Edited by Konstanze Jungbluth, Cornelia Müller, Nicole Richter, Hartmut Schröder
Epistemic Stance in Italian and in the Mind
Andrzej Zuczkowski, Ramona Bongelli, Ilaria Riccioni. 2017. Epistemic stance in dialogue: Knowing, unknowing, believing, (Dialogue studies 29). Amsterdam: John Benjamins.
This review provides a brief summary of each chapter before concluding with an overall evaluation of the book as a contribution to research on epistemic stance in general linguistics. Readers with a primary interest in a purchase recommendation might want to jump directly to the last section.
2.2 Chapter 2
Evidentiality is defined as a linguistic device that refers to source of information, while other possible definitions (source of knowledge, modes of knowing) are rejected due to a lack of clarity about the notion of knowledge (as opposed to belief, imagination, etc.). Moreover, the authors reject the applicability of experiential evidential categories (i.e. reported/hearsay) to past events. Instead, they include cognitive processes (i.e. think, believe, imagine, remember) as modes of acquiring information, some of which are linked to specific temporal relations (remember-past, imagine-future).
The wide range of possible definitions for epistemicity (reliability of information, judgement of likelihood of the proposition, commitment to the truth of a message) are reduced to degrees of certainty. The authors postulate that certainty and uncertainty are communicated through both epistemic and evidential markers. Unfortunately, they fail to define certainty beyond the claim that it is independent of the notion of truth, which they attribute to an extra-linguistic process of corroboration. This leaves us with two negative definitions: epistemicity as a contribution to certainty that is not evidential, and certainty as a claim to truth that has to be verified extra-linguistically in order to become true.
2.3 Chapter 3
markers of evidentiality and epistemicity in a corpus of 780 written texts obtained during a video-description task where participants were asked to describe what they had seen and felt (ital.: ‘visto e sentito’) when watching three recordings of different dances (dervishes and acrobatic contemporary dances). The texts were qualitatively evaluated sentence by sentence for epistemic markers. The authors opt for an exemplification of their methodology with three text examples from the corpus, two of which show no world-constitutive (wc) markers. The first text (T1) is written in the past tense and mentions seeing and hearing as source of information, so the authors attribute it a chain of worlds as in (2):
(2) [I REMEMBER [I have seen and heard [T1]dp]wcp]pp
Whereas the second text does not contain wc-markers either, the third contains five types of world-constitutive markers (4 verbs: penso ‘I think’, credo ‘I believe’, ritengo ‘I think’, mi sembra ‘it seems to me’; 1 adverb: forse ‘perhaps’). It thereby illustrates some lexical possibilities for expressing epistemic meanings in Italian and gives the reader an idea of what forms of expression are relevant for the KUB model.
Subsection 3.2 continues the discussion of the three corpus texts as well as made-up examples, but now with the goal of introducing the relation between evidential and epistemic worlds and the KUB model. Its main proposal is to interpret certainty as a form of communicating a Knowing position, uncertainty as a form of communicating a Believing position, and wh-questions as a form of communicating an Unknowing position.
Subsections 3.3 to 3.6 are dedicated to a summary of the respective markers of these positions. The Knowing position is attributed to verbs of knowing in the 1.PERS.PRST.IND. (such as ricordo ‘I remember’, sento ‘I hear/feel’, vedo ‘I see’, so ‘I know’) and in the 3.PERS.PRST.IND. (as in immagini che richiamano alla mente I vortici corporali di Aldo Mondino ‘images that recall the whirling bodies of Aldo Mondino’). Moreover, verbal expressions such as sono certo ‘I am sure’ or non ho dubbi ‘I have no doubt’, as well as adverbs and adjectives such as indubbiamente ‘undoubtedly’, sicuramente ‘surely’, certo ‘certain’ etc., are subsumed under the Knowing position. The Unknowing position is attributed to indirect wh-questions as in (3), negations of verbs of the Known (4), and positive forms of ignorare (5). It is moreover associated with adjectives like sconosciuto ‘unknown’, ignoto ‘unknown’, a me incomprensibile ‘incomprehensible to me’.
The Believing position covers a particularly large set of expressions, the categories of which are reproduced here:
• Lexical markers
o Verbs (suppongo ‘I suppose’, potere ‘can’, dovere ‘may/must’)
o Adjectives and adverbs (probabile ‘likely’, possibile ‘possible’, forse ‘perhaps’ etc.)
o Nouns (imperssione ‘impression’, dubbio ‘doubt’ etc.)
o Noun expressions (secondo me ‘In my opinion’, a mio parere ‘in my view’ etc.)
• Morphosyntactic markers
o Verbs in the conditional mood (potrei ‘could’, dovrei ‘should’, vorrei ‘would’ etc.)
o Uncertain questions (È forse X? ‘Is it maybe X?’)
o If-clauses and congiuntivo trapassato
o Epistemic future
Finally, we also learn that, in total, about 13 percent of the words in the dervishes sub-corpus were syntactically dominated by a marker of Belief, 2 percent of markers of Unknowing, and the rest was communicated from a Knowing position.
2.4 Chapter 4
The first experiment included 18 university students and two tasks. First, an individual production task had participants produce sentences that were in some way certain or uncertain. This method obtained 23 ways of expressing degrees of certainty and 22 ways of expressing degrees of uncertainty. Then, an inter-observation task asked the participants to discuss and collectively order the sentences along a horizontal bar (printed on paper)
according to their degree of certainty. Finally, the participants concluded the task by drawing a perpendicular line to indicate a division between certainty and uncertainty that would divide the entire set of expressions. The results show that both certainty and uncertainty are perceived as graded, though uncertainty receives a much higher degree of gradedness than certainty (with 2 out of 6 groups of participants considering it a point, that is, not gradable).
The second experiment involved 120 university students who had to rate the degree of certainty in 12 sentences on a 7-point Likert scale. The stimuli consisted of 6 pairs of sentences corresponding to the following categories of markers: (i) sentence structure: declarative – interrogative; (ii) knowing – not knowing whether: I know – I don’t know whether; (iii) doubt – absence of doubt: undoubtedly – I doubt; (iv) certainty – uncertainty: I’m certain – I’m not certain; (v) belief: I believe – I think; (vi) probability: probably – perhaps. All these markers were combined with a carrier sentence about catching the 3 pm train, as in (4):
The results are evaluated by means of Guttman’s deterministic scalogram (1950) and an Extended Logistic Model of the Rasch Model family (see e.g. Andrich 1978a; Andrich 1978b; 2010 among others). They show that all 12 sentences express uncertainty on a single continuum, whereas the interrogative sentence does not lie on a continuum of certainty with the other 11 sentences. The rest of the examples yield continuums for both dimensions with an ‘item separation index’ of 0.69, i.e. ‘good’ (Wright and Masters 1982; Embretson and Reise 2009). Moreover, the function obtained by combining the ratings for the two scales yields an identity function, which means that the two questionnaires measure one latent construct (Bond and Fox 2001).
2.5 Chapter 5
In their previous study, the authors tried to capture the dynamics of troubles talk by focusing on the complementary roles of the confider (the party who talks about their own trouble) and
the confidant (the receiver of confidence) in what they see as an advice-giving triplet: Initiation – Advice – Reaction (IAR). Each position in the triplet has a certain paradigm of forms it can take (see Table 1 for forms and their number of occurrences in a total of 124 triplets).
Table 1. Occurrences of advice giving strategies in troubles talk in Riccioni et al. (2014)
The three possible reaction categories do not sum up to the total of 124 because in 24 cases, there was no clear reaction by the confidants. Instead, they often ignored unrequested and unmitigated advice and simply continued their line of thought (a reaction coined self-continuity). Only seven combinations of IAR-triplets reached a frequency of above 5 percent. They are presented in Table 2.
Table 2. IAR triplet types in Riccioni et al. (2014) with frequency above 5 %
The core tendency is that unrequested advice will lead to misalignment in reactions, with mitigation playing a role in aligning the interlocutors’ stances in some cases. The KUB model is then used as a means of explaining this tendency in the analysis of seven dialogues. They showcase a strong misalignment effect in one case of advice giving where confider and confidant both operate from a Knowing position, whereas different degrees of epistemic negotiation take place in the six other configurations of Knowing, Believing, and Unknowing confider and confidants.
2.6 Chapter 6
In sum, the journalists play two different roles, one displaying trust in the clairvoyant and the other showing skepticism, while the clairvoyant positions herself in a knowing position and aligns her answers only with the trustful interlocutor.
2.7 Chapter 7
The quantitative evaluation of the annotated excerpts shows that the participants live up to their roles by assuming different epistemic positions. The attorney puts himself in a Knowing position by means of declarative sentences and the use of adverbs (absolutely, completely, unequivocally). The journalists, in turn, use a Believing position (consisting of hypotheses and ‘hostile, hyperbolic rhetorical questions’) to raise doubts about the attorney’s view. Finally, the two psychiatrists build up a Believing position using counter-argumentation.
2.8 Chapter 8
Building on Kamio’s theory, Heritage established the distinction between epistemic status and epistemic stance. The former is a role associated with responsibilities and rights in the social construction of truth, while the latter is a position an interlocutor can take in a conversation by choosing specific linguistic forms. Incongruence between the epistemic statuses of interlocutors are then seen as the driving force (the ‘epistemic engine’) behind conversations in which the participants, who can occupy either a more knowledgeable (K+) or a less knowledgeable (K-) position, try to reach an informational equilibrium. The main difference between Heritage’s theory and the KUB model is the absence of a Believing position, represented only in terms of degrees of knowledge on an assumed epistemic gradient between speaker and hearer.
2.9 Chapter 9
Figure 1. KUB provenance and destination of wh-questions (solid line), yes-no questions (dashed line), and rhetorical questions (dash-dotted line).
The analysis of individual utterances or sentences in transcriptions from dialogue corpora, in turn, demands a fine-grained labelling and glossing of the lexical, morphological, and syntactic strategies that contribute to these levels of meaning. Zuczkowski, Bongelli, and Riccioni aim at accounting for both lexical and morphological forms, while taking into account the scope of these markers within specific sentences. In the discussion of specific examples, though, only passages of text are highlighted, without any further information on the grammatical mechanisms involved. From the perspective of linguistics, this reduces the cross-linguistic and theoretical comparability of the phenomena analyzed. From a typological perspective, a higher degree of consciousness about the specific language that forms the object of study, Italian, would also have made it easier to tie the results of this research to other investigations of epistemic and interactional phenomena in Italian (cf. Portner and Zanuttini 2003; Squartini 2010; Cruschina 201), Romance languages (Cornillie 2010; Diewald and Smirnova 2010; Haßler 2010), and genetically unrelated languages (van der Auwera and Ammann 2013).
Finally, the authors make frequent use of prosodic cues to disambiguate the epistemic meanings of the sentences they analyze. The reader can only guess the specific acoustic form of what they call “hostile hyperbolic rhetorical questions” (Zuczkowski, Bongelli and Riccioni 2017:254) and has to believe the authors when they claim that “from the tone with which the question is uttered it is impossible to assume that it is a challenge question (Koshik 2003) or an unmasking question (Vincze et al. 2016).” (Zuczkowski et al. 2017:173). Given the recent surge of interest in epistemic prosody, a more explicit account of such cues would be a next step in this promising line of research (cf. Savino and Grice 2011, Vanrell et al. 2013, Moraes and Rilliard 2014; Frota and Prieto 2015; Bianchi, Bocci and Cruschina 2016; Reich in press).
Andrich, David. 1978a. A rating formulation for ordered response categories. Psychometrika 43(4):561–573. doi:10.1007/BF02293814.
Andrich, David. 1978b. Application of a Psychometric Rating Model to Ordered Categories Which Are Scored with Successive Integers. Applied Psychological Measurement 2(4): 581–594. doi:10.1177/014662167800200413.
Andrich, David. 2010. Sufficiency and Conditional Estimation of Person Parameters in the Polytomous Rasch Model. Psychometrika 75(2): 292–308. doi:10.1007/s11336-010-9154-8.
Austin, John L. 1962. How to Do Things with Words. Oxford: At the Clarendon Press.
Bianchi, Ivana, Roberto Burro, Stefania Torquati, and Ugo Savardi. 2013. The Middle of the Road: Perceiving Intermediates. Acta psychologica 144(1): 121–135. doi:10.1016/j.actpsy.2013.05.005.
Bianchi, Valentina, Giuliano Bocci, and Silvio Cruschina. 2016. Focus fronting, unexpectedness, and evaluative implicatures. Semantics and Pragmatics 9(3): 1-54. doi:10.3765/sp.9.3.
Bond, Trevor G., and Christine M. Fox. 2001. Applying the Rasch Model: Fundamental Measurement in the Human Sciences. Mahwah, N.J., London: L. Erlbaum.
Bongelli, Ramona, and Andrzej Zuczkowski. 2008. Indicatori linguistici percettivi e cognitive, (Vol. 348). Roma: Aracne.
Cornillie, Bert. 2010. An interactional approach to epistemic and evidential adverbs in Spanish conversation. In Gabriela Diewald and Elena Smirnova (eds.), Linguistic realization of evidentiality in European languages, (Empirical approaches to language typology 49). Berlin, New York: De Gruyter Mouton, 309–330.
Cruschina, Silvio. 2011. Discourse-related features and functional projections. Oxford, New York: Oxford University Press.
Diewald, Gabriele, and Elena Smirnova. 2010. Introduction. Evidentiality in European languages: the lexical-grammatical distinction In Gabriela Diewald and Elena Smirnova (eds.), Linguistic realization of evidentiality in European languages, (Empirical approaches to language typology 49). Berlin, New York: De Gruyter Mouton, 1–14.
Embretson, Susan E., and Steven P. Reise. 2009. Item Response Theory for Psychologists, (Vol. 4. 2009th ed.). New York: Psychology Press.
Frota, Sónia, and Pilar Prieto (eds.) 2015. Intonation in Romance. Oxford: Oxford University Press.
Guttman, Louis. 1950. The Basis for Scalogram Analysis. In Samuel A. Stouffer (ed.), Measurement and prediction, (Studies in social psychology in World War II 4). Princeton, N.J.: Princeton University Press, 60–90.
Haßler, Gerda. 2010. Epistemic modality and evidentiality and their determination on a deictic basis: the case of Romance languages. In Gabriela Diewald and Elena Smirnova (eds.), Linguistic realization of evidentiality in European languages, (Empirical approaches to language typology 49). Berlin, New York: De Gruyter Mouton, 223–248.
Heritage, John. 2012a. Epistemics in Action: Action Formation and Territories of Knowledge. Research on Language & Social Interaction 45(1): 1–29. doi:10.1080/08351813.2012.646684 .
Heritage, John. 2012b. The Epistemic Engine: Sequence Organization and Territories of Knowledge. Research on Language & Social Interaction 45(1): 30–52. doi:10.1080/08351813.2012.646685.
Kamio, Akio. 1997a. Evidentiality and Some Discourse Characteristics in Japanese. In Akio Kamio (ed.), Studies in language companion series, Directions in Functional Linguistics. Amsterdam: John Benjamins, 145–172.
Kamio, Akio. 1997b. Territory of information, (new ser. 48). Amsterdam, Philadelphia: John Benjamins.
Koshik, Irene. 2003. Wh-questions used as challenges. Discourse Studies 5(1): 51–77. doi:10.1177/14614456030050010301.
Metzger, Wolfgang. 1975. Psychologie: Die Entwicklung ihrer Grundannahmen seit der Einführung des Experiments, (Vol. 52. 5th ed.). Heidelberg: Steinkopff; Imprint.
Moraes, João A. de, and Albert Rilliard. 2014. Illocution, attitudes and prosody: A multimodal analysis. In Tommaso Raso and Heliana Mello (eds.), Studies in Corpus Linguistics, Spoken Corpora and Linguistic Studies. Amsterdam: John Benjamins, 233–270.
Petöfi, János S. 1973. Towards an Empirically Motivated Grammatical Theory of Verbal Texts. In J. S. Petöfi and Hannes Rieser (eds.), Studies in Text Grammar. Springer Science+Business Media B.V, 205–275.
Portner, Paul, and Rafaela Zanuttini. 2003. Exclamative clauses at the syntax-semantics interface. Language 79(1): 39–81.
Reich, Uli. in press. Presupposed modality. In Marco García García and Melanie Uth (eds.), Studies in language companion series, Focus Realization and Interpretation in Romance and Beyond. Amsterdam/Philadelphia: John Benjamins.
Riccioni, Ilaria, Ramona Bongelli, and Andrzej Zuczkowski. 2014. Mitigation and epistemic positions in troubles talk: The giving advice activity in close interpersonal relationships. Some examples from Italian. Language & Communication 39: 51–72. doi:10.1016/j.langcom.2014.08.001.
Savino, Michelina, and Martine Grice. 2011. The Perception of Negative Bias in Bari Italian Questions. In Sònia Frota, Gorka Elordieta, and Pilar Prieto (eds.), Studies in natural language and linguistic theory, Prosodic Categories: Production, Perception and Comprehension. Dordrecht: Springer Netherlands, 187–206.
Squartini, Mario. 2010. Where mood, modality and illocution meet: the morphosyntax of Romance conjectures. In Martin G. Becker and Eva-Maria Remberger (eds.), Modality and mood in romance: Modal interpretation, mood selection, and mood alternation, (Linguistische Arbeiten 533). Berlin: De Gruyter, 109–131.
van der Auwera, Johan, and Andreas Ammann. 2013. Overlap between Situational and Epistemic Modal Marking: The World Atlas of Language Structures Online. Leipzig: Max Planck Institute for Evolutionary Anthropology.
Vanrell, Maria d. M., Joan Mascaró, Francesc Torres-Tamarit, and Pilar Prieto. 2013. Intonation as an Encoder of Speaker Certainty: Information and Confirmation Yes-No Questions in Catalan. Language and Speech 56(2): 163–190. doi:10.1177/0023830912443942.
Vincze, Laura, Ramona Bongelli, Ilaria Riccioni, and Andrzej Zuczkowski. 2016. Ignorance-unmasking questions in the Royal-Sarkozy presidential debate: A resource to claim epistemic authority. Discourse Studies 18(4): 430–453. doi:10.1177/1461445616646924.
Wright, Benjamin D., and Geoff N. Masters. 1982. Rating Scale Analysis: Rasch measurement. Chicago: Mesa Pr.