1.Introduction

The focus of the present study is on the reflexive use of language as observed in Persian indirect reports. Indirect reports are complicated language games that require (intricate) online cognitive processing (e.g., an appraisal of the event as it occurs) and social awareness (Capone 2016Capone, Alessandro 2016 The Pragmatics of Indirect Reports: Socio-philosophical Considerations. Cham: Springer., 2019Capone, Alessandro 2019 Pragmatics and philosophy: Connections and ramifications. Cham: Springer. ; Morady Moghaddam and Capone 2020Morady Moghaddam, Mostafa, and Alessandro Capone 2020 “Metalanguage and Subjectivity in Indirect Reports.” Lingua, 236: 102784. ). The idea of reflexivity closely deals with guiding listeners through the proper interpretation of utterances (Lucy 1993aLucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. ). With regard to utterance interpretation, Capone (2019Capone, Alessandro 2019 Pragmatics and philosophy: Connections and ramifications. Cham: Springer. , 226) points out that “there are pragmatic increments to utterance interpretation which are fuelled through explicatures.” Likewise, the practice of indirect reporting is under the control of the Paraphrase Principle, concerning which Capone (2010Capone, Alessandro 2010 “The Social Practice of Indirect Reports.” Journal of Pragmatics 42: 377–391. , 382) states that “[t]he that-clause embedded in the verb ‘say’ is a paraphrase of what Y said if it meets the following constraint: Should Y hear what X said Y had said, Y would not take issue with it, but would approve of it as a fair paraphrase of the original utterance” (emphasis added). When it comes to a ‘fair paraphrase’, the reporter is cautioned to provide sufficient linguistic (and paralinguistic) markers to help the hearer distinguish between the reporting speaker’s and the reported speaker’s voice (Salmani Nodoushan 2015Salmani Nodoushan, Mohammad A. 2015 “The Secret Life of Slurs from the Perspective of Reported Speech.” Rivista Italiana di Filosofia del Linguaggio 9(2): 92–112.; Capone 2016Capone, Alessandro 2016 The Pragmatics of Indirect Reports: Socio-philosophical Considerations. Cham: Springer.). According to Wettstein (2016Wettstein, Howard 2016 “Speaking for Another.” In Indirect Reports and Pragmatics: Interdisciplinary Studies (vol. 5), ed. by Alessandro Capone, F. Kiefer, and F. L. Piparo, 405–434. Heidelberg: Springer. , 421), “The reporter must be faithful to the original speaker’s remark. At the same time the reporter needs to choose a sentence that in the current context conveys the original speaker’s point. And there may well be no uniquely correct way to satisfy both desiderata.” On this account, reflexivity empowers the reporter to use metalanguage as an index of extralinguistic clues to facilitate the simultaneous interpretation process. In this case, Capone (2019Capone, Alessandro 2019 Pragmatics and philosophy: Connections and ramifications. Cham: Springer. , 7) rightly mentions that “linguistic resources and contextual clues and cues that direct the hearer towards the recovery of the implicature.”

That said, acknowledging the reflexive part of indirect reporting (metapragmatic comments) is in itself an intriguing task, since the reporter does not explicitly mention which parts are added, modified, or deleted. Thus, the hearer is left to his/her own devices to guess which parts of the report are related to the original discourse and which parts have been manipulated (the hearer should separate the voices). One way to ease the hearer’s burden in distinguishing between voices in indirect reporting is the use of metalanguage (or contextualisation cues, according to Gumperz [1982]Gumperz, John J. 1982 Discourse Strategies. Cambridge: Cambridge University Press. ). Contextual cues in the indirect reporter’s utterance can communicate “[t]he functions of metapragmatic expressions [that] include monitoring, commenting on, or assessing the appropriateness of one’s own or another speaker’s discourse, expressing a negative or positive attitude toward a portion or aspect of the discourse, […] and negotiating potentially problematic stances” (Blackwell 2016Blackwell, Sarah E. 2016 “Porque in Spanish Oral Narratives: Semantic Porque, (Meta)Pragmatic Porque or Both?” In Interdisciplinary Studies in Pragmatics, Culture and Society, ed. by Alessandro Capone, and Jacob L. Mey, 615–652. Heidelberg, Springer. , 619). From the researchers’ viewpoint, reflexivity is an important element of indirect reporting, since it helps researchers better to analyse the underlying features of opaque structures (although the practice of reflexivity may not be [properly] acknowledged by the hearer).

The aim of this study is to analyse the reflexive strategies used by the reporter and the original speaker in line with the ‘third sense’ of metapragmatics proposed by Caffi (2009)Caffi, Claudia 2009 “Metapragmatics.” In Concise Encyclopedia of Pragmatics (2nd ed.), ed. by Jacob L. Mey, 625–630. Amsterdam: Elsevier, Amsterdam., who proposes three senses of metapragmatics. After providing some background knowledge on the concept of metapragmatics, I discuss the third sense of metapragmatics (Caffi 2009Caffi, Claudia 2009 “Metapragmatics.” In Concise Encyclopedia of Pragmatics (2nd ed.), ed. by Jacob L. Mey, 625–630. Amsterdam: Elsevier, Amsterdam.), which is the most relevant to my study of indirect reporting. In Section 2 and 3, I deal with the concept of opacity and the practice of indirect reporting. A brief discussion is provided in Section 4 on the indirect reports in Persian. In Section 5, I explain methodology and the procedure of data collection. In Sections 6 and 7, I analyse the qualitative data and discuss the main findings. Finally, in Section 8, I briefly talk about the main results and the implication of the study.

2.Background

Metapragmatics is closely related to and preceded by the concept of metalanguage (a term introduced by Jakobson [1960]Jakobson, Roman 1960 “Closing Statement: Linguistics and Poetics.” In Style in Language, ed. by T. A. Sebeok, 350–377. Cambridge, MA: MIT Press.), which is anchored in the contextualisation cues defined by Gumperz (1982Gumperz, John J. 1982 Discourse Strategies. Cambridge: Cambridge University Press. , 77) as “a class of verbal signs […] that serve to relate what is said on any particular time-bound occasion to knowledge acquired through past experience.” The idea of metalanguage is a crucial element of linguistics (Jakobson 1960Jakobson, Roman 1960 “Closing Statement: Linguistics and Poetics.” In Style in Language, ed. by T. A. Sebeok, 350–377. Cambridge, MA: MIT Press.). As pointed out by Ilie (2016Ilie, Cornelia 2016 “Metadiscursive Strategies in Dialogue: Legitimising Confrontational Rhetoric?” In Interdisciplinary Studies in Pragmatics, Culture and Society, ed. by Alessandro Capone, and Jacob L. Mey, 601–614. Heidelberg: Springer. , 611), “[m]etadiscursive utterances represent concomitant and accompanying speech acts that comment on, refer to and/or complement particular speech acts that they evaluate at the same time.” Metalanguage is particularly important because it is closely germane to the concept of reflexivity, which is a crucial feature of verbal interaction and “create[s] frames of interpretation for social interaction” (Gal 2009Gal, Susan 2009 “Linguistic Anthropology.” In Concise Encyclopedia of Pragmatics (2nd ed.), ed. by Jacob L. Mey, 519–533. Amsterdam, Elsevier., 520). Metalanguage refers to how the decoder and encoder manage the utterances to create presuppositions that are implicit assumptions about the world or background belief relating to a specific utterance.

Equally relevant to the concept of metapragmatics is Lucy’s (1993a)Lucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. discussion on reflexive language. A good point to start with is the definition of reflexive language: “[s]peech is permeated by reflexive activity as speakers remark on language, report utterances, index and describe aspects of the speech event, invoke conventional names, and guide listeners in the proper interpretation of their utterances. This reflexivity is so pervasive and essential that we can say that language is, by nature, fundamentally reflexive” (Lucy 1993aLucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. , 11). As argued by Lucy, reflexive language “tell[s] listeners how to interpret the speech they are hearing” (Lucy 1993aLucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. , 10). Additionally, Mey (2001)Mey, Jacob L. 2001 Pragmatics: An Introduction. Oxford: Blackwell. states that reflexivity conceptualises “the way language is able to reflect on itself, make statements about itself, question itself, improve itself, quote itself and so on” (Mey 2001Mey, Jacob L. 2001 Pragmatics: An Introduction. Oxford: Blackwell., 177).

Metapragmatics has ‘therapeutic potential’ (Caffi 2009Caffi, Claudia 2009 “Metapragmatics.” In Concise Encyclopedia of Pragmatics (2nd ed.), ed. by Jacob L. Mey, 625–630. Amsterdam: Elsevier, Amsterdam.) that is related to making corrections, clarifying, and adjusting the discourse. Considering that “the means and functions of metapragmatics […] are a sub-category of reflexive language” (Hübler and Wolfram 2007Hübler, Axel, and Wolfram Bublitz 2007 “Introducing Metapragmatics in Use.” In Metapragmatics in Use, ed. by W. Bublitz, and A. Hübler, 1–26. Amsterdam: John Benjamins. , 5), “[m]etapragmatic acts, inter alia, assign (either explicitly or implicitly) particular pragmatic functions to speech acts or discourse types” (Hübler and Wolfram 2007Hübler, Axel, and Wolfram Bublitz 2007 “Introducing Metapragmatics in Use.” In Metapragmatics in Use, ed. by W. Bublitz, and A. Hübler, 1–26. Amsterdam: John Benjamins. , 5). Metapragmatics can be conceptualised as ‘consciousness’, “whereby speakers analyze and evaluate linguistic facts by taking into consideration their own participation, intentions, and deliberate use of language in discourse” (Blackwell 2016Blackwell, Sarah E. 2016 “Porque in Spanish Oral Narratives: Semantic Porque, (Meta)Pragmatic Porque or Both?” In Interdisciplinary Studies in Pragmatics, Culture and Society, ed. by Alessandro Capone, and Jacob L. Mey, 615–652. Heidelberg, Springer. , 616). In this sense, Wettstein (2016Wettstein, Howard 2016 “Speaking for Another.” In Indirect Reports and Pragmatics: Interdisciplinary Studies (vol. 5), ed. by Alessandro Capone, F. Kiefer, and F. L. Piparo, 405–434. Heidelberg: Springer. , 419) states that “[t]he reporter’s ability to convey the speaker’s point would be increased substantially were we to allow her to alter her formulation dynamically.” In the present study, I adhere to Caffi’s metapragmatics 3, namely that of “management of discourse”. Metapragmatics 3 “[…] is concerned with the investigation of that area of the speakers’ competence that reflects judgments of appropriateness on one’s own and other people’s communicative behavior. In this sense, metapragmatics deals with the ‘know-how’ regarding the control and planning of, as well as feedback on, the ongoing interaction” (Caffi 2009Caffi, Claudia 2009 “Metapragmatics.” In Concise Encyclopedia of Pragmatics (2nd ed.), ed. by Jacob L. Mey, 625–630. Amsterdam: Elsevier, Amsterdam., 625, for further information refer to Caffi 2009Caffi, Claudia 2009 “Metapragmatics.” In Concise Encyclopedia of Pragmatics (2nd ed.), ed. by Jacob L. Mey, 625–630. Amsterdam: Elsevier, Amsterdam.).

Indirect reports are examples of mutual manifestness, to use Sperber and Wilson’s (1986)Sperber, Dan, and Deirdre Wilson 1986 Relevance: Communication and Cognition. Oxford: Blackwell. conceptualisation, in that the hearer or the addressee is expected to construct the communicator’s intention either via the immediate physical environment or based on previous assumptions/experiences (of course, interpretation clash may happen in indirect reports as well). Relevance Theory is of importance since it combines performance and competence in a specific theory of communication. Indirect reports, as a mixed game, can be conceptualised as a sociocognitive event that takes into account Sperber and Wilson’s Relevance Theory. Indirect reports are cases of ostensive behaviour because, by reporting others, the reporting speaker has an intention in mind (we do not cite others without any specific reason) that he/she wants the audience to notice. This intention, however, could be fake, meaning that the reporting speaker may deliberately change the original speaker’s utterances to provide a particular intention (to convey a certain impression, according to Goffman).

More specifically, Lucy refers to two major forms of reflexivity. The first form is when one talks about the general regularities of language use in examples such as ‘get is an irregular verb’ and ‘cilantro is another name for coriander’ (Lucy 1993aLucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. ). In this case, languages refer to their own structure and use (attending to speech as opposed to attending from speech [Hübler and Bublitz 2007Hübler, Axel, and Wolfram Bublitz 2007 “Introducing Metapragmatics in Use.” In Metapragmatics in Use, ed. by W. Bublitz, and A. Hübler, 1–26. Amsterdam: John Benjamins. ]). The second form deals with the representation of the speech that is remembered or imagined through particular acts of speech. The instantiation of this form of reflexivity is observed in examples such as ‘he said [that] I got a great haircut’ and ‘Tom complimented me today’ (Lucy 1993aLucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. ). Lucy further proposes the ‘literary-performance approaches’ to reflexive language, concerning which he states that “there has been concern with the ways verbal art can effectively represent different perspectives simultaneously” (Lucy 1993aLucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. , 18). Lucy (1993a)Lucy, John A. 1993a “Reflexive Language and the Human Disciplines.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 9–32. Cambridge: Cambridge University Press. differentiates between direct and indirect reports based on the frame of orientation, whereby the indirect reporter adopts the perspective of the original speaker. The distinction between direct and indirect reporting contributes to a theory of reflexive language, since the indirect reporter can manipulate the language to convey his/her own point of view during the report using contextual clues and cues.

3.Persian indirect reports

Persian reported speech has received scant attention in the literature. To the best of my knowledge, only few research articles have been published in this regard (Capone and Salmani Nodoushan 2014Capone, Alessandro, and Mohammad A. Salmani Nodoushan 2014 “On Indirect Reports and Language Games: Evidence from Persian.” Rivista Italiana di Filosofia del Linguaggio 8(2): 26–42.; Morady Moghaddam and Capone 2020Morady Moghaddam, Mostafa, and Alessandro Capone 2020 “Metalanguage and Subjectivity in Indirect Reports.” Lingua, 236: 102784. ; Salmani Nodoushan 2015Salmani Nodoushan, Mohammad A. 2015 “The Secret Life of Slurs from the Perspective of Reported Speech.” Rivista Italiana di Filosofia del Linguaggio 9(2): 92–112.; Salmani Nodoushan 2018Salmani Nodoushan, Mohammad A. 2018 “Which View of Indirect Reports Do Persian Data Corroborate?” International Review of Pragmatics 10: 76–100. ). Capone and Salmani Nodoushan (2014)Capone, Alessandro, and Mohammad A. Salmani Nodoushan 2014 “On Indirect Reports and Language Games: Evidence from Persian.” Rivista Italiana di Filosofia del Linguaggio 8(2): 26–42. argue that Persian indirect reports do not allow syntactic transformations, where samesaying is 100 per cent perfect in Persian quotations and indirect reports (Capone and Salmani Nodoushan 2014Capone, Alessandro, and Mohammad A. Salmani Nodoushan 2014 “On Indirect Reports and Language Games: Evidence from Persian.” Rivista Italiana di Filosofia del Linguaggio 8(2): 26–42.). In this sense, they argue, “whatever fills the ‘quotative slot’ after ‘ke’ /ke/ (or is wrapped inside ‘giume’ /giːjuːme/ following a colon) is a demonstration of the original speaker’s assertions” (Capone and Salmani Nodoushan 2014Capone, Alessandro, and Mohammad A. Salmani Nodoushan 2014 “On Indirect Reports and Language Games: Evidence from Persian.” Rivista Italiana di Filosofia del Linguaggio 8(2): 26–42., 34). In the same vein, Salmani Nodoushan (2018)Salmani Nodoushan, Mohammad A. 2018 “Which View of Indirect Reports Do Persian Data Corroborate?” International Review of Pragmatics 10: 76–100. uses some evidence from Persian to support the semantico-pragmatic nature of indirect reports. Salmani Nodoushan states that ‘quotation’ and ‘paraphrasis’ are exactly the same in Persian. On this account, Salmani Nodoushan (2018Salmani Nodoushan, Mohammad A. 2018 “Which View of Indirect Reports Do Persian Data Corroborate?” International Review of Pragmatics 10: 76–100. , 89) states: “[i]t seems as if Persian DIs [indirect reports] are monophonic in that only the voice of the original speaker permeates all aspects of content sentences in DI reports.” By emphasising the monophonic nature of indirect reports, Salmani Nodoushan refutes the role of ‘polyphony’ (the existence of two voices) in Persian indirect reports. That said, Salmani Nodoushan (2015Salmani Nodoushan, Mohammad A. 2015 “The Secret Life of Slurs from the Perspective of Reported Speech.” Rivista Italiana di Filosofia del Linguaggio 9(2): 92–112., 107) clearly mentions that considering “indirect reports on purely semantic grounds is doomed, and that a true interpretation of the truth conditional nature of indirect reports will have to be based on an interplay between semantics and pragmatics.” A discussion on the reflexive nature of indirect reporting (presented in Section 5) will refute some of the previous findings on Persian indirect reports. My approach to investigating indirect reports is based on the analysis of authentic talk during interaction, which is a major difference between this study and previous studies.

4.Method

This study is based on the observation of naturally occurring talk in interaction. The data were gathered by collecting 117 Persian indirect reports from an Iranian national TV programme called ‘Goft-o Guː-je Viːʒe-je Xæbær-iː’ (Special News Talk), which is broadcast every night with at least one expert guest (and at the most two) in attendance, along with a male interviewer who asks technical questions and guides the interaction.11.The videos are archived and available on the following website: http://​www​.telewebion​.com​/program​/5094 [retrieved September 2018]. The expert guests were all officials (senior managers) who had executive responsibilities in the government. All the talks were conducted in Persian. The indirect reports analysed here were generated by both the expert guest and the interviewer. Approximately 800 minutes of the talks were observed and transcribed for analysis. The aim of the analysis was to identify traces of reflexive language use in indirect reports. The data were transcribed and analysed by the author.

The indirect reports are analysed in line with the reflexive use of language as realised through communicative know-hows and contextual clues. These contextual clues show whether the report aligns with the pragmatic force of the utterance or whether the reporter has changed the original utterances in a way that leads to disagreement on the part of the original speaker. The criterion here is the pragmatic force: whether the report respects the original pragmatic force or whether the reporter has intervened in this regard and changed the pragmatic force.

5.The metalinguistic function of reported speech

Language has the ability to refer and predicate about itself (Lucy 1993bLucy, John A. 1993b “Metapragmatic Presentationals: Reporting Speech with Quotatives in Yucatec Maya.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 91–125. Cambridge: Cambridge University Press. ). That said, confusion arises “where metalanguage and object language are implemented out of one and the same language” (Lucy 1993bLucy, John A. 1993b “Metapragmatic Presentationals: Reporting Speech with Quotatives in Yucatec Maya.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 91–125. Cambridge: Cambridge University Press. , 94). In this case, it is not clear which portions of the utterance are addressed by metalanguage and which portions serve as object language (the context to be referred to). According to Capone (2016)Capone, Alessandro 2016 The Pragmatics of Indirect Reports: Socio-philosophical Considerations. Cham: Springer., contextual clues, or metalinguistic signals, are useful in distinguishing between these intermingled voices. In this regard, reported speech can be ‘fully reflexive’ or ‘degenerately reflexive’ (which is closely related to the concept of opacity mentioned earlier in this paper). Lucy (1993bLucy, John A. 1993b “Metapragmatic Presentationals: Reporting Speech with Quotatives in Yucatec Maya.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 91–125. Cambridge: Cambridge University Press. , 94) explains these two concepts as follows:

Individual utterances are multifunctional in the sense of achieving multiple social and communicational goals at the same time. In reporting such utterances speakers may single out certain functions for emphasis. If these reports focus on the reference-and-predicational aspect of the utterance, then the reports are fully reflexive from a functional point of view since they involve reference-and-predication about reference-and predication. If these reports focus on the specific form or the nonreferential pragmatic (or indexical) qualities of the utterance then, by comparison, the reports are only degenerately reflexive from a functional point of view in that they attempt to refer to and predicate about aspects of the reported utterance which are not themselves reference-and-predicational (or propositional).

In line with Lucy (1993b)Lucy, John A. 1993b “Metapragmatic Presentationals: Reporting Speech with Quotatives in Yucatec Maya.” In Reflexive Language: Reported Speech and Metapragmatics, ed. by J. A. Lucy, 91–125. Cambridge: Cambridge University Press. , in the case of nonreferential pragmatic qualities, metapragmatic refers to metalinguistic forms that describe pragmatic aspects of language. In this article, ‘reflexivity’ is used in the broad sense, where interactants may choose to judge or comment on the speech, making amendments and clarifications to how an utterance should be interpreted. Previous research on Persian indirect reports indicates that such metadiscourse amendments and clarifications are not permitted within the structure of Persian indirect reports (see Salmani Nodoushan 2018Salmani Nodoushan, Mohammad A. 2018 “Which View of Indirect Reports Do Persian Data Corroborate?” International Review of Pragmatics 10: 76–100. ). However, the present research shows that indirect reports in Persian are not solid constructs in which samesaying is a hundred per cent perfect (cf. Capone and Salmani Nodoushan 2014Capone, Alessandro, and Mohammad A. Salmani Nodoushan 2014 “On Indirect Reports and Language Games: Evidence from Persian.” Rivista Italiana di Filosofia del Linguaggio 8(2): 26–42.). Persian indirect reports allow for transformations, in line with the reflexive use of language as realised through communicative know-hows and contextual clues. In what follows, extracts from the transcripts are presented to show how reflexivity is applied in Persian indirect reports.

5.1Fully reflexive indirect reports

A report is considered fully reflexive when the specific form or content of the reported utterance is (more or less) fully reproduced in line with the perspective of the original speaker (S = original speaker; R = reporter). Full reflexivity is distinguishable by the contextualisation clues provided by either the reporter or the original speaker.

Extract 1.22.The International Phonetic Alphabet (IPA) is used to transcribe Persian language pronunciation. Refer to https://​en​.wikipedia​.org​/wiki​/Help:IPA​/Persian for further information and examples. The use of the IPA allows the researcher to provide a more exact transcription of Persian alphabet.

S: jæʔniː ʃomɒː miːgiːd iːn tærh zɒːmen-e hoquːq-e mæsræf konænd-e hæst?

•That is, you say (that) this plan guarantees consumers’ rights?

R: sæd dær sæd zɒːmen-e hoquːq-e mæsræf konænd-e hæst.

•A hundred per cent it guarantees the consumers’ rights. (Episode:33.Episodes are the title of the Special Talks. I have provided the topic of the episodes to allow further research and investigation. Tariffs on automobile import; a support or a disaster for national industry?)

Extract (1) clearly shows that the (distinct) voices can cooperate with each other during the process of indirect reporting. The reporter in Extract (1) paraphrases the original speaker’ utterances (the speaker is present when the report is made) by using an interrogative tone. The question mark at the end of the report may indicate that the reporter is not sure whether he has reported the original speaker accurately enough (or questioning the truth of the claim; or expressing surprise at the meaning of the claim). Anyway, the presence of the original speaker allows for the negotiation of the report by the original speaker, and the possibility of a supportive or opposing follow-up on the part of the original speaker. In the above case, the original speaker uses the metapragmatic comment ‘sæd dær sæd’ (meaning ‘a hundred per cent’) to emphasise that the report is justified based on the perspective of the original speaker.44.In this regard, Levinson (1988)Levinson, Stephan C. 1988 “Putting Linguistics on a Proper Footing: Explorations in Goffman’s Participation Framework.” In Goffman: Exploring the Interaction Order, ed. by P. Drew and, A. Wootton, 161–227. Polity Press: Oxford. points out that what is not said is not said, if someone reports a sentence literally, it can be implied that the interlocutor only said what is literally reported and nothing else. Therefore, it must be an opt-out that more enrichments were made or were to be made in context. Therefore, the report is completely in line with the evaluative load of the original speaker’s utterance. The use of ‘a hundred per cent’ by the original speaker is meta-utterance, which shows that the report is fully reflexive.

An interesting issue in Extract (1) is the use of the polite word ‘miːgiːd’ (conveying the superiority of the original speaker), which is a face-enhancing verb of saying. In Persian, speakers are allowed to change the verb of saying (in line with communicative know-hows) in the process of indirect reporting and to use a more polite and face-giving verb of saying, such as ‘miːgiːd’ or ‘miːfærmɒːjiːd’ (both these verbs are polite versions of the word ‘say’), thus showing respect for the original speaker. In English, such inherently polite verbs of saying are absent. An English speaker should add the adverb ‘politely’ to the verb of saying (e.g. ‘you said politely’) to make a face-giving phrase (though I doubt any English speaker would do this in everyday conversation).55.However, one way to show respect when reporting speech in English would be to use the pronoun “himself” or “herself”. For example, “Amber herself said that the price of fuel is too high”. This would convey respect for the original speaker. It is relatively rare, but one can sometimes hear it – for example, when someone is reporting speech that was originally produced by a professor or the prime minister (and the reporter wants to show some sort of humility). This is a good topic for further investigation though, for politeness has not been sufficiently addressed in reported speech. That said, ‘saying politely’ differs slightly from ‘miːgiːd’ or ‘miːfærmɒːjiːd’ (these two phrases convey that, maybe insincerely, the person who has been reported is superior and deserves respect). In Persian, however, some verbs of saying inherently provide the assumption that the original speaker deserves respect (this is what happens in a formal context). This is largely because Persian speakers are under the influence of Persian cultural schemas, such as ‘adab va ehteram’66.As noted by Sharifian (2011Sharifian, Farzad 2011 Cultural Conceptualisations and Language: Theoretical Framework and Applications (vol. 1). Amsterdam: John Benjamins. , 106), ‘adab va ehteram’ are “roughly glossed as ‘courtesy and respect’ in English”, which “encourages Iranians to constantly place the presence of others at the centre of their conceptualizations and monitor their own ways of thinking and talking to make them harmonious with the esteem that they hold for others.” (courtesy and respect), which does not permit speakers to address others (nonfriends and nonacquaintances) by third person singular verbs and pronouns (they must use ‘iːʃuːn’ [‘they’] instead of ‘uː’ [he/she]). Extracts (2) and (3) show other cases of metapragmatic comments in Persian indirect reports:

Extract 2.

R: terɒːmp mætræh kærd ke ruːs-hɒː ɒːmɒːde-je muːʃæk-hɒː-je be qol-e xod-eʃ ‘ziːbɒː væ huːʃmænd-e’ ɒːmriːkɒːjiː-hɒː bɒːʃænd.

•Trump suggested that Russians be ready for, according to him, the ‘beautiful and smart’ American missiles. (Episode: The newest changes in Syria)

Extract 3.

R: dolæte ɒːmriːkɒː miːxɒːd hæme-je keʃvær-hɒː-ro, jæʔniː fæqæt uːn hæft keʃvær-æm niːst-eʃ, hæme-je, ‘every country’, hær keʃvær-iː ro miːxɒːd mædʒbuːr bokon‑e.

•The American government wants to oblige every country, that is, not only those seven countries, ‘every country’. (Episode: Investigating the new policies of the USA and their consequences)

As shown in Extracts (2) and (3), the reporter refers to some of the original words (‘ziːbɒː væ huːʃmænd-e’ and ‘every country’) of the source utterance and adds them to his report in order to adhere to the original speaker’s perspective and, at the same time, directly highlights the important features of the original utterance. What happens in Extract (2) is an instance of code-switching, whereby the reporter employs President Trump’s words (‘ziːbɒː væ huːʃmænd’, meaning ‘beautiful and smart’ in English) to imply that the reporter is aware of and has seen Trump’s exact words regarding his objection towards Russian military policies in Syria. The meta-utterance ‘be qol-e xod-eʃ’ (according to him) is a contextualisation clue that shows full reflexivity. In Extract (3), the reporter code-switches from Persian to English and uses the English phrase ‘every country’ to put emphasis on full reflexivity and to suggest that President Trump is indeed forcing every country (‘hæme-je keʃvær-hɒː-ro’) to act against Iran. In Extract (3), the indirect reporter does not use any verb of saying. In Extract (3), the reporter uses the phrase ‘every country’ in his sentence as a meta-utterance. The phrase ‘every country’ is a verbatim report of Trump’s original words. The phrase ‘dolæte ɒːmriːkɒː’ (the American government) represents President Trump. This was a rare use of implicit indirect reporting observed in the data.

Comparing Extracts (1) and (2), it is possible to see more vividly how politeness works in Persian indirect reports, which is possibly a unique feature of the Persian language. When referring to President Trump in Extract (2), the reporter uses the neutral verb of saying ‘mætræh kærd’ (it is an informal verb) instead of a more polite one such as ‘mætræh færmuːd’ or ‘mætræh færmuːdænd’ (which are more formal verbs of saying in Persian when it comes to nonfriends and nonacquaintances). On this account, reporting senior managers in Iran mostly requires a polite verb of saying, whereas reporting official people of other countries (especially those arouse opposition) does not require the reporter to use a polite verb of saying.

A fully reflexive indirect report does not inhibit the reporter from using syntactic transformations. The reporter can change the syntax of the original utterance at his/her will. The following extract shows syntactic transformations in line with full reflexivity:

Extract 4.

S: ɒːxæriːn soʔɒːl ro miːxɒːm beporsæm ke ælbæte xejliː sæxt-e væ tuːlɒːniː.

•I am going to ask you the last question, which is, of course, a very hard and long question.

R: xob hæmɒːntor ke færmuːd-iːd pɒːsox-e iːn xejliː dær zæmɒːn-e kuːtɒːh emkɒːn nædɒːre.

•Well, as you said, it is not possible to answer this (question) in a short period of time. (Episode: The newest changes in Syria)

In line with Extract (4), the phrase ‘xejliː sæxt-e væ tuːlɒːniː’ (‘very hard and long’) in the original speech is turned into ‘dær zæmɒːn-e kuːtɒːh emkɒːn nædɒːre’ (‘is not possible in a short time’) in the report, yet the perspective of the original speaker remains intact. Contrary to previous research, different cases of syntactic transformations were observed in Persian indirect reports, but these changes were only a close paraphrase of the original speech in which the pragmatic force of the original utterance was not manipulated by the reporter. On the other hand, the reporter has respected the ‘principle of economy’ mentioned above.

Another form of fully reflexive indirect reports among Persian speakers is observed in cases of self-report. Self-quotation happens when the reporter and the original speaker are the same, and where the original speaker reports one of his/her own previous utterances. The following extracts provide examples of self-quotation (and politeness in self-quotation) in Persian:

Extract 5.

S: pæs mæn hæm boʔd-e siːjɒːsiː væ tɒːriːx-iː ro xedmæt-e ʃomɒː ærz kærd-æm, hæm bo?d-e reqɒːbæt-iː ro ærz kærdæm væ hæm boʔd-e ertebɒːtɒːt-e mɒːliː-ʃ bɒː æræbestɒːn.

•Then, I told you about both the historical and political dimensions, I told you about both the competitive dimension and economical connections with Saudi Arabia. (Episode: The newest changes in Syria)

Extract 6.

S: bænde dær hæmiːn bærnɒːme je bɒːr eʔlɒːm kærdæm uːn ro biːjoteroriːsm miːdɒːnæm.

•I once declared to you in this very programme (that) I would call it bioterrorism. (Episode: The newest changes in Syria)

Extract 7.

S: hæmɒːntor ke goft-æm, jek-iː æz tæqiːrɒːt-e omde dær qebɒːl-e iːrɒːn xɒːhæd buːd.

•As I told you, (it) would be one of the main changes in acting against Iran. (Episode: Dismissing the USA’s Minister of Foreign Affairs and its consequences)

Extract 8.

S: ærz kærd-æm mohemtær-iːn dærxɒːst-e mærdom tuː iːn æjɒːm æmniːjæt-e ʔomuːm-iː æst.

•I told you (that) the most important request from people in these days is for general security. (Episode: The security of roads in holidays)

Of particular importance in the above extracts is the politeness work observed in the verbs of saying used by the reporter or the original speaker. Persian allows the inclusion of the politeness work in the verbs of saying (unlike in English and possibly in other languages). Persian speakers can change the verbs of saying, as well as pronouns, to convey a version of inferiority or posterity that is anchored in Persian cultural schemas such as adab va ehteram (courtesy and respect) and shekasteh-nafsi 77.According to Sharifian (2005Sharifian, Farzad 2005 “The Persian Cultural Schema of Shekasteh-nafsi: A Study of Compliment Responses in Persian and Anglo-Australian Speakers.” Pragmatics & Cognition 13(2): 337–361. , 337), “[t]he schema [shekasteh-nafsi] motivates the speakers to downplay their talents, skills, achievements, etc. while praising a similar trait in their interlocutors.” (modesty). In Extract (5), the phrase ‘xedmæt-e ʃomɒː’ (‘at your service’) is a face-giving act, whereas ‘ærz kærd-æm’ (a version of ‘told you’, whereby the reporter downgrades himself/herself) is both an indication of complete reflexivity and a sign of shekasteh-nafsi. A distinction between ‘ærz kærd-æm’ (in Extract [5]) and ‘færmuːd-iːd’ (in Extract [4]) seems relevant and informative here. Both of these terms are related to verbs of saying in Persian, yet there are two main differences between them, as is evident in the above extracts. Firstly, ‘ærz kærd-æm’ is most often used in the original speaker’s speech and refers back to the original speaker’s previous utterances (the reporter and the speaker are the same), whereas ‘færmuːd-iːd’ is mostly observed in the reporter’s speech and refers back to the original speaker’s previous utterances (the reporter and the speaker are not the same). Secondly, ‘ærz kærd-æm’ is a sign of shekasteh-nafsi, which indicates meekness on the part of the reporter, whereas ‘færmuːd-iːd’ is a face-giving act and a sign of politeness, whereby respect is shown for the original speaker by the reporter.

In Extracts (6–8), full reflexivity is indicated by the use of different verbs of saying (‘eʔlɒːm kærdæm’; ‘goft-æm’; ‘ærz kærd-æm’). The verbs ‘eʔlɒːm kærdæm’ (announced) and ‘goft-æm’ (said) are more neutral in terms of politeness than the verb ‘ærz kærd-æm’ (the ‘humble’ version of saying ‘told you’). Self-quotation is a common phenomenon in other languages such as English. That said, in English, there is no way to indicate politeness or humbleness by changing the verbs of saying.

5.2Degenerately reflexive indirect reports

In degenerately reflexive indirect reports, the original speaker’s utterance is manipulated beyond its illocutionary force. In the case of Persian indirect reports, the data in this study provide evidence for Borg’s (2012)Borg, Emma 2012 Pursuing Meaning. Oxford: Oxford University Press. concept of ‘free pragmatic enrichment’, which refers to interpretations that are not characterised by any lexico-syntactic element of the source information. These kinds of so-called ‘implicit arguments’ (Capone 2013Capone, Alessandro 2013 “Consequences of the Pragmatics of ‘De Se’.” In Attitudes ‘De Se’: Linguistics, Epistemology and Metaphysics, ed. by Alessandro Capone, and N. Feit, 209–244. Stanford: CSLI.) are more often than not seen in Persian indirect reports.88.In this regard, Capone (2019Capone, Alessandro 2019 Pragmatics and philosophy: Connections and ramifications. Cham: Springer. , 224) mentions that “[c]oncerning indirect reports in disguise, we can say that a syntactic analysis is coupled with a certain semantic-pragmatic interpretation and helps guide the interpretation.” In this sense, Capone (2018Capone, Alessandro 2018 “On the Social Praxis of Indirect Reporting.” In Indirect Reports and Pragmatics in the World Languages, ed. by Alessandro Capone, M. Garcia-Carpintero, and A. Falzone, 3–20. Heidelberg, Springer., 11) refers to the concept of ‘opacity’, and states that “[i]t is well known that that-clauses are intensional contexts, that is to say contexts in which it is not licit to substitute an NP (but it could also be another element of the sentential structure, such as a verb, for example) with a coextensive one (one which denotes the same object).” Capone continues by stating that opacity “is imported into indirect reports only because pragmatically they can be seen as representing the voice of the original speaker and the original speaker may object to the substitution of certain words” (2018Capone, Alessandro 2018 “On the Social Praxis of Indirect Reporting.” In Indirect Reports and Pragmatics in the World Languages, ed. by Alessandro Capone, M. Garcia-Carpintero, and A. Falzone, 3–20. Heidelberg, Springer., 15). The following extracts emphasise Capone’s notion of opacity by opposing illicit transformations.99.In discussing implicit indirect reports, Capone (2019Capone, Alessandro 2019 Pragmatics and philosophy: Connections and ramifications. Cham: Springer. , 224) argues that “[w]e assume that more examples than are under discussion in this paper can be subjected to a similar analysis, as most cases in which one cannot directly know another person’s mind but gets to know it through what a speaker has said, potentially constitute cases that can be analysed as implicit indirect reports.” Degenerately reflexive indirect reports ignore opacity, since in this case “one would distort the truth about the attitudes of the (original) speaker towards the referents” (Capone 2018Capone, Alessandro 2018 “On the Social Praxis of Indirect Reporting.” In Indirect Reports and Pragmatics in the World Languages, ed. by Alessandro Capone, M. Garcia-Carpintero, and A. Falzone, 3–20. Heidelberg, Springer., 16). In the following extracts, which may be considered degenerately reflexive indirect reports, the original speaker disagrees or challenges the report and its pragmatic force by explicitly stating that he is not satisfied with the reporter’s interpretation:

Extract 9.

R: ʃomɒː miːgiːd mɒː bɒːjæd beriːm be sæmt-e reqɒːbæt-iː kærdæn-e bɒːzɒːr ?

•You say we should move towards a competitive market?

S: mæn miːxɒːm ʔærz bokonæm ke mɒː lozuːmæn bæxʃ-iː æz iːn zændʒiːr-e rɒː toliːd koniːm.

•I want to say that we necessarily produce part of this cycle. (Episode: Tariffs on automobile import; a support or a disaster for national industry?)

With regard to Extract (9), the reporter uses an interrogative tone, which indicates that the reporter is not sure about his interpretation of the original utterance. The reporter uses ‘miːgiːd’, which is a polite version of saying ‘goft-iː’ in Persian. One interesting point is that Persian speakers sometimes use a present continuous version of the verb of saying, which is not common in other languages such as English. In English, more often than not the past-tense of the term ‘say’ or ‘tell’ is used, whereas in Persian it is quite common that other tenses, such as the present continuous, are used to talk about a past event. Extract (9) shows clearly that the report is not fully reflexive. The original speaker does not accept the reporter’s interpretation, and states ‘mæn miːxɒːm ʔærz bokonæm ke’ (I want to say that …), which indicates that the original speaker’s perspective was different to the one that the reporter is trying to communicate. In Extract (10), the conflict between the views is more evident:

Extract 10.

R: færmuːd-e buːdiːd jek miːljuːn tomæn hær metr kɒːheʃ dɒːʃt-e dær tehrɒːn… ɒːre?

•You had said it has been decreased by one million Toman (the official currency of Iran) in every metre in Tehran… Yes?

S: mæn tʃeniːn ʔærz nækærd-æm. væliː dær mædʒmuːʔ miːʃe goft æz sɒːl-e 1392 be bæʔd miːtuːnæm ʔærz bokonæm ke mɒː æfzɒːjeʃ-e qejmæt-e mæhsuːs dær qejmæt-e mæskæn nædɒːʃtiːm.

•I didn’t say any such thing. But, overall, it is possible to say that from 1392 (according to Persian calendar) onwards we haven’t seen a tangible increase in housing prices. (Episode: Strategies to save housing from stagnation)

The metapragmatic comment,1010.According to Weizman and Dascal (1991)Weizman, Elda, and Marcelo Dascal 1991 “On Clues and Cues: Strategies of Text-Understanding.” Journal of Literary Semantics 20: 18–30. , clues relevant to meta-linguistic contexts play a pivotal role in the construction of indirect reports. ‘ɒːre?’, which is an interrogative ‘yes’, in Extract (10) can be a sign of uncertainty, which can reveal that the reporter is not sure about the information conveyed through the report. As argued by Weizman and Dascal (1991)Weizman, Elda, and Marcelo Dascal 1991 “On Clues and Cues: Strategies of Text-Understanding.” Journal of Literary Semantics 20: 18–30. , a clue in indirect reports can show that a specific word includes part of the linguistic repertoire of a certain individual. Likewise, Salmani Nodoushan (2017Salmani Nodoushan, Mohammad A. 2017 “Lexemes, Practs, and Those Who Have Yet to Decide.” Linguistik Online 81: 77–93. , 80) reveals that “[h]uman agents who produce linguistic utterances have certain intentions (in so doing), which are mutually manifest for both themselves and their interlocutors.” Extracts (9) and (10) are different in one important aspect, in that in Extract (9) the original speaker does not boldly disagree with the pragmatic load of the report, whereas in Extract (10), the original speaker says ‘mæn tʃeniːn ʔærz nækærd-æm’ (I didn’t say any such thing), which vividly shows the speaker’s dissatisfaction with the reported information. The reports in Extracts (9) and (10) are not in line with the original speaker’s perspective, and hence these are degenerately reflexive. Extract (11) reveals another instance of reflexive language use:

Extract 11.

R: je nokte-jiː tuːje hæste-je mætlæb-e iːʃuːn iːn buːd ke ægær modiːr-iː ro ke xejliː hæm bɒː kefɒːjæt hæst, mɒː bærɒːje iːn-ke negæh-eʃ dɒːriːm bɒːjæd be-heʃ hoquːq-e bɒːlɒː-tær bediːm. iːndʒuːriː buːd?

•One hint in the centre of his discussion was that if we are going to keep a manager who has many qualifications, we should pay him/her a higher salary. Am I right?

S: næ, næ, tæfsiːr-e mæn lɒː-jærzɒː mo’ælef nækon-iːd. ʔærz mæn iːn æst ke mɒː je bɒːr qɒːnuːn-gozɒːr-iː kærd-iːm, modiːriːjæt-e xædæmɒːt-e keʃvær-iː æz tuːʃ dær ɒːvord-iːm.

•No, no, I am not satisfied with this kind of interpretation. I am saying that we settled the rules once, and the management of national service came out of it. (Episode: Discussing the reasons for rejecting the plan for the transparency of managers’ salaries)

In Extract (11), another marker of uncertainty is used by the reporter, which is the phrase ‘iːndʒuːriː buːd?’ (Am I right?). The original speaker does not accept the illocutionary force of the report, and uses a bold-on-record expression, which is an Arabic proverbial ‘tæfsiːr-e mæn lɒː-jærzɒː mo’ælef nækon-iːd’, meaning ‘I am not satisfied with this kind of interpretation’. Extract (12) shows another metapragmatic comment that is more complicated than the previous cases:

Extract 12.

R: ɒːqɒː-je mælekiː miːgæn ke, be hær hɒːl, sænʔæte xodro hæsɒːs-e.

•Mr Maleki says that, in any case, the automobile industry is sensitive. (Episode: Tariffs on automobile import; a support or a disaster for national industry?)

Another important feature of reflexivity in Persian indirect reports can be seen in Extracts (12) and (13). It is not a common and acceptable practice to use ambiguous words (e.g., ‘be hær hɒːl’) in the report (Capone 2016Capone, Alessandro 2016 The Pragmatics of Indirect Reports: Socio-philosophical Considerations. Cham: Springer.). As discussed by Capone (2016)Capone, Alessandro 2016 The Pragmatics of Indirect Reports: Socio-philosophical Considerations. Cham: Springer., the reporter should not report those parts which s/he believes are ambiguous, since this would increase the cognitive load on the part of the hearer. In Extract (12), the reporter uses an ambiguous phrase (‘be hær hɒːl’ [in any case]) after the complementiser that [ke in Persian]). However, the phrase ‘be hær hɒːl’ seems out of place (according to Weizman and Dascal [1991]Weizman, Elda, and Marcelo Dascal 1991 “On Clues and Cues: Strategies of Text-Understanding.” Journal of Literary Semantics 20: 18–30. , these clues can reveal how indirect reports are constructed and deconstructed), and it is clear that this phrase does not belong to the original speaker (it does not contribute to the truth-conditionality of the utterance). Why, then, has the reporter used this ambiguous phrase (i.e., ‘be hær hɒːl’) after the complementiser ke? There are two possible interpretations. One is that the reporter has used a prefabricated phrase to buy some time. The reporter uses this phrase to allow him to think more and remember the information that he is going to convey through the indirect report. Another interpretation is that the reporter tries to show that the report is not a strict interpretation of the original speaker’s utterance (and hence it is degenerately reflexive). Another case of an ambiguous report is observed in Extract (13):

Extract 13.

R: ælɒːn miːfærmɒːjiːd ke ælɒːn bædæne-je vezɒːræt-e bɒːzærgɒːn-iː, iːmiːdro, iːrɒːn xodro, tʃiː, tʃiː, tʃiː,…

•Now you say that the body of the Ministry of Commerce, IMIDRO, Iran Xodro, blah, blah, blah, … (Episode: The separation of the Ministry of Commerce from the Ministry of Industry and Mining)

In Extract (13), the reporter uses a succession of exclamations (‘tʃiː’, which means ‘blah’ in English), which are employed to substitute for actual words in contexts where the words are felt to be too tedious or too lengthy to give in full (or perhaps the speaker cannot remember the exact words). As mentioned by Saka (2017Saka, Paul 2017 “Blah, Blah, Blah: Quasi-quotation and Unquotation.” In The Semantics and Pragmatics of Quotation, ed. by P. Saka, and M. Johnson, 35–64. Heidelberg, Springer. , 47), ‘blah’ “connote[s] uninterest on the part of the reporter: material is omitted not just because it is irrelevant but, especially in the case of blah, because it is tedious, and the original source is portrayed as a blatherer.” Boredom is not always the reason why a word is deleted. One can delete some parts of the report because s/he thinks the information is irrelevant, or the deletion may happen to reduce the cognitive load when retrieving the exact information becomes burdensome. Whatever the reason, exclamations such as ‘blah’ and ambiguous words such as ‘in any case’ can indicate that the report is not fully reflexive.

6.Discussion

The results of this study direct us towards the concept of ‘evidentiality’, which is closely related to two types of reflexivity observed in Persian indirect reports. As noted by Ifantidou (2001Ifantidou, Elly 2001 Evidentials and Relevance. Amsterdam: John Benjamins. , 2), “the notion of evidential is basically a functional one: that is, it is applied to linguistic expressions in virtue of their pragmatic function.” Ifantidou continues by stating that “evidentials are devices used by speakers to mark the source and the reliability of their knowledge” (2001Ifantidou, Elly 2001 Evidentials and Relevance. Amsterdam: John Benjamins. , 3). On this account, evidentiality refers to “things people are less sure of, and some things they think are only within the realm of possibility. Languages typically provide a repertoire of devices for conveying these various attitudes towards knowledge” (Chafe and Nichols 1986Chafe, Wallace L., and Johanna Nichols 1986 “Evidentiality in English Conversation and Academic Writing.” In Evidentiality: The Linguistic Coding of Epistemology, ed. by W. Chafe, and J. Nichols, 261–272. New Jersey: Ablex., vii). Persian indirect reports exhibit elaborated evidential systems that require conscious attention. As the results of this study revealed, the contextual cues used in indirect reports enable the reporter and the original speaker to vary the certainty/amount of the information being communicated during the interaction. Persian indirect reports showed evidentiality in two respects: ‘the source of knowledge’ and ‘degree of certainty’. Let’s elaborate on each in turn.

With regard to the source of knowledge, evidentiality was clearly observed in Persian indirect reports. In Extract (3), the reporter code-switches to English and uses the phrase ‘every country’ to highlight the original speaker (President Trump). Here, ‘every country’ is not only an evidential expression but a phrase to indicate that the reporter has heard (and is cognisant of) the exact words of President Trump. Thus, the reporter is able to convince the hearer more greatly by mentioning the source of the information. The phrase ‘every country’ seems syntactically independent from the main clause, hence the phrase is syntactically parenthetical. Another kind of evidentiality was observed in self-quotations in Persian indirect reports. Verbs of saying, such as ‘eʔlɒːm kærdæm’, ‘goft-æm’, ‘ærz kærd-æm’, ‘færmuːd-iːd’, ‘miːfærmɒːjiːd’, and ‘færmuːd-e buːdiːd’, are evidential markers that reveal the author of the report. In this regard, Goffman (1981Goffman, Erving 1981 Forms of Talk. Philadelphia: University of Pennsylvania Press., 144–145) distinguishes between the different roles of the speaker as follows:

In terms of the verbs of saying used in Persian indirect reports, those referring to the reporter (i.e., ‘færmuːd-iːd’, ‘miːfærmɒːjiːd’, and ‘færmuːd-e buːdiːd’) show that the author and the principal are the original speaker, whereas the animator is the reporter. Regarding the verbs of saying referring to the original speaker (i.e., ‘eʔlɒːm kærdæm’, ‘goft-æm’, ‘ærz kærd-æm’), the original speaker plays the role of animator, author, and principal at the same time.

Persian speakers resort to many devices to show the degree of certainty in their indirect reports (and they do it in many different ways). The first category of evidentials in this regard is related to ambiguous words used in Persian indirect reports. These “implicit assumptions about the source of the information might play a role in the interpretation of the utterance itself” (Ifantidou 2001Ifantidou, Elly 2001 Evidentials and Relevance. Amsterdam: John Benjamins. , 8). These assumptions can help the hearer to distinguish between fully reflexive and degenerately reflexive reports. The instances are frequent in this regard; evidentials such as ‘jæʔniː’, ‘ɒːre?’, ‘iːndʒuːriː buːd?’, ‘be hær hɒːl’, and ‘tʃiː, tʃiː, tʃiː’ all show that the reporter is not entirely certain about the information being conveyed through the report. Of particular interest regarding the evidentials observed in this category is the phrase ‘be hær hɒːl’ (see Extract [12]), which is an evidential expression that is syntactically parenthetical and hence independent from the main clause. According to Ifantidou (2001)Ifantidou, Elly 2001 Evidentials and Relevance. Amsterdam: John Benjamins. , true parentheticals, such as ‘John is, I think, in Berlin’, should be distinguished from the main-clause counterparts, such as ‘I think John is in Berlin’. The phrase ‘be hær hɒːl’ is a true evidential, where the “relation is […] established at the level of utterance interpretation rather than in the syntax” (Ifantidou 2001Ifantidou, Elly 2001 Evidentials and Relevance. Amsterdam: John Benjamins. , 13). True parentheticals such as ‘be hær hɒːl’ seem to make no contribution to the truth conditions of the sentences in which they appear. Another source of ambiguity comes from what Blakemore (1994)Blakemore, Diane 1994 “Evidence and Modality.” In The Encyclopedia of Language and Linguistics, ed. by R. Asher, and J. Simpson, 1183–1186. Oxford: Pergamon Press. refers to as ‘nonlinguistic evidentials’ (such as intonation and prosodic features). Extract (9) provides a good example of this. Here, the interrogative tone indicates uncertainty on the part of the reporter.

The second category related to the evidentials showing the degree of certainty is words that do not indicate uncertainty, but precision. Among the extracts, examples such as ‘sæd dær sæd’ (a hundred per cent), ‘be qol-e xod-eʃ’ (according to him), ‘hæmɒːntor ke færmuːd-iːd’ (as you mentioned), and ‘mæn tʃeniːn ʔærz nækærd-æm’ (I didn’t say any such thing) show precision. It can be concluded that both precision and uncertainty were covered by metapragmatic comments in Persian indirect reports.

7.Concluding remarks

The findings of this study showed some culture-specific features of indirect reports in Persian. Amongst them is the modification of the verb of saying (‘færmuːd-iːd’ instead of ‘goftiːd’), and the use of certain personal pronouns (‘iːʃuːn’ instead of ‘uː’). Moreover, Persian speakers use syntactic clues to indicate ‘full’ or ‘loose’ reflexivity. Besides fully reflexive and degenerately reflexive indirect reports, Persian speakers use self-reporting as a paraphrase of some previous speech. Self-reporting can be direct or indirect, and yet the use of direct speech in self-quotation was not observed in the present study data. That said, although the literature on reported speech indicates that direct reports are the verbatim transfer of the original speaker’s utterances, recent research (see Itakura 2018Itakura, Hiroko 2018 “Accuracy in Reported Speech: Evidence from Masculine and Feminine Japanese Language.” In Indirect Reports and Pragmatics in the World Languages, ed. by Alessandro Capone, M. Garcia-Carpintero, and A. Falzone, 315–332. Heidelberg, Springer.) has proved that subjective manipulation is not impossible in direct reports, and hence, there is no difference between direct and indirect reports with regard to their manipulative power. Self-quotation is mostly used to refresh the mind of the hearer, to buy some extra time for further cognitive processing, or (as a metapragmatic comment) to attract the attention of the hearer towards a particular issue (that the report is not fully reflexive). Persian speakers self-quote in two situations. First, and in the case of degenerately reflexive reports, they may self-quote when the reporter does not provide a faithful paraphrase of the original speaker’s utterances. In this case, the self-quote acts as a further explanation, letting the hearer know the intention and perspective of the original speaker (evidentials in terms of the degree of certainty). Second, they may self-quote when the original speaker wants to put emphasis on some of his/her previous remarks or to refresh the mind of the hearer (evidentials in terms of the source of information).

Indirect reports are complex social events (language games according to Wittgenstein). It is logical to say that indirect reports are under the control of specific principles (Capone 2016Capone, Alessandro 2016 The Pragmatics of Indirect Reports: Socio-philosophical Considerations. Cham: Springer.), but individuals may also breach the rules and maintain their own justifications in conveying (distorted) indirect reports. As endorsed by Wieland (2010)Wieland, Nellie 2010 “Context Sensitivity and Indirect Reports.” Philosophy and Phenomenological Research 81: 40–48. , the aim of indirect reports is to report the speaker’s meaning (if not, the interpretation may be wrong). Sometimes the rules are not broken but manipulated to better fit the context (and metapragmatic comments are useful in this regard). To elaborate more, individuals break the rules in line with their perceptions of the situation (situational necessity) and according to their experiences (pragmatic opacity). It is always possible that individuals will modify the utterances as a matter of self-interest or as a matter of politeness. Regarding (the differences between) Wittgenstein’s and Weigand’s views on the concept of the game, Wittgenstein’s definition of the language game is realised in social interaction (DeAngelis 2007DeAngelis, William James 2007 Ludwig Wittgenstein-A Cultural Point of View: Philosophy in the Darkness of this Time. Cornwall: Ashgate.) and is a rule-governed process, whereas Weigand sees the language game through a ‘sociocognitive paradigm’, which is less confined to rules. To this end, fully reflexive and degenerately reflexive indirect reports can be accounted for based on Weigand’s view regarding the process of meaning-making during the interaction (the language game is cyclic, never-ending [Weigand 2010Weigand, Edda 2010 Dialogue: The Mixed Game. Amsterdam: John Benjamins. ]). For Wittgenstein, there is a “speaking outside language games” (Cavell 1988Cavell, Stanley 1988 “Declining Decline: Wittgenstein as a Philosopher of Culture.” Inquiry 31(3): 253–264. , 73), which means “repudiating our shared criteria” (Cavell 1988Cavell, Stanley 1988 “Declining Decline: Wittgenstein as a Philosopher of Culture.” Inquiry 31(3): 253–264. , 73). By contrast, for Weigand, speaking outside language games is impossible.

The results of this study contradict Anton (1998)Anton, Corey 1998 “ ‘About Talk’: The Category of Talk-reflexive Words.” Semiotica 121: 193–212. , for Persian speakers did not necessarily attend to speech (using metapragmatic comments) only “in moments of communicative breakdown or dysfunction” (Anton 1998Anton, Corey 1998 “ ‘About Talk’: The Category of Talk-reflexive Words.” Semiotica 121: 193–212. , 202). Persian speakers employed metadiscourse to show politeness and uncertainty/precision, or to facilitate the retrieval of relevant information, among other things. It was discovered that when the reporter encounters a communicative breakdown, he can indicate that this is the case by using contextual clues such as question marks, exclamations, ambiguous prefabricated phrases, etc. Moreover, the present study findings provide evidence against the monophonic nature of Persian indirect reports, and this is also corroborated by Salmani Nodoushan (2018)Salmani Nodoushan, Mohammad A. 2018 “Which View of Indirect Reports Do Persian Data Corroborate?” International Review of Pragmatics 10: 76–100. . Additionally, Persian speakers changed the syntax of the report, which helped them reduce the cognitive load and show reflexivity (both degenerately and fully). It seems that Persian indirect reports have many things to offer a theory of reported speech. That said, further research is needed across languages to help researchers propose an all-inclusive theory of indirect reports.