübingen, April 25-26, 2024.

The two days of the workshop saw very lively debates, touching terminology, methodologies, and the resulting analyses plus, of course, also interesting data from different Tibetic languages and beyond. The details of these data can be found in the individual presentations (https://uni-tuebingen.de/fakultaeten/philosophische-fakultaet/fachbereiche/asien-orient-wissenschaften/indologie/personen/bettina-zeisler/international-workshop-april-25-26-2024-ego-evidentiality-and-the-rights-to-know-better/). With this report, I should like to summarise the, in my view, most important points. These concern: 1. Terminological issues, 2. The data, 3. Analyses and disputes, and 4. Further perspectives. I should also like to add some further comments in italics.

1. Terminological issues

The notion of egophoric marking and the so-called anticipation rule, first de­scribed for a couple of Tibetic languages by Nicolas Tournadre (1994), have been reformulated in the crosslinguistic literature in a way that does not suit the Tibetic languages. ‘Egophoricity’ is now typically interpreted as a syn­­tactic category in the sense that only the egophoric marker can be used for the first person subject in statements and the second person subject in questions (cf. Floyd et al. 2018). However, in the Tibetic languages, there is no fixed correlation between the egophoric marker and the epistemic origo (the first person subject in statements or the second person in questions). As Nicolas Tournadre emphasised in his presentation, and as several other presenta­tions showed, all kinds of evidential markers can be used for the epistemic origo as subject. By contrast, egophoric markers can also be used for non-origo subjects in particular contexts. – Whether one should call the corre­sponding marker ‘egophoric’ is another question.

Given this inbuilt flexibility, egophoric marking in Tibetic languages cannot be conceived as a however definable person category in terms of ‘egophoric­ity’. The related concept of conjunct-disjunct marking, as originally proposed by Hale (1980) for Newari and taken up by DeLancey (1990) for Standard Spoken Tibetan, is likewise unsuitable for the Tibetic languages, as long as it is conceived as syntactic category. – One might, however, talk about a con­junct perspective of identification or involvement, associated with egophoric marking, and a disjunct perspective of non-identification or non-involvement, the latter allowing different evidential or epistemic markers.

The reference to statements and questions alone could likewise be to narrow: assessments and requests, e.g., may have their own effects, as suggested by Ilana Mushin. Apart from this, questions have different purposes: rhetorical questions do not necessitate a perspective shift from the speaker to the ad­dressee, as mentioned by Bettina Zeisler. Nicolas Tournadre further showed that tag questions, depending on the verb semantics and context, may or may not allow a perspective shift.

Ilana Mushin further highlighted that from the point of conversation analy­sis, it would not make sense to talk of ‘standard’ and ‘non-standard’ usages or ‘deviations’ from a paradigm. One should further not to talk of ‘upgrad­ing’ and ‘downgrading’ a speaker’s stance, as proposed by Bettina Zeisler. A speaker would typically use the contextually appropriate form. Bettina Zeisler, however, insisted that the non-prototypical, lesser used, or marked usage of the forms in contrast to their prototypical, more frequent use, or unmarked usage needs some illustrative description along a scale, if only metaphorically.

Ilana Mushin further pointed to the fact that languages with grammatical evidentiality will also have non-grammaticalised expressions of evidentiality, while Henrik Bergqvist added that the question of epistemic access and/or authority cannot be treated like temporal reference and tense.

Henrik Bergqvist, taking up the notion of the epistemic origo pointed to the fact that this notion originally only represents one side, either the speaker or the addressee in questions. Similarly, when talking about information source or access rather than of knowledge, the approach is one-sided – and the in­formation flow always goes only in one direction, from the origo to the non-origo. –  Knowledge, on the other hand, could be represented on both sides. Like Ilana Mushin, Henrik Bergqvist argued that speech acts always hap­pen in a dynamic dialogical setting, and that the participants need to negoti­ate their epistemic authority. In this interaction, one has to differentiate be­tween how knowledge is acquired and how it is claimed – or presented.

Juha Yliniemi argued for a negative definition of what he calls a ‘neutral’ marker (the Denjongke counterpart of factual red in Standard Spoken Ti­betan): it would not indicate integrated knowledge or foregrounding or emo­tional involvement, and it would also not imply sensory access. It turned out, that this marker need not necessarily be described only in negative terms, but may well have functions for which a positive description is possible, such as backgrounding or dissociating. Nevertheless, Juha Yliniemi would still think that such terms are only negatively defined.

It should be noted, however, that the term ‘neutral’ may imply that a corre­sponding marker can cover all the positively defined functions, which is not the case with this particular marker in the Tibetic languages.

Zoe Tribur pointed to the fact that in Chinese linguistics, the egophoric markers are associated with ‘subjectiveness’. As a consequence, both the sensory and the factual markers are associated with ‘objectiveness’. This was to some extent problematised in the discussion.

While the egophoric markers are associated with the personal, and thus subjective, perspective of the epistemic origo, the notion of ‘objectiveness’ brings in a quality of higher reliability, which is problematic, particularly for the sensory markers.

Nevertheless, in quite a similar manner Hiroyuki Suzuki described that the Japanese factual construction in naru is used “to indicate that the statements are objective”, which “helps to prevent any potential influence from subjec­tive or authoritative opinions”.

2. The data

All presentations demonstrated the flexibility in the system, but it became also obvious that the Tibetic languages are flexible in different ways. That is, individual contexts do not trigger the same marked strategies. Not only do the different languages have different strategies for marking endopathic sensations, as shown by Nicolas Tournadre (egophoric yod, experiential ḥdug, or non-visual *grag), but the cut-off points for using egophoric or other markers may also differ considerably.

Nicolas Tournadre presented the new observation that tag questions may lead to a perspective shift with the factual and the sensory markers, but not with the egophoric and endopathic markers.

As for the different use of markers, Hiroyuki Suzuki showed, i.a., that in the dialect of Zhollam (Sems-kyi-nyila Khams), red is still a full verb with the meaning ‘be well’, and the factual function is expressed through the addition of (the experiential marker) snaṅ. – This corresponds to the combination yin + ḥdug in some western Tibetan varieties, yielding a ‘factual’ counterpart of sorts.

Zoe Tribur argued that the dialects of Amdo Tibetan, particularly of Golok Tibetan, may possibly not have a direct counterpart to sensorial ḥdug, as the corresponding form -kə would allow more applications in “Asserter-subject sentences” (i.e., the speaker in statements, the addressee in questions) than ḥdug would do in Lhasa Tibetan. More particularly, jo-kə would not only be used when the speaker realises unexpectedly that she has some money, but also when everybody else has knowledge about this fact. Nevertheless, with respect to the shift between egophoric yin and factual red, similar factors as observed in other languages apparently play a role: the speaker being the in­stigator, being emotionally involved, being related as family member or having immediate access allows using yin also for third persons, while plurality, cor­recting, or an attitude of empathy allow red for the self-descriptions. 

Wang Jiahong similarly showed that in Golok Tibetan, the ego copula yin may be used for third persons to express the epistemic origo’s asserted cer­tainty, e.g., through close relatedness as family member and personal experi­ence. She further showed that the ‘non-ego’ marker -ku (related to the above ‘objective’ marker -kə) is used for self-descriptions in dreams – a common fea­ture in Tibetic languages – and to express surprises concerning oneself. She further pointed to the fact that one could not anticipate the addressee’s sur­prise in questions, while it is possible to ask the addressee about his or her dream by anticipating the answer form. Wang Jiahong also gave examples for counterfactuals (irrealis mood), hardly discussed in the literature. While non-controllable verbs, such as fall down, usually do not allow egophoric marking, but require evidential, i.e. sensory, marking – like in other Tibetic languages –, Wang Jiahong showed that in Golok Tibetan an egophoric marker is allowed, when the speaker who lacks sensory perception of the sit­uation, e.g., because of having a blackout, comes to know by other means.

This appears to be quite different from what can be observed in other Tibetic languages, where an inferential or epistemic marker would be used in such situations. To some extent, however, it could be compared to the evidential shift observed in Dolpo Tibetan, where the non-visual marker ɖak is used when becoming aware of a sensation, while the visual marker duk/gyik is used some time later, as described by Nicolas Tournadre.

Juha Yliniemi pointed to the fact that a modern novel writer can use ego­phoric forms in Denjongke when describing third persons. Egophoric markers also seem to be preferred when information needs to be established or might be contested. By contrast, a speaker may use the ‘neutral’ copula for him-/ herself, when focusing more on the implications of the self-identification than about his/her identity. Most strikingly, Denjongke speakers do not differ­entiate between controllable and non-controllable verbs, that is, unlike in many other Tibetic languages, non-controllable verbs such as mthoṅ ‘(happen to) see’ take egophoric marking with the epistemic origo.

Inclusive plural restrictions to egophoric marking, as shown by Bettina Zeisler, are apparently only found in the Ladakhi dialects. She further showed how in Ladakhi, at least, the speaker’s identification or non-identifi­cation with the situation, the addressee’s knowledge, and in sum, the ‘upgrad­ing’ and ‘downgrading’ of one’s stance of epistemic authority play a constitu­ent role in the choice of the markers, overriding knowledge access to the ex­tent that one may claim one’s own possession (it’s mine) with the egophoric copula yin, but may need to ‘downtune’ one’s epistemic authority by using the ‘factual’/inferential marker man.ḥog when telling the addressee it doesn’t seem to be yours, referring to exactly the same item. She further highlighted that the flexible modification of the speaker’s stance is a feature of state­ments, but cannot be anticipated in information-seeking questions. A speaker may nevertheless ‘downtune’ his or her question in order to avoid an inquisi­tive tone or not to hold the addressee be responsible.

Searching for egophoric or factual marking among Non-Tibetic languages could be helpful when reconstructing the developmental path. Comparing Khams Tibetan with Japanese, Hiroyuki Suzuki showed that the develop­ment of factual markers may be based on verbs of becoming (red or ḥgyur in Tibetic languages, naru in Japanese). However, the resulting applications of the factual category differ: while a factual self-identification in Japanese points to shared knowledge, factual self-identification would be rather un­common in Khams Tibetan, except when stating that on has taken up a cer­tain role.

The Qiangic language of Western Minyag as described by Tamdrin Lhamo shows some similarities with the Tibetic system (most likely due to areal in­fluence): evidential marking for direct or sensory, indirect or inferential and two reportative markers depending on whether the report is based on oral or written communications, an egophoric marker for first person activities, plus two copulas differentiating between new and established knowledge.

Christian Huber discussed features in the Westhimalyan language Shumcho that have at least some family likeness to egophoric marking in Tibetan, but are restricted to statements and questions about third persons. To talk of egophoric marking thus appears to be far-fetched at first sight, which led to sharp criticisms. However, the motivations for using these ego-like markers for third persons correspond to quite some extent to the use of Tibetic ego­phoric markers for non-first persons, when the speaker has a close relation­ship with the person spoken about. Additionally, perspective shift in ques­tions can be observed, when the speaker expects the addressee to have the corresponding ego-like knowledge access through closeness or involvement concerning a third person.

As discussed after the workshop over some cups of tea and coffee, this kind of marking was most likely influenced by Tibetan, and one could think of a partial adaptation that helped to close a gap in the paradigm, namely the lack of 3rd person marking. That the system did not develop further into a Tibetan-type egophoric plus evidential marking may well have been blocked by the person markers for 1st and 2nd person subjects and objects.

One could possibly say that the ‘non-canonical’ or ‘non-prototypical’ use of egophoric and egophoric-like markers for the epistemic non-origo and the likewise ‘non-canonical’ or ‘non-prototypical’ use of any ‘non-egophoric’ marker for the epistemic origo, be it sensory, factual, or ‘objective’, result from similar factors that can be summarised as a speaker’s closeness to, and involvement in, elements of the situation and the absence or rejection thereof. In my view, this seems to be the particular effect of the egophoric-(like) systems, rather than of evidentiality in the crosslinguistic sense.

However, it also seems that not all participants would subscribe to this interpretation.

Adding a different perspective from languages far away from the sphere and influence of Tibetic languages, Henrik Bergqvist argued that the epistemic prefixes in the Kogi language, spoken in the Andes, indicate two dimensions, that of access symmetry or asymmetry and that of epistemic authority, the latter to be found either on the speaker’s side or on the addressee’s side. While neither the level of commitment nor the quality of the cognitive access would be at issue, a “speaker’s willingness to claim knowledge and to simultaneous­ly make assumptions concerning the addressee’s relation to the same knowl­edge is encoded in the prefixes.”

3. Analyses and disputes

Wang Jiahong suggested that the observable flexibility may be grounded in the combination of the origo’s epistemic authority and the discrepancy between the proposition and the origo’s knowledge schema. That is, a proposi­tion which is inconsistent with, contradicts, or has not yet been integrated in­to the origo’s knowledge schema, such as surprise upon own situations, would lead to the use of non-egophoric markers, and by contrast, well-ac­quaintedness of a third person’s situation through involvement or closeness would lead to the use of egophoric markers. In addition to this psychological interpretation, Wang Jiahong tried to show that while egophoric marking is definitely not a syntactic category, there might be a syntactic position in the left periphery of the syntax tree for speech acts, epistemic modals, evidential­ity, and thus also for egophoric marking. Her syntactic analyses, however, were not seen as being useful for explaining the psychological effects that she had shown.

With respect to the question whether the egophoric is an evidential category the opinions were divided. While Nicolas Tournadre emphatically asserted that it is an evidential category of access, Wang Jiahong, Zoe Tribur, and Tamdrin Lhamo differentiated between evidential and egophoric markers, while Bettina Zeisler was rather sceptical with respect of a superordinate evidential category, given the pragmatic underpinning of all markers.

Bettina Zeisler suggested thus a binary model in terms of a speaker’s having or not having the rights and the willingness to claim exclusively personal knowledge. Hence, the (equative) copula yin and the existential yod, indicat­ing highest epistemic authority and the speaker’s commitment, that is, the speaker’s identification with the situation relayed or his/her claim of respon­sibility or personal involvement, are contrasted with everything else. The lat­ter domain would then have the subdivisions for evidential (in the narrower cross-linguistic sense), epistemic, and factual marking. According to her, the observable flexibility in the system draws upon these main contrasts: yin vs. factual (Standard Spoken Tibetan red or Ladakhi yin.ḥog and related com­pounds) and yod vs. sensorial ḥdug (or rag). Nicolas Tournadre, however, objects any bifurcation and prefers a model where everything is about access and thus horizontally or­dered on one level, and possibly only the factual sticks out.

Nicolas Tournadre’s rejection of a binary model might have to do with the justified rejection of the binary concept of conjunct-disjunct in terms of a syntactic category. But this is not the only way of conceiving a binary con­trast.

Nicolas Tournadre further claimed that “egophoric and factual do not dif­fer in terms of epistemic authority”, a position that was objected by Bettina Zeisler as being theoretically highly problematic, – none the least as the al­leged ‘factuality’ and/ or evidential ‘neutrality’ of the marker stands in con­trast to crosslinguistic conceptualisations of factuality, which would well en­compass egophoric marking (see Kittilä 2019). It depends, of course, on how one defines ‘epistemic authority’: as being only an issue of access on the origo’s side or as being a question of rights to present knowledge in a particular way which is to be established in the dialogical setting.

Nicolas Tournadre tried to supported his claim with a striking example of conflict: a stepfather telling his stepson I am your father using either the ego­phoric copula yin or the factual marker red, and the stepson telling his step­father you are not my father, using either the negated egophoric copula ma.yin or the corresponding factual ma.red.

In this context, Nicolas Tournadre pointed to the subjective stance that the speaker has when using the egophoric copula yin combined with a notion of decision or future orientation ‘from now on’, which should then imply that the factual marker would convey a more ‘objective’ or generic stance. – Which brings us back to the terminological issue about ‘objectiveness’. Nicolas Tournadre, however, denied that the epistemic force of the ‘subjec­tive’ stance would be weaker.

Nicolas Tournadre’s claim that “egophoric and factual do not differ in terms of epistemic authority” might not necessarily be true in other contexts, as available descriptions also show that the so-called factual can be used for inferences, assumptions, even irrealis. If the factual marker were really ‘fac­tual’ or ‘objective’, one could expect that its use could be less aggressive than the ‘subjective’ ‘decisive’ copula in the above context.

Nicolas Tournadre’s description would certainly not hold for the Ladakhi counterpart of the factual marker. As I had the opportunity to test his example in Ladakh with two speakers, right now, a (step-) father’s use of ‘factual’ yin.ḥog instead of the egophoric copula yin would bolster up the father’s rejection of complaints on the part of the son, insisting on his authority within the family hierarchy, while a stepson’s use of ‘factual’ man.ḥog instead of the negated egophoric copula man, apart from indicating a sudden realisation upon viewing the relevant documents, would signal aggression or that there is “more behind”, because the ‘factual’ form would present the family member as a complete stranger. The egophoric markers would thus be pragmatically more neutral or common and thus less aggressive in these contexts. (I have similarly observed the preference for the egophoric marker in Ladakhi for introduc­tions, where the Standard Spoken Tibetan factual marker may be more polite – so much for different pragmatic strategies.)

Henrik Bergqvist pointed to the need to reanalyse the epistemic origo in dialogical settings, including both the speaker and the addressee in their relationship(s) to the object (or situation) talked about. He further sug­gested to use ‘epistemicity’ as a cover term for evidential and egophoric mark­ing or in other words: a “speaker’s representation of knowledge and the at­tribution of knowledge to either/both speech-act participant(s) is at the heart of epistemicity”.

In this context I should like to point to Bergqvist & Grzech’s recent (2023) article, where they argue convincingly that the focus on evidentiality as in­formation source or access is one-sided and an oversimplification. “[E]viden­tials situate events in an on-going linguistic exchange and the characterization of evidentials and evidentiality must therefore be grounded in the dialogical exchange between interlocutors and in the inter-personal context” (2023: 2f.)

Similarly, Ilana Mushin highlighted that conversation is not just a genre like story telling or teaching, rather it is the most basic language use and the “seedbed of grammar”. Pragmatics thus comes first.

Ilana Mushin further critically reviewed that most descriptions of evidential­ity concentrate on grammatical evidentials, but there will also be other, i.e., non-grammaticalised expressions. She further objected that a lot is usually said about the speaker’s choices, but little or nothing about the recipients. Why should one evaluate the context only in terms of normative politeness; why talk only about assertions, questions, and reported speech, and neglect other conversational actions, such as assessments and requests? Talking about normative politeness would also frame “epistemic stances as moral concern”; and while “specific rights and obligations may be culturally shaped”, “the foundation of epistemics is a feature of human social interaction.” Since conversations are dynamic, and the interlocutors are constantly adapt­ing their stance, it would be necessary to focus on the whole event and its flow instead of analysing only individual sentences. She exemplified this with the contrast between butterflies fixed in a display box and a video clip of but­terflies dancing in the air.

This prompted Juha Yliniemi’s remark that glossing necessarily breaks down the flow (or, to remain in Ilana Mushin’s analogy, it would kill the butter­flies) and he does not see a solution to this dilemma.

I should like to add, that it would hardly be possible to handle full conversa­tions and the description of turns and contexts, given the limited space usu­ally allowed for articles. Another problem is that the analyses of such record­ings usually only reflect the researcher’s interpretation, and even if one could ask the participants about their motivations, they would probably hardly re­member what they said and why. There are thus limitations, as in other methods, and the best method might be to combine different methods and different approaches to analysis.

4. Further perspectives

Since at least one participant and a few other scholars who could not attend the workshop had asked whether there might not be a publication on the workshop topic, this question was also brought to the forum. However, in­terest in a joint publication appeared to be rather non-existent.

Instead and in contrast to the workshop theme, Nicolas Tournadre sug­gested a collaborative effort to define the function of the five ‘core’ egophoric-evidential markers according to their standard usages and to develop a com­mon terminology.

Bettina Zeisler, however, insisted that one cannot define these markers without taking account of their ‘non-standard’ functions. She would think that it is high time to focus more on these latter usages, particularly since the ‘standard’ usages have been described times and again (see Bickel 2000, 2001 and Gawne & Hill 2017, as well as in the various individual publications on different Tibetic languages). While the ‘standard’ usages appear to be fairly common across the Tibetic languages, not withstanding the different handling of endopathic sensations, the pragmatics that lead to ‘non-standard’ usages might differ from speech community to speech community, and we should try to figure out where and why speech communities created different solutions to the communicative needs in interaction by documenting and comparing these solutions.

Neither suggestion, however, was met with enthusiasm.

With Ilana Mushin, I would further think that there are no ‘standard’ and ‘non-standard’ usages. These classifications or judgements are only due to our setting up of the paradigm(s), and do not do justice to the linguistic facts. Maybe one should rather think of redefining the paradigm(s), e.g., in terms of a speaker’s rights to present a piece of knowledge in a particular manner or not. While the workshop proposal aimed at coming closer to such redefini­tion, the workshop itself apparently failed to do so.

References

Bergqvist, Henrik and Karolina Grzech. 2023. The role of pragmatics in the definition of evidentiality. STUF – Language Typology and Uni­versals, 76.1: 1–30.

doi.org/10.1515/stuf-2023-2004.

Bickel, Balthasar (ed.) 2000. Person and evidence in Himalayan languages. Part I. Special issue of Linguistics of the Tibeto-Burman Area, 23.2.

Bickel, Balthasar (ed.) 2001. Person and evidence in Himalayan languages. Part II. Special issue of Linguistics of the Tibeto-Burman Area, 24.1

DeLancey, Scott. 1990. The historical status of the conjunct/disjunct pat­tern in Tibeto-Burman. Acta Linguistica Hafniensia, 25: 39–62.

Floyd, Simeon, Elisabeth Norcliffe, & Lila San Roque. 2018. Egophoric­ity. Amsterdam/ Philadelphia: Benjamins.

doi.org/10.1075/tsl.118.

Gawne, Lauren; and Nathan W. Hill (eds.) 2017. Evidential systems of Tibetan languages. Berlin; Boston: De Gruyter Mouton.

doi.org/10.1515/9783110473742.

Hale, Austin. 1980. Person markers: finite conjunct and disjunct verb forms in Newari. In: R. Trail (ed.), Papers in South-East Asian linguistics 7, 95-106.

Kittilä, Seppo. 2019. General knowledge as an evidential category. Linguistics, 57.6: 1271–1304.

doi.org/10.1515/ling-2019-0027.

Tournadre, Nicolas. 1994. Personne et médiatifs en tibétain. Faits de langues 3: 149–158.

doi.org/10.3406/flang.1994.918.

Further reading

Evans, Nicholas, Henrik Bergqvist, & Lila San Roque. 2018a. The gram­mar of engagement I: framework and initial exemplification. Language and Cognition, 10: 110–140.

doi.org/10.1017/langcog.2017.21.

Evans, Nicholas, Henrik Bergqvist, & Lila San Roque. 2018b. The grammar of engagement II: typology and diachrony. Language and Cognition, 10: 141–170.

doi.org/10.1017/langcog.2017.22.

Grzech, Karolina. 2020. Epistemic primacy, Common Ground management, and epistemic perspective. In: Henrik Bergqvist and Seppo Kittilä (eds.), Evidentiality, egophoricity and engagement. Berlin: Language Science Press, pp. 23–60.

Häsler, Katrin Luise. 2001. An empathy-based approach to the description of the verb system of the Dege dialect of Tibetan. In B. Bickel (ed.), Person and evidence in Himalayan languages. Part II. Special issue of Linguistics of the Tibeto-Burman Area, 24.1: 1–34.

Hintz, Daniel J. and Diane M. Hintz. 2014/17. The evidential category of mutual knowledge in Quechua. Lingua (2017), 186–187: 88–109.

doi.org/10.10.16/j.lingua.2014.07.014.

Kamio, Akio. 1997. Territory of information. Amsterdam, Phil.: Benjamins.

Mushin, Ilana. 2012. “Watching for witness”. Evidential strategies and epis­temic authority in Garrwa conversation. Pragmatics and Society, 3.2: 270–293.

Tournadre, Nicolas. and Randy. LaPolla. 2014. Towards a new approach to evidentiality. Issues and directions for research. Linguistics of the Tibeto-Burman Area, 37.2: 240–262.

Zeisler, Bettina. 2023. Beyond evidentiality, the case of Ladakhi inok & sib­lings. Himalayan Linguistics, Archive 13.

escholarship.org/uc/item/0tm2v2sr.

Zeisler, Bettina. 2024. Facts and attitudes: on the so-called ‘factual’ markers of the modern Tibetic languages. Himalayan Linguistics, Archive 14. (To appear shortly.)